JP7460725B2 - Information processing device, information processing method, and information processing program - Google Patents

Description

The present invention relates to a method of providing information regarding properties such as houses and buildings.

Conventionally, real estate information providing systems that provide information about real estate through websites and applications are known. In such systems, it is common for a user to first input the conditions for a desired property. Examples of such conditions include sales price or rent, type of floor plan (e.g., 2LDK, 3LDK, etc.), type of building, age, exclusive floor area, area, railway line, travel time to a station, presence or absence of specific facilities, etc. The real estate information providing system then searches for and provides property information that matches the input conditions. The provided property information includes, for example, various attributes of the property and floor plans. For example, Patent Document 1 discloses a real estate information search device that acquires real estate information based on search conditions such as price, land area, building area, age, number of rooms, etc.

Japanese Patent Application Publication No. 2015-129980

However, even if property information that matches the search conditions specified by the user is presented, the internal appearance of the property shown by the floor plan included in the property information, such as the floor plan, may not match the user's wishes. . The reason for this is that the floor plan may represent various features that are not specified by the above-mentioned search conditions. Therefore, the user is required to search for property information that meets the user's wishes from a plurality of pieces of property information that match the specified search conditions while looking at the floor plan. On the other hand, even if the user latently has a desire regarding the interior of the property, he or she may not recognize this as a clear condition. Therefore, even if a new search condition that can be specified regarding the interior of a property is established, it may be difficult for the user to determine whether or not to specify the condition.

The present invention has been made in view of the above points, and one example of the problem is that information used to identify a property of the user's preference is based on the user's understanding of the characteristics of the rooms in the property and the floor plan of the property. An object of the present invention is to provide an information processing device, an information processing method, and an information processing program that can be associated with evaluations .

In order to solve the above problem, the invention according to claim 1 provides feature information generated based on a floor plan image of a property, wherein feature information indicating the characteristics of a room in the property is for each of the plurality of properties stored in the storage means in association with property identification information for identifying the screen of a display device having a display unit having a screen, and a detection unit detecting a contact position on the screen; display control means for displaying the floor plan image on a screen, and evaluation information acquisition means for acquiring evaluation information indicating an evaluation of the displayed floor plan image for each of the plurality of properties, the evaluation information acquisition means for obtaining evaluation information indicating an evaluation of the displayed floor plan image for each of the plurality of properties; If a movement of the contact position in a predetermined first direction is detected on the screen on which an image is displayed, acquiring the evaluation information indicating a positive evaluation of the displayed floor plan image; evaluation information acquisition means for acquiring the evaluation information indicating a negative evaluation of the displayed floor plan image when movement of the contact position in a predetermined second direction different from the first direction; Evaluation information storage control for storing the acquired evaluation information in the storage means for each of the plurality of properties in association with user identification information that identifies a user of the display device and the property identification information that identifies the property. It is characterized by comprising means .

In the invention described in claim 2, the evaluation information acquisition means, when a predetermined first aspect of movement of the contact position in the first direction is detected, acquires the evaluation information indicating the positive evaluation of the entirety of the property for which the floor plan image is displayed on the screen, when the first aspect of movement of the contact position in the second direction is detected, acquires the evaluation information indicating the negative evaluation of the entirety of the property for which the floor plan image is displayed on the screen, and when a predetermined second aspect of movement of the contact position in a predetermined third direction, which second aspect of movement is different from the first aspect of movement, acquires the evaluation information indicating the negative evaluation of the entirety of the property for which the floor plan image is displayed on the screen, and when the movement of the touch position in the second aspect in a predetermined fourth direction different from the third direction is detected, acquires the evaluation information indicating the negative evaluation of the room at the touch position among at least one room shown in the displayed floor plan image, and when the evaluation information indicating the evaluation of the room at the touch position is acquired by the evaluation information acquisition means, the evaluation information storage control means associates the acquired evaluation information with the user identification information, the property identification information, and room identification information that identifies the room at the touch position, and stores the association in the storage means.

In the invention according to claim 3, the storage means stores, for each of at least one room in the property, the characteristic information indicating the characteristics of the room, the property identification information, and a room identification for identifying the room. It is characterized by being stored in association with information.

The invention described in claim 4 is characterized in that it further comprises a positive evaluation information acquisition means for acquiring evaluation information that indicates the positive evaluation from the evaluation information stored in the storage means, and a negative evaluation information acquisition means for acquiring evaluation information that indicates the negative evaluation from the evaluation information stored in the storage means.

The invention described in claim 5 is characterized in that it further comprises a positive evaluation model generation means for generating a positive evaluation model indicating a probability distribution in which the feature information for a room in a property for which the acquired evaluation information indicates the positive evaluation is a random variable based on the evaluation information acquired by the positive evaluation information acquisition means and the feature information associated with the property identification information associated with the evaluation information, and a negative evaluation model generation means for generating a negative evaluation model indicating a probability distribution in which the feature information for a room in a property for which the acquired evaluation information indicates the negative evaluation is a random variable based on the evaluation information acquired by the negative evaluation information acquisition means and the feature information associated with the property identification information associated with the evaluation information.

The invention described in claim 6 is an information processing method executed by a computer, comprising: a display control step of displaying the floor plan image on a screen of a display device having a display unit having a screen and a detection unit that detects a contact position on the screen, for each of a plurality of properties in which feature information generated based on a floor plan image of a property, the feature information indicating features of a room within the property, is stored in a storage means in association with property identification information that identifies the property; an evaluation information acquisition step of acquiring evaluation information indicating an evaluation of the displayed floor plan image for each of the plurality of properties, the evaluation information acquiring step acquiring the evaluation information indicating a positive evaluation of the displayed floor plan image when movement of the contact position in a predetermined first direction is detected on the screen on which the floor plan image is displayed, and acquiring the evaluation information indicating a negative evaluation of the displayed floor plan image when movement of the contact position in a predetermined second direction different from the first direction is detected; and an evaluation information storage control step of storing the acquired evaluation information in the storage means, for each of the plurality of properties, in association with user identification information that identifies a user of the display device and the property identification information that identifies the property.

The invention described in claim 7 is an information processing program that causes a computer to function as: a display control means for displaying a floor plan image on a screen of a display device having a display unit having a screen and a detection unit for detecting a contact position on the screen, for each of a plurality of properties in which feature information generated based on a floor plan image of a property, the feature information indicating features of a room within the property, is stored in a storage means in association with property identification information that identifies the property; an evaluation information acquisition means for acquiring evaluation information indicating an evaluation of the displayed floor plan image for each of the plurality of properties, the evaluation information acquisition means acquiring, when movement of the contact position in a predetermined first direction is detected on the screen on which the floor plan image is displayed, the evaluation information indicating a positive evaluation of the displayed floor plan image, and acquiring, when movement of the contact position in a predetermined second direction different from the first direction is detected, the evaluation information acquiring means acquiring, when movement of the contact position in a predetermined second direction different from the first direction is detected, the evaluation information indicating a negative evaluation of the displayed floor plan image; and an evaluation information storage control means for storing, for each of the plurality of properties, the acquired evaluation information in the storage means in association with user identification information that identifies a user of the display device and the property identification information that identifies the property.

According to the present invention, as information used to identify a user's preferred property, the characteristics of a room in the property can be associated with the user's evaluation of the floor plan of the property .

1 is a diagram illustrating an example of a schematic configuration of a property information providing system S according to an embodiment. 1 is a block diagram showing an example of a schematic configuration of an information providing server 1 according to an embodiment. 3 is a diagram showing an example of information stored in a database of the information providing server 1. FIG. 1 is a diagram showing an example of functional blocks of a system control unit 11 of an information providing server 1. FIG. 13A is a diagram showing an example of a floor plan screen display, and FIG. 13B is a diagram showing an example of a property details screen display. 1A is a diagram showing an example of inputting a positive evaluation for a floor plan image and an example of an evaluation for each room, and FIG. 1B is a diagram showing an example of inputting a negative evaluation for a floor plan image and an example of an evaluation for each room. 13A and 13B are diagrams showing examples of input of partial evaluations and overall evaluations for a floor plan image and examples of evaluations for each room. FIG. 3 is a diagram illustrating an example of generation of segmented floor plan information. FIG. 13 is a diagram showing an example of a feature vector of a property acquired from a floor plan image. FIG. 13 is a diagram illustrating an example of the configuration of a property classifier. It is a figure which shows the example of calculation of a feature item score. FIG. 7 is a diagram illustrating an example of a difference between the probability distribution of a feature amount for an evaluation of “like” and the probability distribution of a feature amount for an evaluation of “dislike.” FIG. 10 is a flowchart showing an example of a floor plan analysis process by a system control unit 11 of the information providing server 1 according to an embodiment. 10 is a flowchart showing an example of a property recommendation process performed by a system control unit 11 of the information providing server 1 according to an embodiment. 10 is a flowchart showing an example of a property selection process by a system control unit 11 of the information providing server 1 according to an embodiment. 5 is a flowchart illustrating an example of a floor plan display process performed by the CPU of the user terminal 2 according to an embodiment. 10 is a flowchart illustrating an example of a learning process performed by a system control unit 11 of the information providing server 1 according to an embodiment.

The following describes in detail an embodiment of the present invention with reference to the drawings.

[1. Configuration of property information provision system]
First, the configuration and functional outline of the property information providing system S according to this embodiment will be explained using FIG. 1. FIG. 1 is a diagram showing an example of a schematic configuration of a property information providing system S according to the present embodiment.

As shown in FIG. 1, the property information providing system S includes an information providing server 1 and a plurality of user terminals 2. The information providing server 1 and each user terminal 2 are connected to each other via a network NW. The network NW is constructed, for example, from the Internet, a dedicated communication line (e.g., a CATV (Community Antenna Television) line), a mobile communication network (including base stations, etc.), and a gateway, etc.

The information providing server 1 is a server device that delivers information on real estate properties to the user terminals 2. The property to which information is provided may be, for example, either a residence (for example, a house, a room in an apartment, a room in a condominium, etc.) or an office. Further, the property for which information is provided may be, for example, a newly built property, a used property, or a rental property. The information providing server 1 may transmit property information in response to a request from the user terminal 2, for example. The property information may include, for example, property name, address, price, exclusive area, floor plan type, and other various information. Furthermore, the property information may include a floor plan image. The floor plan image is an image showing the floor plan of the target property. The information providing server 1 may transmit property information via a web page, for example.

The information providing server 1 executes a process of recommending properties to a user. For example, the information providing server 1 may provide property information of multiple recommended properties to each user, for example, when the user uses the property information providing system S for the first time that day. The information providing server 1 in particular causes a floor plan image to be displayed on the user terminal 2. The information providing server 1 then causes the user to input an evaluation of the floor plan image. The information providing server 1 generates a model that identifies properties preferred by the user by learning using the floor plan image and the input evaluation. This model is called a property classifier. The information providing server 1 then determines the property to recommend to the user based on the generated property classifier.

Each user terminal 2 is a terminal device that can be carried by a user who can use the property information providing system S. Examples of the user terminal 2 include a mobile information terminal such as a smartphone or a tablet computer, a mobile phone, and a PDA (Personal Digital Assistant). Each user terminal may include a display having a screen and a sensor that detects a contact position on the screen. Contact with the screen means that the user's finger, stylus pen, etc. contacts the screen. The contact position is the coordinate on the screen where the user's finger, stylus pen, etc. contacts. An example of a device that integrates such a sensor and a display is a touch panel. An application dedicated to the property information providing system S may be installed on each user terminal 2. The user can view the property information using this application. In addition, a web browser may be installed on each user terminal 2. It may be possible to view the property information by using the web browser. A stationary terminal device such as a personal computer may be used as the user terminal 2.

2. Configuration of Information Providing Server
Next, the configuration of the information providing server 1 will be described with reference to Fig. 2 and Fig. 3. Fig. 2 is a block diagram showing an example of a schematic configuration of the information providing server 1 according to this embodiment. As shown in Fig. 2, the information providing server 1 includes a system control unit 11, a system bus 12, an input/output interface 13, a storage unit 14, and a communication unit 15. The system control unit 11 and the input/output interface 13 are connected via the system bus 12.

The system control unit 11 is composed of a CPU (Central Processing Unit) 11a, a ROM (Read Only Memory) 11b, a RAM (Random Access Memory) 11c, etc.

The input/output interface 13 performs interface processing between the memory unit 14 and the communication unit 15 and the system control unit 11.

The storage unit 14 includes, for example, a hard disk drive. This storage unit 14 stores databases such as a member DB 14a, property DB 14b, evaluation information 14c, model DB 14d, and the like. "DB" is an abbreviation for database.

FIG. 3 is a diagram showing an example of information stored in the database of the information providing server 1. In the member DB 14a, member information regarding users who can use the property information providing system S is stored for each user. For example, the member DB 14a stores user ID, name, gender, date of birth, address, telephone number, e-mail address, and other user attributes in association with each other as member information. The user ID is identification information that identifies the user.

In the property DB 14b, property information on the property that is the subject of information provision is stored for each combination of the property and the real estate agent who brokers or sells the property. Specifically, in the property DB 14b, property information such as a property ID, agent ID, registration date, property name, detailed information, property image, and floor plan image is stored. The property ID is identification information that identifies the combination of the property and the real estate agent. The agent ID is identification information that identifies the real estate agent who brokers or sells the target property. The registration date indicates the date on which the property information was registered in the property information providing system S. The detailed information is detailed information on the target property. The detailed information includes attributes of the target property. Specifically, the detailed information may include the address, the nearest station, the travel time to the nearest station, the building type, the age of the building, the layout type, the price, the exclusive area, the floor on which the property is located, and other attributes. The building type indicates the type of the target property or the building in which the property is located (for example, a house, condominium, apartment, etc.). The layout type indicates the type and number of rooms that the property has. Examples of floor plan types include "1R", "1K", "1LDK", "2K", "2DK", "2LDK", "3LDK", and "3SLDK". The price indicates the selling price or rent. Other examples of attributes include management fee, security deposit, key money, whether the bathroom and toilet are separate, whether there is flooring, whether there is a place to put a washing machine, whether there is a system kitchen, the type of stove, whether there is a bathroom dryer, whether there is a bath reheating function, whether there is a separate washroom, whether there is a heated toilet seat, whether there is an air conditioner, whether there is a floor heating function, whether there is a walk-in closet, whether it is a corner room, whether there is an auto-lock, whether there is a parking lot, whether there is a delivery box, whether pets are allowed, whether musical instruments are allowed, whether foreigners are allowed to live, whether the property is available for private lodging, the distance to a convenience store, the distance to a supermarket, and the like. Property images are photographs of the interior and exterior of the property.

Furthermore, the property DB 14b may store segmented floor plan information in association with the property ID. The segmented floor plan information is information obtained from a floor plan image of the target property. The segmented floor plan information is information indicating the position and range of each element constituting the target floor plan image. The segmented floor plan information may be, for example, information in which a label indicating an element corresponding to the position of each pixel of the floor plan image is associated with the pixel. The components of the floor plan image are divided into the inside and outside of the property. The components of the inside of the property may be broadly divided into partitions and internal spaces. A partition is something that divides the space inside the property or something that divides the inside and outside of the property. A partition may physically divide a space, or may divide a space imaginarily. Examples of partitions include walls and openings. Examples of openings include entrances and windows. An internal space may be an empty space surrounded by a partition. An internal space may be broadly divided into a room and other spaces. In this embodiment, the term "room" does not refer to a room in an apartment or a condominium, but may refer to an internal space of the target property that is used for a relatively long time as a place of living or working and can be classified according to its purpose. Examples of rooms include a general room, living room, dining room, kitchen, dining kitchen, LDK (a room in which a living room, dining room, and kitchen are integrated), a storeroom, a loft, etc. A general room may be a room that does not fall under any of the living room, dining room, kitchen, dining kitchen, LDK, storeroom, and loft. A general room may be a room that does not have a specific purpose. A general room is called a Japanese-style room or a Western-style room, or a bedroom. Other examples of spaces include a washroom, a toilet, a bath, a corridor, a hallway, a veranda, etc. At least one of the washroom, toilet, and bath may be included in the room.

Furthermore, the property DB 14b may store a feature vector in association with the property ID. The feature vector may be a two-dimensional array of features of a room in the target property. For example, the feature vector may be information that arranges one-dimensional arrays of features of each room in the target property. The one-dimensional array of features of each room may be associated with, for example, a label indicating the room type of the room and the room number of the room. The room type indicates the type of room. For example, room types may include a general room, living room, dining room, kitchen, dining kitchen, living/dining/kitchen area, storeroom, and loft. Room numbers may be assigned sequentially starting from 1 for the rooms in the property.

In the evaluation information DB 14c, evaluation information indicating the user's evaluation of the floor plan image displayed on the user terminal 2 is stored each time an evaluation is performed. Specifically, in the evaluation information DB 14c, a user ID, a property ID, a display date, and an overall evaluation are stored in association with each other as evaluation information. The user ID indicates the user who performed the evaluation. The display date indicates the date on which the floor plan image of the evaluated property was displayed. The date on which the target floor plan image was displayed and the date on which the evaluation was performed usually coincide. The overall evaluation is an evaluation of all rooms in the property shown by the floor plan image. Either "like" or "dislike" may be set as the overall evaluation. "Like" is a positive evaluation of the target floor plan image. "Dislike" is a negative evaluation of the target floor plan image. In the evaluation information DB 14c, one or more pieces of partial evaluation information may be further stored. The partial evaluation information is information regarding the evaluation of one of the rooms in the property shown by the target floor plan image. When a user specifies a part of the target floor plan image and performs an evaluation, the partial evaluation information is stored. The partial evaluation information includes the target room number, the room type, and the partial evaluation. The target room number is the number of the room specified by the user. The room type indicates the type of room specified by the user. The partial evaluation indicates the user's evaluation of the specified room. As with the overall evaluation, either "like" or "dislike" may be set as the partial evaluation. Furthermore, the evaluation information DB 14c may store questionnaire responses. The questionnaire responses are information indicating, when the overall evaluation is "dislike," as the user's response, which part of the target property the user entered the evaluation "dislike."

The model DB 14d stores, for each user, parameters of a property classifier that identifies properties preferred by the user. Specifically, the model DB 14d stores the user ID and the parameters of the property classifier in association with each other. The user ID indicates the user whose preferences are identified by the property classifier. For example, when using a probability density function as a model, the parameters may include coefficients of the function. For example, for a normal distribution, the parameters may include the mean and variance. In the case of a mixed normal distribution, the parameters may include the mean, variance, and weight of each normal distribution that makes up the mixed normal distribution. When a neural network is used as a model, the parameters may include weights of intermediate layers included in the network.

The storage unit 14 further stores various programs such as an operating system, a DBMS (Database Management System), and a server program. The server program is a program that causes the system control unit 11 to perform processing related to providing property information, such as providing floor plan images, obtaining evaluations from users, generating property classifiers, and determining properties to recommend to users. It is. For example, the server program may be obtained from another device via the Internet NW, or may be recorded on a recording medium such as a magnetic tape, an optical disk, or a memory card and read via a drive device. Good too.

The communication unit 15 is composed of, for example, a network interface card. The communication unit 15 connects to the user terminal 2 and the like via the network NW, and controls the communication state with these devices.

[3. Functional overview]
Next, an overview of the functions of the system control unit 11 in the information providing server 1 will be explained using FIGS. 4 to 12. FIG. 4 is a diagram showing an example of functional blocks of the system control unit 11 of the information providing server 1. The system control unit 11 includes a display control unit 111, an evaluation information acquisition unit 112, a feature information acquisition unit 113, and a generation unit, as shown in FIG. 114, a determining unit 115, etc.

[3-1. Display of floor plan image]
The display control unit 111 displays floor plan images of each of the multiple properties on the user terminal 2. The property for which the floor plan image is displayed may be a property recommended to the target user. The recommended property is determined using a property classifier, as described below. However, when the property classifier is not generated, the recommended property may be any property. For example, a property for which the number of days elapsed since the registration date of the property is less than a predetermined value may be recommended. Alternatively, a search condition for a property may be input and saved in advance by the user. As the search condition, the same information as the information included in the detailed information of the property information may be specified. For example, the search condition may be stored in the member DB 14a or the like. Then, a property that matches the saved search condition may be recommended. In addition, the display control unit 111 may redisplay the floor plan image of at least one property among the multiple properties for which the floor plan image has been displayed once after a predetermined number of days (for example, one week, one month, etc.) have passed since the floor plan image was displayed. While searching for a property, the user's preferences for properties may change as time passes or as the user views information on various properties. Therefore, a rating once input by a user may change thereafter. In order to reflect the change in the user's preference in the property classifier, the floor plan image may be redisplayed. Here, the display control unit 111 may redisplay the floor plan image only for the property for which the target user has input a positive rating of "like," which will be described later.

The display control unit 111 may display the floor plan image on the user terminal 2, for example, when the target user logs in to the property information providing system S for the first time on that day. Further, the display control unit 111 may also display property information such as detailed information indicating attributes of the property on the user terminal 2. For example, the display control unit 111 may transmit the property name, detailed information, property image, and floor plan image of the recommended property to the user terminal 2. The display control unit 111 may further transmit segmented floor plan information to the user terminal 2.

The user terminal 2 displays the floor plan image received from the information providing server 1. FIG. 5(a) is a diagram showing an example of a display of a floor plan screen. As shown in FIG. 5(a), the user terminal 2 displays a floor plan screen 100. The floor plan screen 100 includes a floor plan image 110. FIG. 5(b) is a diagram showing an example of a display of a property details screen. When the floor plan image 110 is displayed and the user performs a predetermined operation, the user terminal 2 displays a property details screen 200 as shown in FIG. 5(b). For example, the property details screen 200 may be displayed by the user swiping the floor plan screen 100 to the right or left. Swiping may be an operation in which the user presses the screen with a finger and slides the finger. The property details screen 200 is a screen that displays detailed information about the property. For example, the property details screen 200 may display a property image, detailed information, and information about a real estate agent who handles the target property. For example, the property details screen 200 may display various information by the user scrolling the property details screen 200 up and down. Additionally, on the property details screen 200, the user may be able to register the target property as a favorite, or request information or make inquiries to real estate agents.

[3-2. Input evaluation]
The evaluation information acquisition unit 112 acquires, for each of a plurality of properties whose floor plan images are displayed by the user terminal 2, evaluation information indicating the user's evaluation of the displayed floor plan images. The evaluations that can be input by the user may include at least positive evaluations and negative evaluations. The positive evaluation may be an evaluation that the user likes the room or floor plan in the property shown by the floor plan image. The negative evaluation may be an evaluation that the user dislikes the interior or floor plan of the property shown by the floor plan image. A positive evaluation is higher than a negative evaluation. The user may be able to input any evaluation from three or more levels of evaluation. Even if a user potentially has a desire regarding a room or floor plan within a property, the user may not be aware of it. Alternatively, even if the user has a desire in mind, it may be difficult to express it in words. By allowing an intuitive evaluation to be input, the user does not have to input which points are good or bad for specific reasons.

The evaluation information acquisition unit 112 may acquire evaluation information indicating a positive evaluation when a movement in a predetermined first direction of a position touched by a finger or the like is detected on the screen on which the floor plan image is displayed. This allows the user to easily input an evaluation. Also, the evaluation can be input by an intuitive operation. The operation at this time may be, for example, a flick. A flick may be an operation in which the user flicks the screen with a finger or the like. The difference between a flick and the above-mentioned swipe may be at least one of the movement distance and the length of time from when the finger or the like starts to move until it stops or until the finger or the like is released from the screen. For example, when the movement distance is less than a predetermined distance, the operation may be a flick, and when the movement distance is equal to or greater than the predetermined distance, the operation may be a swipe. Also, when the length of time is less than a predetermined time, the operation may be a flick, and when the length of time is equal to or greater than the predetermined time, the operation may be a swipe. The first direction may be any of up, down, right, left, and diagonal. In this embodiment, a movement in the right direction is a positive evaluation.

Further, the evaluation information acquisition unit 112 provides a negative evaluation when a movement in a second direction different from the first direction of the position touched by a finger or the like on the screen on which the floor plan image is displayed is detected. You may also acquire evaluation information shown. The operation at this time may also be a flick. Further, the second direction may be any direction different from the first direction. In this embodiment, leftward movement is a positive evaluation.

The evaluation information acquisition unit 112 may acquire the evaluation based on the above-described operation as the overall evaluation. That is, the input evaluation is associated with all rooms in the property shown by the floor plan image.

FIG. 6A is a diagram showing an example of inputting a positive evaluation for a floor plan image and an example of evaluation for each room. As shown in FIG. 6A, the user flicks the floor plan image 110 to the right. As a result, a positive evaluation "like" is input. The property shown by the floor plan image 110 has an LDK, a general room A of 7 tatami mats, and a general room B of 5.5 tatami mats. Since the evaluation in this case is for the entire floor plan 110, the evaluation "like" may be set for all the features of the LDK, general room A, and general room B, respectively. FIG. 6(b) is a diagram showing an example of inputting a negative evaluation for a floor plan image and an example of evaluation for each room. As shown in FIG. 6(b), the user flicks the floor plan image 110 to the left. As a result, a negative evaluation "dislike" is input. In this case, the evaluation "dislike" may be set for all the characteristics of the LDK, general room A, and general room B, respectively.

The evaluation information acquisition unit 112 may further acquire partial evaluation information indicating the user's evaluation of a portion specified by the user in the floor plan image of at least one of the properties for which the floor plan image is displayed by the user terminal 2. This allows the user to input a different evaluation for only some of the rooms in the property. Even if the user wishes to input a specific evaluation for the property as a whole, there is a possibility that the user may wish to input a different evaluation for some of the rooms.

For example, the evaluation information acquisition unit 112 may acquire partial evaluation information indicating a positive evaluation when a movement in a predetermined third direction of a position touched by a finger or the like is detected on a screen on which a floor plan image is displayed, which is an operation in a different manner from the operation in the case of the overall evaluation. An operation in a different manner from the operation in the case of the overall evaluation may be, for example, a different direction of movement of a finger or the like. For example, when a flick to the right is a positive evaluation as an overall evaluation, a flick to the upward direction may be a positive evaluation as a partial evaluation. Alternatively, when a user touches a finger or the like on the screen and stops for a predetermined time or more, and then performs a flick, the operation may be determined to be an input of a partial evaluation. In this case, the third direction may be, for example, a rightward direction. This flick is called a hold flick. In contrast to the hold flick, a flick performed by a user without stopping for a predetermined time or more after touching a finger or the like on the screen is called a normal flick. A normal flick is an operation for inputting an overall evaluation. The room at the position of the finger or the like when the user starts to touch the screen with the finger or the like may be the subject of the partial evaluation. For example, as described above, segmented floor plan information may be transmitted to the user terminal 2 together with the property information. The user terminal 2 may determine where the user specified in the floor plan image by acquiring, from the segmented floor plan information, a label corresponding to the pixel at the position touched by the user. Furthermore, the evaluation information acquisition unit 112 may acquire partial evaluation information indicating a negative evaluation when an operation different from the operation in the case of an overall evaluation is detected, that is, when movement of the position touched by a finger or the like on the screen on which the floor plan image is displayed in a predetermined fourth direction different from the third direction is detected. This operation may also be a hold flick. Furthermore, the fourth direction may be to the left.

FIG. 7 is a diagram showing an input example of partial evaluation and overall evaluation for a floor plan image, and an example of evaluation for each room. As shown in FIG. 7, the user placed his finger on a portion 120 on the displayed floor plan image 110 and flicked it toward the right. Portion 120 is a point on the LDK. Thereafter, the user normally flicked left on the floor plan image 110. In this case, the evaluation “like” may be set for all the features of the LDK, and the evaluation “dislike” may be set for all the features of the general room A and the general room B, respectively.

When the user inputs "dislike" as the overall evaluation, the user terminal 2 may display a questionnaire screen. The questionnaire screen is a screen for the user to select which aspect of the target property he/she dislikes. Options include "floor plan" in addition to information items included in the detailed information (for example, address, nearest station, travel time, building type, age, floor plan type, price, exclusive area, etc.). "Floor plan" is an option when you do not like the room or floor plan in the property shown by the floor plan image. The evaluation information acquisition unit 112 may acquire the user's selection from the user terminal 2 as a questionnaire response. This questionnaire is used to determine whether or not information about the property whose floor plan image is displayed is to be used for model learning.

The evaluation information acquisition unit 112 may acquire evaluation information when the user performs an operation other than a slide operation such as a flick. For example, an element such as a button or a menu for selecting an evaluation may be displayed on the floor plan screen 100. The evaluation information acquisition unit 112 may acquire a table or information indicating an evaluation according to the user's operation on the held element.

[3-3. Feature extraction]
The characteristic information acquisition unit 113 acquires characteristic information generated based on the floor plan images for each of a plurality of properties whose floor plan images are displayed by the user terminal 2, which indicates the characteristics of the rooms in the properties. Get feature information. This feature may be a feature for each room. Moreover, the acquired features are not particularly limited as long as they are features that can be identified from the floor plan image. The acquired feature may be a feature that cannot be specified or is difficult to specify from the detailed information of the property information. The acquired feature information may be a feature amount.

The feature information acquisition unit 113 may identify an area corresponding to a room from the floor plan image. Alternatively, the room type may be identified from the floor plan image for each room included in the target property. The feature information acquisition unit 113 may generate segmented floor plan information from the floor plan image. For example, the feature information acquisition unit 113 may generate segmented floor plan information using semantic segmentation.

For example, the feature information acquisition unit 113 may generate segmented floor plan information using machine learning. For example, a convolutional neural network (CNN) that can be used for semantic segmentation may be used as a model for generating segmented floor plan information. By inputting a floor plan image to this model, segmented floor plan information will be output from this model. For example, in the learning stage, a floor plan image and segmented floor plan information may be prepared as training data. The segmented floor plan information in this case may be created manually, for example. As preprocessing, the feature information acquisition unit 113 recognizes characters indicating the area of an internal space such as a room from the floor plan image, for example, by character recognition. For example, characters such as "7 jo", "5 tatami", and "20 m2" indicate the area. The feature information acquisition unit 113 may remove characters indicating area from the floor plan image. Then, the feature information acquisition unit 113 trains the model using the preprocessed floor plan image as input data and the prepared segmented floor plan information as the ground truth. Then, the feature information acquisition unit 113 generates segmented floor plan information from the floor plan image using the trained model. In the segmented floor plan information generated at this time, if there are multiple rooms of the same type, the same label may be attached to all of the rooms. In that case, the feature information acquisition unit 113 may relabel each room separately (for example, labels such as "general room A", "general room B", "general room C", etc.).

The feature information acquisition unit 113 may generate segmented floor plan information from the floor plan image, for example, on a rule basis. For example, the feature information acquisition unit 113 may remove characters indicating the area from the floor plan image. In addition, the feature information acquisition unit 113 may emphasize necessary edges and remove unnecessary edges by performing morphological transformation or binarization on the floor plan image. The feature information acquisition unit 113 may identify the type of each internal space by pattern recognition from the floor plan image processed in this way. For example, the feature information acquisition unit 113 may identify the type by character recognition. For example, the kitchen is given letters such as "kitchen" and "K". The dining room is given letters such as "dining" and "D". The general rooms are given letters such as "Western-style room", "Japanese-style room", "Western", "Japanese", and "bedroom". The living/dining/kitchen area is given letters such as "LDK". The entrance is given letters such as "entrance" and "entrance". The storage is given letters such as "storage", "storage", and "closet". The feature information acquisition unit 113 may also identify the type by image recognition such as object detection. For example, a picture of a sink is drawn in the kitchen. A picture of a toilet bowl is drawn in the toilet. A picture of a bath heater is drawn in the bath. The feature information acquisition unit 113 may then remove the recognized characters and pictures from the floor plan image. The feature information acquisition unit 113 may also remove the sector from the floor plan image. The sector drawn in the floor plan image usually indicates the direction in which a hinged door opens. The feature information acquisition unit 113 may also identify figures formed by line segments from the floor plan image. Then, the feature information acquisition unit 113 may label the area surrounded by each figure. The figure including the part where the above-mentioned character or picture is recognized usually corresponds to the internal space of the property. The feature information acquisition unit 113 may associate a label indicating the type identified from the recognized character or picture with all pixels in the area corresponding to the internal space. A figure in which neither character nor picture is recognized may also correspond to the internal space. For example, a location adjacent to the entrance and connected to another internal space through an opening is likely to be a corridor. Other figures adjacent to the figure corresponding to the internal space and having a relatively thin shape are usually partitions. Among figures corresponding to partitions, rectangles with one long side that contacts only one internal space and the other long side that contacts only one internal space or contacts the outside of the property usually correspond to openings. Among rectangles corresponding to openings, rectangles with both long sides that contact only one internal space correspond to doorways. Other openings are windows. However, an opening that leads from the entrance to the outside of the property is a doorway, and an opening that leads to a balcony is a window. Also, an opening that leads to a storage area is a storage door. In this way, the feature information acquisition unit 113 performs labeling and generates segmented floor plan information.

FIG. 8 is a diagram illustrating an example of generation of segmented floor plan information. As shown in FIG. 8, segmented floor plan information 300 is generated from the floor plan image 110. The segmented floor plan information 300 includes the exterior of the property 310, walls 320-1 to 320-5, doorways 330-1 to 330-7, storage doors 330-8 to 330-10, and windows 340-1 to 340-4. , consists of rooms 350-1 to 350-3, entrance 360-1, washroom 360-2, toilet 360-3, bath 360-4, balcony 360-5, storage 370-1 to 370-3, etc. . Room 350-1 is an LDK, room 350-2 is room A, and room 350-3 is room B.

When the area corresponding to the room is thus identified from the floor plan image, the feature information acquisition unit 113 may identify the features of the room based on the identified area. For example, the feature information acquisition unit 113 may identify the features of the room based on the features of the area. The feature information acquisition unit 113 may extract a feature amount as the feature of the room. For example, the feature information acquisition unit 113 may identify the area, number of vertices, length of sides, and angle of interior angles (for example, only minor angles) of the area (figure forming the area). Then, the feature information acquisition unit 113 may identify the area, number of vertices, length of sides, and angle of the corners of the room as the feature amount of the room. The area, vertices, sides, and interior angles of the room are the area, vertices, sides, and interior angles of a figure corresponding to the range that the room occupies in the property when, for example, the inside of the property is viewed from above. In this case, the side of the partition corresponds to the side or part of the side of the figure. For example, at the stage of generating segmented floor plan information, the area of at least one interior space is identified by character recognition. The characteristic information acquisition unit 113 may calculate the ratio of the area of the region corresponding to the internal space to the area of the region corresponding to the other internal space in the floor plan image. Then, the characteristic information acquisition unit 113 may calculate the area of the other internal space based on the area of the internal space and the calculated ratio. Also, the characteristic information acquisition unit 113 may calculate the length of the side of the figure surrounding the region corresponding to each room. Also, the characteristic information acquisition unit 113 may calculate the square root of the ratio between the area of the region and the area of the room corresponding to the region. Then, the characteristic information acquisition unit 113 may calculate the length of the side surrounding the room based on the length of the side of the region and the calculated square root. For example, the characteristic information acquisition unit 113 may calculate the length of only the shortest side and the smallest side among the sides surrounding the room. The characteristic information acquisition unit 113 may specify the number of vertices of the region and the angle between the sides as the number of vertices of the region and the angle between the sides, respectively. The characteristic information acquisition unit 113 may specify only the minimum angle and the maximum angle. The characteristic information acquisition unit 113 may identify at least one of the area, vertices, sides, and interior angles of the room. Alternatively, the characteristic information acquisition unit 113 may identify at least one of the vertices, sides, and interior angles.

Further, the feature information acquisition unit 113 may specify the adjacency status between the area corresponding to the target room and other areas. Based on the identified adjacency situation, the feature information acquisition unit 113 may specify the adjacency situation of other interior spaces or partitions to the target room as the feature amount of the room. Here, the feature information acquisition unit 113 may specify a feature related to an opening provided in the target room. For example, the feature information acquisition unit 113 may specify the number of windows and the number of doorways in the room as the feature amount. In the floor plan image, a window that adjoins a room is a window of that room. For example, in the segmented floor plan information 300, room 350-1 is adjacent to windows 340-1 and 340-2. Therefore, the number of windows is two. In the floor plan image, the doorway that adjoins a room is the doorway of that room. For example, room 350-1 is adjacent to doorways 330-2, 330-6, and 330-7. Therefore, the number of entrances and exits is three. The feature information acquisition unit 113 may specify the number of rooms adjacent to the room as the feature amount. In the floor plan image, other rooms that are adjacent to the target room across a partition are adjacent to the target room. There is no need to consider whether or not the rooms are connected by a doorway. For example, in the segmented floor plan information 300, room 350-2 is adjacent to room 350-1 across entrance 330-7, and adjacent to room 350-3 across wall 320-6. Therefore, the number of adjacent rooms is 2. The feature information acquisition unit 113 may specify the area of storage in the room as the feature amount. In the floor plan image, a storage space that is adjacent to the target room with a storage door in between is a storage space in that room. For example, room 350-2 is adjacent to storage 370-2 with storage door 330-9 in between. Therefore, room 350-2 has storage 370-2. The area of storage 370-2 may be specified as the area of storage in room 350-2. In the case of a room having a plurality of storages, the feature information acquisition unit 113 may calculate the total area of the storages as the area of the storages in the room.

The feature information acquisition unit 113 may also specify the positional relationship between the area corresponding to the target room and the area corresponding to other internal spaces. The feature information acquisition unit 113 may also specify the positional relationship between the target room and other internal spaces as the feature of the room. For example, the feature information acquisition unit 113 may specify whether the room is connected to a corridor, whether the room is connected to a living room or living/dining/kitchen area, whether the room is connected to a balcony, etc., as features. In the floor plan image, the internal space adjacent to the target room across the entrance is connected to the room. The feature information acquisition unit 113 may also specify whether it is necessary to pass through an internal space other than the corridor to go from the room to the entrance. This feature can be specified by a path search algorithm such as Dijkstra's algorithm. The feature information acquisition unit 113 may also specify whether the entrance to the room is a hinged door or a sliding door as the feature of the room. In the floor plan image, the entrance where the area corresponding to the entrance is in contact with the sector may be a hinged door. The other entrances may be sliding doors. For features that are indicated by whether or not a certain condition is met, for example, a feature value may be set to 1 if the condition is met, and a feature value may be set to 0 if the condition is not met.

The feature information acquisition unit 113 may acquire feature amounts of the same items for all types of rooms. The feature amount items are, for example, the categories of room features such as "room area" and "number of vertices" when the room features are divided by name. Alternatively, the feature information acquisition unit 113 may make all or part of the acquired feature quantity items different from the feature quantity items acquired in other rooms, depending on the room type.

9 is a diagram showing an example of the configuration of a feature vector of a property acquired from a floor plan image. The feature information acquisition unit 113 may generate a feature vector composed of the extracted features. Here, as an example, the area, number of vertices, the length of the longest side, the length of the shortest side, the maximum angle, the minimum angle, the number of windows, the number of entrances, the number of adjacent rooms, and the area of the storage are extracted as the features of the room. As shown in FIG. 9, in the feature vector 400, for each room, 10 items of feature values for the room are stored. In FIG. 9, feature values are stored for the living room, dining room, kitchen, and kitchen area, general rooms A and B identified from the floor plan image 110. Here, the area affects the livability of the room and how much furniture, etc., can be placed in it. The number of vertices indicates the number of corners. The length of the longest side affects whether or not relatively long furniture, such as a bed, desk, and TV stand, can be installed. The longer the maximum angle of the shortest side is greater than 90 degrees and farther from 90 degrees, the more likely it is that dead space will be created when furniture, etc. is installed. The smaller the minimum angle is and the further away it is from 90 degrees, the more likely it is that dead space will be created when furniture, etc. is installed. The number of windows affects the amount of light. The more entrances there are, the more paths there are to enter and exit the room, but the more limited the space available for installing furniture, etc. The fewer the number of adjacent rooms, the more likely it is that the room is located at a corner of the property. The storage area affects the amount of items that can be stored.

The feature information acquisition unit 113 may generate the segmented floor plan information and feature vectors when or after the floor plan image of the target property is displayed. Alternatively, the feature information acquisition unit 113 may generate the segmented floor plan information and feature vectors in advance and store the information in the property DB 14b.

Note that the feature information acquisition unit 113 itself does not need to generate the feature information. For example, a device other than the information providing server 1 may generate the feature information, and the feature information acquisition unit 113 may acquire the feature information from that device.

[3-4. Model Generation]
The generating unit 114 generates a property classifier, which is an example of a model for identifying properties that a user likes, based on the evaluation information acquired by the evaluation information acquiring unit 112 and the feature information acquired by the feature information acquiring unit 113. The generated property classifier may be a classifier that uses feature information of all rooms of a property as input data and outputs estimated evaluation information indicating an evaluation that is estimated to be input by a user for the property as output data. The output estimated evaluation information may indicate either a positive evaluation of "like" or a negative evaluation of "dislike", or may indicate the degree or probability that the user likes the property. In addition, the generated property classifier may be either a statistical model or a machine learning model, as long as it uses evaluation information and feature information as learning data or training data.

The generation unit 114 may generate a property classifier so as to identify a property that has relatively many features that the user likes and relatively few features that the user does not like as a property that the user likes. . The generation unit 114 determines that there is a room that has a certain feature in which the percentage of features that are given a "like" rating is higher than the percentage of features that are given a "dislike" rating. A property classifier may be generated that identifies properties preferred by the user from among properties. Even if the user likes a specific characteristic of a certain item of a specific type of room among the rooms in a certain property, the user does not necessarily input an evaluation "like" for the floor plan image of the property. If the user does not like other features or other rooms, the user may enter a "dislike" rating. However, as the number of properties evaluated by the user increases, the proportion of that specific feature among the features for which a characteristic item of a specific type of room is rated "like" converges to, for example, the first proportion. . Along with this, the proportion of the specific feature among the features for which the characteristic item of the specific type of room has been given an evaluation of "dislike" converges to, for example, the second proportion. Here, the first ratio is higher than the second ratio. For example, suppose a user likes a regular room with two windows. However, there are fewer rooms with two windows than there are rooms with one window. Therefore, it is highly likely that the ratio of two windows to the number of windows in a general room that has been given a "like" rating is smaller than the ratio of one window. Similarly, it is highly likely that the ratio of two windows to the number of windows in a general room that has been given a rating of "dislike" is smaller than the ratio of one window. Therefore, when comparing the ratings "like" and "dislike", the ratio of two windows to the number of windows for ratings "like" is greater than the ratio of two windows to the number of windows for ratings "dislike". also becomes larger. On the other hand, the ratio of one window to the number of windows with an evaluation of "like" is smaller than the ratio of one window to the number of windows with an evaluation of "dislike." From this, it can be inferred that the user relatively "likes" a general room with two windows, and relatively "dislikes" a general room with one window. If at least one feature does not satisfy such conditions, it is considered that the user does not like the property. When using a statistical model as a property classifier, it is relatively easy to design such a property classifier. When using a machine learning model as a property classifier, it is considered that the parameters are optimized during the learning stage, so that the property classifier automatically learns this point.

The generation unit 114 may also generate a property classifier that shows multiple probability distributions corresponding to multiple ratings including "like" and "dislike," with the property feature information from which rating information indicating the rating corresponding to the probability distribution has been obtained being a random variable. In this case, the property classifier is a model that identifies properties that the user prefers based on the multiple probability distributions. Each probability distribution can be said to be information indicating the proportion of each feature among the features that have been assigned a rating corresponding to that probability distribution.

FIG. 10 is a diagram showing an example of the configuration of a property classifier. As shown in FIG. 10, the property classifier 500 may include multiple room models 510. Each room model 510 is a model for estimating a user's evaluation of a room of a type corresponding to that room model 510. By generating a model for each room type, the user's evaluation of each room can be reflected in the output of the property classifier. In FIG. 10, the room models 510 include a model for a general room, a model for a living room, a model for a dining room, a model for a kitchen, a model for an LDK, a model for a dining kitchen, a model for a storeroom, and a model for a loft. The property classifier 500 may further include an evaluation score calculation unit 520. The evaluation score calculation unit 520 generates estimated evaluation information used to identify a property preferred by the user based on information indicating the evaluation of the room output from each of the all-room models 510.

Each room model 510 may include a plurality of feature item models 530. Each feature item model 530 is a model for estimating the user's evaluation of the feature amount of the item corresponding to the feature item model. Each room model 510 outputs a plurality of feature item scores output from each of a plurality of feature item models 530 described later to the evaluation score calculation unit 520. Each feature amount has no relationship or low relevance with other feature amounts. Therefore, by generating a model for each feature, the structure of the model can be simplified. In FIG. 10, the feature item models 530 include a model for the area of the room, a model for the number of vertices, a model for the length of the longest side, a model for the length of the shortest side, and a model for the maximum angle. , a model for the minimum angle, a model for the number of windows, a model for the number of doorways, a model for the number of adjacent rooms, and a model for the storage area.

Each feature item model 530 may include a like model 540 and a dislike model 550. The like model 540 may output a like score indicating the degree to which it is estimated that the user will input an evaluation of "like" for the feature amount of the target feature item of the target room type. The dislike model 550 may output a dislike score indicating the degree to which it is estimated that the user will input an evaluation of "dislike" for the feature amount of the target feature item of the target room type. Each feature item model 530 may further include a feature item score calculation unit 560. The feature item score calculation unit 560 may integrate the like score and dislike score that are output from the like model 540 and the dislike model 550, and output the feature item score that is output from the feature item model 530.

The generation unit 114 generates a room model 510 for each of the plurality of room models 510 based on the feature amount of the type of room corresponding to the room model 510 and the evaluation information corresponding to the property including the room. It's okay. For example, the generation unit 114 generates the liking model 540 for each combination of room type and feature item, using the feature amount of that item of the type of room for which the user inputs the evaluation "like" as learning data. Good too. In addition, for each combination of room type and feature item, the feature information acquisition unit 113 generates a dislike model 550 using the feature amount of the item of the type of room for which the user inputs the evaluation "dislike" as learning data. You may. However, the generation unit 114 may exclude from the learning data the feature amounts for properties for which responses other than "floor plan" have been input as questionnaire responses for which the evaluation "dislike" has been input. In this case, it is considered that the user has not given any evaluation to the floor plan image. It is more appropriate not to use information on such properties for learning purposes. When using a normal distribution or a mixed normal distribution as the liking model 540 and the disliking model 550, the feature information acquisition unit 113 may generate the models using, for example, a maximum likelihood estimation method. The number of normal distributions that make up the mixed normal distribution is not particularly limited. As a result of generating all the liking models 540 and all the disliking models 550, the models 530 for all feature items are generated, and the models 510 for all rooms are generated. As a result, a property classifier 500 is generated.

For a property for which a partial evaluation has been input by the user, the generation unit 114 may generate a property classifier using the partial evaluation included in the evaluation information for the rooms included in the property that are specified by the user, and using the overall evaluation included in the evaluation information for the rooms included in the property other than the room specified by the user. For example, the generation unit 114 may use the feature amount of the room to generate a model corresponding to the partial evaluation among the like model 540 and the dislike model 550 for the all feature item model 530 of the room model 510 corresponding to the room specified by the user. On the other hand, the generation unit 114 may use the feature amount of the room to generate a model corresponding to the overall evaluation among the like model 540 and the dislike model 550 for the all feature item model 530 of the room model 510 corresponding to a room not specified by the user. In this way, for a property in which the evaluation of some rooms differs from the overall evaluation, the evaluation of the rooms can be reflected in the property classifier 500.

The generation unit 114 may generate the property classifier 500 for each user. This makes it possible to personalize which properties are recommended to the user.

When a neural network is used as the property classifier, the property classifier may include, for example, multiple fully connected layers as intermediate layers. The generation unit 114 inputs feature information to this property classifier and causes it to output estimated evaluation information. The generation unit 114 may train the property classifier by backpropagation using the evaluation information acquired by the evaluation information acquisition unit 112 as ground truth.

[3-5. Deciding on recommended properties]
The determining unit 115 determines the property to be presented to the user based on the property classifier 500 generated by the generating unit 114. The determining unit 115 selects candidate properties, for example, according to predetermined conditions. For example, the determining unit 115 may select candidates at random, or may select properties for which the number of days that have passed since the registration date of the property is less than a predetermined value as a candidate. Further, the determining unit 115 may select properties that match the saved search conditions as candidates. The determining unit 115 acquires feature information, such as a feature vector, of the selected candidate. The determining unit 115 inputs the feature quantity included in the feature vector to each of the liking model 540 and disliking model 550 in the feature item model 530 of the room model 510 corresponding to the feature quantity. Then, the determining unit 115 obtains a liking score and a disliking score from the liking model 540 and the disliking model 550, respectively. Note that if the candidate property does not have a specific type of room, all the liking models 540 and all the disliking models 550 of the room model 510 corresponding to that type each output a score of 0. Good too.

The determination unit 115 may calculate the feature item score as the feature item score calculation unit 560 of each feature item model 530, for example, by subtracting the dislike score from the like score. FIG. 11 is a diagram showing an example of calculation of feature item scores. In FIG. 11, a graph 610 shows a probability distribution shown by the trained like model 540. A graph 620 shows a probability distribution shown by the trained dislike model 550. Each of the graphs 610 and 620 shows a mixed normal distribution composed of two normal distributions. In a mixed normal distribution, the integral of the probability density is 1. The probability density y of the feature amount x satisfies 0<y<1. It is assumed that the graphs 610 and 620 show a probability distribution corresponding to, for example, the area of a living room. Here, the area x of the living room of a certain property is input as a feature amount to the model for room area included in the model for the living room. In the graph 610, the probability density of the area x is 0.3. The like model 540 may output this probability density as a like score. The higher the like score, the more likely it is that the user likes the area of the living room. In graph 620, the probability density of area x is 0.1. Dislike model 550 may output this probability density as the dislike score. The higher the dislike score, the more likely it is that the user dislikes the area of the living room. From these scores, determination unit 115 determines 0.2 as the feature item score for the area of the living room.

When the feature item score is greater than 0, it is likely that the user likes the target feature of the target room. On the other hand, when the feature item score is less than 0, it is likely that the user dislikes the target feature of the target room. The larger the absolute value of the feature item score, the more likely it is that the user likes or dislikes the target feature of the target room. Therefore, the smaller the absolute value, the more likely it is that the user neither likes nor dislikes the target feature of the target room.

The difference in probability density (the difference between the like score and the dislike score) is the overall The larger the value, the higher the possibility that the user has input the evaluation with emphasis on the characteristics of the target room. For example, the larger the difference between the like score and the dislike score for the feature that has the maximum like score, or the greater the difference between the like score and the dislike score for the feature that has the largest dislike score. , there is a high possibility that users consider it important. FIG. 12 is a diagram illustrating an example of the difference between the probability distribution of the feature amount for the evaluation "like" and the probability distribution for the feature amount for the evaluation "dislike." The upper part of FIG. 12 shows a graph 610 and a graph 620 overlapping each other, each showing the probability density corresponding to the area of the living room. The upper part of FIG. 13 shows a graph 630 and a graph 640 overlapping each other, each showing the probability density corresponding to the storage area of a general room. A graph 630 shows the probability distribution represented by the preference model 540 included in the storage area model in the general room model. A graph 640 shows the probability distribution shown by the dislike model 550 included in the storage area model in the general room model. Compared to the difference between the like score and dislike score for the storage area of a general room, the difference between the like score and dislike score for the area of the living room is generally large. Therefore, it is highly likely that the user values the area of the living room more than the storage area of a general room.

The determination unit 115, as the evaluation score calculation unit 520, may integrate the feature item scores output from the all-feature item model of the all-room model to calculate an evaluation score as estimated evaluation information. For example, the evaluation score calculation unit 520 may calculate the total or average value of all feature item scores as the evaluation score. Alternatively, the evaluation score calculation unit 520 may extract feature item scores in which at least one of the like score and the dislike score before calculating the feature item score is equal to or greater than a first predetermined value, and the absolute value of the feature item score is equal to or greater than a second predetermined value. Then, the evaluation score calculation unit 520 may calculate the total or average value of only the extracted feature item scores as the evaluation score. This is a measure to prevent evaluations of features that the user does not consider important from being reflected in the final evaluation score.

If the evaluation score is equal to or greater than a predetermined threshold, the determination unit 115 may determine that the selected candidate property is the user's favorite property and present the property to the user. The threshold may be, for example, 0 or a value greater than 0. Setting the threshold to 0 or greater ensures that the property preferred by the user is identified from among properties that have rooms with a certain characteristic in which the proportion of the certain characteristic among the characteristics that are rated "like" is higher than the proportion of the certain characteristic among the characteristics that are rated "dislike."

The determination unit 115 may determine a property to be presented to the target user based on the property classifier 500 generated for other users having the same attributes as the target user's attributes. Users with the same attributes may have the same or similar property preferences. For example, when a property classifier 500 corresponding to the target user has not been generated, or when the estimation accuracy of the property classifier 500 is low due to a small number of properties for which the target user has input ratings, the property preferred by the target user can be appropriately identified by using the property classifier 500 corresponding to a user having the same attributes. For example, the determination unit 115 may calculate the estimation accuracy of the generated property classifier 500 using characteristic information of an already-rated property and its evaluation information. That is, the determination unit 115 may input characteristic information to the property classifier 500 to obtain estimated evaluation information. The determination unit 115 may calculate the rate at which the estimated evaluation information and the evaluation information match as the estimation accuracy. When the estimation accuracy is less than a predetermined value, the determination unit 115 may use the property classifier 500 generated for another user.

The determination unit 115 may determine which user's property classifier 500 to use, for example, using attributes such as gender, date of birth, and address stored in the member DB 14a. The determination unit 115 may also determine which user's property classifier 500 to use, based on the attributes of the target user registered in a system other than the property information system S. For example, a user who uses the property information system S may be able to use other systems using the same user ID as the user's user ID. In this case, the determination unit 115 may obtain information on other users who have the same attributes as the target user's attributes by sending a request including the user ID to that system. The other system may be, for example, a website, or a system that can be used through a dedicated application that can be installed on the user terminal 2. Examples of such systems include an online shopping mall, a hotel reservation system, a job hunting support system, a marriage activity support system, a web search, a flea market, a ticket sales site, and the like. For example, the target user is using a job hunting support system. Therefore, the determination unit 115 may identify a user who is on the same campus at the same university as the target user, who is in the same industry as the company to which the target user is applying, and who is located in the same location as the company to which the target user is applying. The determination unit 115 may obtain, for example, information indicating other users who have the same attributes as the attributes registered by the user in the system from the system. The determination unit 115 may also obtain, for example, information indicating an action taken by the target user in the system or information indicating other users who have the same attributes as the target user's attributes as identified from the history of the actions taken from the system.

The determination unit 115 may determine the property for which the floor plan image is to be displayed when the display control unit 111 causes the user terminal 2 to display floor plan images of multiple properties. Alternatively, the determination unit 115 may determine the property to be presented to the user, for example, every time a predetermined time has elapsed (for example, every 12 hours, every day, every three days, etc.).

The system control unit 11 presents the information of the property determined by the determination unit 115 as the property to be presented to the user to the target user in some manner. The determined property may be presented as a property recommended to the user. For example, as described above, the display control unit 111 may display a floor plan image and property information of the determined property on the user terminal 2 of the user when the user logs in for the first time that day. Alternatively, the display control unit 111 may display the property information and floor plan image of the determined property on the user terminal 2 at any timing after the user logs in to the property information providing system S, or when the user performs an operation to request a recommended property, or may display a list of the determined properties on the user terminal 2. Alternatively, the system control unit 11 may send an email including the property information and floor plan image of the determined property to the user. Alternatively, the system control unit 11 may send a message to the user terminal 2 by push notification when a property to be presented to the user is found, indicating that a recommended property exists. When the user selects the message displayed on the user terminal 2, the user terminal 2 starts an application dedicated to the property information providing system S, and in response, the system control unit 11 may cause the user terminal 2 to display property information and a floor plan image of the selected property.

[4. Operation of property information provision system]
Next, the operation of the property information providing system S will be explained using FIGS. 13 to 17. The system control unit 11 of the information providing server 1 executes the processes shown in FIGS. 13, 14, 15, and 17 according to various program codes included in the server program.

Figure 13 is a flowchart showing an example of floor plan analysis processing by the system control unit 11 of the information providing server 1 according to this embodiment. For example, when property information input by a real estate agent is stored in the property DB 14b, the system control unit 11 may execute floor plan analysis processing on the property information.

13, the feature information acquisition unit 113 acquires a floor plan image from the property information (step S101). Next, the feature information acquisition unit 113 generates segmented floor plan information based on the floor plan image (step S102). Next, the feature information acquisition unit 113 selects one of the rooms in the property indicated by the segmented floor plan information (step S103). Next, the feature information acquisition unit 113 selects one of a plurality of predetermined feature items (step S104). Next, the feature information acquisition unit 113 identifies an area corresponding to the selected room from the segmented floor plan information. For example, the feature information acquisition unit 113 identifies the positions of all pixels labeled with the selected room. Then, the feature information acquisition unit 113 acquires a feature amount corresponding to the selected feature item based on the identified area (step S105). Next, the feature information acquisition unit 113 determines whether or not there is any feature item that has not yet been selected among a plurality of predetermined feature items (step S106). If there is an unselected feature item (step S106: YES), the process proceeds to step S104, where the feature information acquisition unit 113 selects one of the unselected feature items. On the other hand, if there is no unselected feature item (step S106: NO), the feature information acquisition unit 113 determines whether there is an unselected room among the rooms in the property indicated by the segmented floor plan information (step S107). If there is an unselected room (step S107: YES), the process proceeds to step S103, where the feature information acquisition unit 113 selects one of the unselected rooms. On the other hand, if there is no unselected room (step S107: NO), the feature information acquisition unit 113 generates a feature vector (step S108). For example, the feature information acquisition unit 113 generates a one-dimensional array consisting of feature amounts of a predetermined number of feature items for each room. At this time, the feature information acquisition unit 113 associates the room number and room type of the corresponding room with each one-dimensional array. Then, the feature information acquisition unit 113 arranges the one-dimensional array to generate a feature vector. Next, the feature information acquisition unit 113 associates the generated segmented floor plan information and segmented feature vector with property information and stores them in the property DB 14b (step S109), and the floor plan analysis process ends.

Figure 14 is a flowchart showing an example of a property recommendation process by the system control unit 11 of the information providing server 1 according to this embodiment. For example, the target user launches an application dedicated to the property information providing system S. In response, the user terminal 2 transmits a login request including, for example, a user ID and password stored in the user terminal 2 to the information providing server 1. If it is determined that the login is successful based on the login request, the system control unit 11 determines whether or not the current login by the target user is the first login for the day. If the current login is the first login, the system control unit 11 may execute a property recommendation process.

As shown in FIG. 14, the determining unit 115 determines whether the parameters of the property classifier 500 are stored in the model DB 14d in association with the user ID of the target user (step S201). If the parameter is not stored (step S201: NO), the determining unit 115 identifies another user who has the same attributes as the target user (step S202). For example, the determining unit 115 may acquire at least one attribute of the target user's attributes stored in the member DB 14a as a search condition. The determining unit 115 may then search the member DB 14a for users who have attributes that match the search conditions. Alternatively, the determining unit 115 may request a server device of a system different from the property information providing system S for a list of user IDs of users having the same attributes as the target user. At this time, the determining unit 115 may specify an attribute item as a search condition. Next, the determining unit 115 determines whether the parameters of the property classifier 500 are stored in the model DB 14d in association with the user ID of the other specified user (step S203). If the parameters of the target user's property classifier 500 or the parameters of the property classifier 500 of another user are stored (step S201: YES or step S203: YES), the determining unit 115 uses the parameters as a model. It is acquired from the DB 14d (step S204). After step S204, or if the parameters of the property classifier 500 of another user are not stored (step S203: NO), the determining unit 115 sets the number of properties i to 0 (step S205). Next, the determining unit 115 executes property selection processing (step S206).

Figure 15 is a flowchart showing an example of a property selection process by the system control unit 11 of the information providing server 1 according to this embodiment. As shown in Figure 15, the determination unit 115 determines whether or not parameters of the property classifier 500 have been acquired in the property recommendation process (step S301). That is, the determination unit 115 determines whether or not step S204 has been executed. If parameters have not been acquired (step S301: NO), the determination unit 115 selects any property whose property information is stored in the property DB 14b as a display property (step S302), and the property selection process ends. For example, the determination unit 115 acquires a property ID of the display property. At this time, the determination unit 115 refers to the evaluation information DB 14c and excludes properties for which the target user has already input an evaluation from the display properties.

On the other hand, if the parameter is acquired (step S301: YES), the determining unit 115 determines whether the acquired parameter is a parameter of the object classifier 500 of the target user (step S303). If the acquired parameter is a parameter of the property classifier 500 of the target user (step S303: YES), the determining unit 115 determines that the acquired parameter is a parameter of the property classifier 500 of the target user (step S303: YES), the determination unit 115 determines that the target user has received a predetermined number of days or more from the day the floor plan image was displayed to the target user. It is determined whether there is any property that has passed (step S304). For example, the determining unit 115 extracts evaluation information stored in association with the user ID of the target user from the evaluation information DB 14c. The determining unit 115 searches the extracted evaluation information for evaluation information in which the number of days that have elapsed from the display date of the floor plan image to today is a predetermined number of days or more. Here, the determining unit 115 may search only for evaluation information whose overall evaluation is set to "like". If the corresponding evaluation information is retrieved, the determining unit 115 determines that there is a property for which a predetermined number of days or more have passed since the day the floor plan image was displayed (step S304: YES). In this case, the determining unit 115 generates pseudo-random numbers within the range of, for example, 0 to 100 (step S305). Next, the determining unit 115 determines whether the generated pseudo-random number is less than a predetermined probability (step S306). This probability may be set in advance based on the percentage of properties presented to the user that include properties for which the user has input evaluations in the past. If the pseudo-random number is less than the predetermined probability (step S306: YES), the determining unit 115 selects any of the properties for which a predetermined number of days or more have passed since the day the floor plan image was displayed as the display property. (Step S307), the property selection process ends. For example, the determining unit 115 acquires the property ID of the displayed property from any of the searched evaluation information.

If the acquired parameter is not a parameter of the target user's property classifier 500 (step S303: NO), if there is no property for which a predetermined number of days or more have passed since the day the floor plan image was displayed (step S304) : NO), or if the pseudo-random number is greater than or equal to the predetermined probability (Step S306: NO), the determining unit 115 selects any property whose property information is stored in the property DB 14b as a candidate property (Step S308). . At this time, the determining unit 115 refers to the evaluation information DB 14c and excludes properties for which the target user has already input an evaluation from the candidate properties.

Next, the feature information acquisition unit 113 obtains the feature vector of the candidate property from the property DB 14b. The determining unit 115 converts each feature included in the acquired feature vector into a favorite item included in a model 530 for a feature item corresponding to the feature in a model 510 for the type of room corresponding to the feature. They are input into the model 540 and the dislike model 550 to obtain a like score and a dislike score (step S309). For example, the determining unit 115 uses the parameters of the liking model 540 from among the acquired parameters to calculate the probability density of the feature amount as the liking score. Further, the determining unit 115 uses the parameters of the disliking model 550 to calculate the probability density of the feature amount as a disliking score.

Next, the determination unit 115 excludes combinations of room types and feature items for which both the like score and the dislike score are less than the first predetermined value from combinations for which feature item scores are calculated (step S310). Next, the determination unit 115 calculates feature item scores by subtracting the dislike score from the like score for each combination of room types and feature items that was not excluded in step S310 (step S311). Next, the determination unit 115 excludes combinations for which the absolute value of the feature item score is less than the second predetermined value from combinations for which the feature item score is used in the calculation of the evaluation score (step S312). Next, the determination unit 115 calculates the evaluation score by adding up the feature item scores of the combinations of room types and feature items that were not excluded in step S312 (step S313).

Next, the determining unit 115 determines whether the evaluation score is greater than or equal to the threshold (step S314). If the evaluation score is less than the threshold (step S314: NO), the process advances to step S303. On the other hand, if the evaluation score is equal to or greater than the threshold (step S313: YES), the determining unit 115 selects the candidate property as the display property (step S315), and the property selection process ends.

As shown in FIG. 14, when the property selection process ends, the display control unit 111 acquires property information, floor plan information, and segmented floor plan information of the displayed property from the property DB 14b. Then, the display control unit 111 transmits the acquired information to the user terminal 2 of the target user (step S207). Next, the evaluation information acquisition unit 112 receives evaluation information transmitted from the user terminal 2 of the target user (step S208). Next, the evaluation information acquisition unit 112 adds the user ID of the target user, the property ID of the displayed property, and the display date indicating today's date to the received evaluation information. Then, the evaluation information acquisition unit 112 stores the evaluation information in the evaluation information DB 14c (step S209). If the evaluation information of the displayed property is already stored in the evaluation information DB 14c, the evaluation information acquisition unit 112 may rewrite the evaluation information with new evaluation information.

Next, the display control unit 111 determines whether the number i of properties is less than the predetermined upper limit number N of properties (step S210). If the number of properties i is less than the upper limit number N (step S2120: YES), the display control unit 111 increases the number of properties i by 1 (step S211), and the process proceeds to step S206. On the other hand, if the number of properties i is equal to or greater than the upper limit number N (step S2120: NO), the property recommendation process ends.

FIG. 16 is a flowchart showing an example of floor plan display processing by the CPU of the user terminal 2 according to the present embodiment. For example, after a target user starts an application dedicated to the property information providing system S, the information providing server 1 transmits property information, floor plan information, and segmented floor plan information to the user terminal 2 of the user. When the user terminal 2 receives this information from the information providing server 1, it may execute the floor plan display process.

As shown in FIG. 16, the user terminal 2 displays the received floor plan image on the display (step S401). Next, the user terminal 2 acquires the coordinates of the pixel that is touched on the screen at a predetermined time interval, for example, from a touch panel provided in the user terminal 2. The user terminal 2 determines the operation by the user based on the acquired coordinate information.

Then, the user terminal 2 determines whether the user operation is a charging flick to the right or left (step S402). If the user operation is a hold flick (step S402: YES), the user terminal 2 identifies the room specified by the user (step S403). For example, the user terminal 2 specifies the coordinates of the pixel that is first touched in the charging flick operation. The user terminal 2 obtains the label of the pixel at the specified coordinates from the segmented floor plan information. The user terminal 2 identifies the room number and room type of the designated room based on the acquired label. Next, the user terminal 2 determines whether the direction of the charging flick is to the right (step S404). If the direction of the charging flick is to the right (step S404: YES), the user terminal 2 generates partial evaluation information including the room number and room type of the specified room and the partial evaluation "like" (step S405). ), the process proceeds to step S402. On the other hand, if the direction of the charging flick is not to the right (step S404: NO), the user terminal 2 generates partial evaluation information including the room number and room type of the specified room and the partial evaluation "dislike" ( Step S406), the process proceeds to step S402.

If the user operation is not a hold flick (step S402: NO), the user terminal 2 determines whether the user operation is a normal flick to the right or left (step S407). If the user operation is not a normal flick (step S407: NO), the user terminal 2 determines whether the user operation is another valid operation (step S408). If the user operation is another operation (step S408: YES), the user terminal 2 executes processing according to the operation (step S409), and the processing proceeds to step S401. For example, if the user operation is a swipe to the right or left, the user terminal 2 displays property information such as the property name, property image, and detailed information on the property details screen. Then, when the user swipes to the right or left again, the processing proceeds to step S401.

If the user operation is a normal flick (step S407: YES), the user terminal 2 determines whether the direction of the normal flick is to the right (step S410). If the normal flick direction is rightward (step S410: YES), the user terminal 2 sets the overall evaluation to "like" (step S411). On the other hand, if the direction of the normal flick is not rightward (step S410: NO), the user terminal 2 sets the overall evaluation to "dislike" (step S412). Next, the user terminal 2 displays a questionnaire screen (step S413). Next, the user terminal 2 obtains the questionnaire responses input by the user on the questionnaire screen (step S414).

After step S411 or step S414, the user terminal 2 generates evaluation information including the overall evaluation. If partial evaluation information has been generated in step S405 or S406, the user terminal 2 adds the partial evaluation information to the evaluation information. Furthermore, if a questionnaire response has been acquired in step S414, the user terminal 2 adds the questionnaire response to the evaluation information. Then, the user terminal 2 transmits the generated evaluation information to the information providing server 1 (step S415), and the floor plan display process ends.

FIG. 17 is a flowchart showing an example of a learning process by the system control unit 11 of the information providing server 1 according to this embodiment. For example, after the property recommendation process is completed, the system control unit 11 may execute a learning process for the target user to whom the property was presented.

As shown in FIG. 17, the generation unit 114 acquires all evaluation information associated with the user ID of the target user from the evaluation information DB 14c (step S501). Next, the feature information acquisition unit 113 acquires feature vectors associated with the property IDs included in each piece of acquired evaluation information from the property DB 14b (step S502). Next, the generation unit 114 selects one of a plurality of predetermined room types (step S503). Next, the generation unit 114 selects one of a plurality of predetermined feature items (step S504).

Next, the generation unit 114 acquires partial evaluation information in which the selected room type and the partial evaluation "like" are set from the evaluation information acquired in step S501. Then, for each piece of acquired partial evaluation information, the generation unit 114 acquires a feature value corresponding to the combination of the room number included in the partial evaluation information and the selected feature item from the feature vector of the property corresponding to that partial evaluation information (step S505). Next, the generation unit 114 acquires evaluation information in which the overall evaluation "like" is set from the evaluation information acquired in step S501. The feature information acquisition unit 113 identifies a room of the selected room type for the property corresponding to each piece of evaluation information acquired here. The feature information acquisition unit 113 acquires a feature value corresponding to the combination of the room number and the selected feature item for the property that does not have partial evaluation information for that room among the identified rooms (step S506). Next, the generation unit 114 uses the feature values acquired in steps S505 and S506 to determine the parameters of the like model 540 (step S507). For example, the generation unit 114 may use maximum likelihood estimation to calculate the mean, variance, and weight of each normal distribution that constitutes a mixed normal distribution that indicates the probability distribution of the acquired feature quantities.

Next, the generation unit 114 acquires partial evaluation information in which the selected room type and the partial evaluation "dislike" are set from the evaluation information acquired in step S501. Then, for each piece of acquired partial evaluation information, the generation unit 114 acquires a feature value corresponding to the combination of the room number included in the partial evaluation information and the selected feature item from the feature vector of the property corresponding to that partial evaluation information (step S508). Next, the generation unit 114 acquires evaluation information in which the overall evaluation "dislike" is set from the evaluation information acquired in step S501. The feature information acquisition unit 113 identifies a room of the selected room type for the property corresponding to each piece of evaluation information acquired here. The feature information acquisition unit 113 acquires a feature value corresponding to the combination of the room number and the selected feature item for the property that does not have partial evaluation information for that room among the identified rooms (step S509). Next, the generation unit 114 uses the feature values acquired in steps S508 and S509 to determine the parameters of the dislike model 550 in the same manner as in step S507 (step S510).

Next, the generation unit 114 determines whether or not there is any feature item that has not yet been selected among the multiple predetermined feature items (step S511). If there is any feature item that has not yet been selected (step S511: YES), the process proceeds to step S504, and the generation unit 114 selects one of the feature items that have not yet been selected. On the other hand, if there is no feature item that has not yet been selected (step S511: NO), the feature information acquisition unit 113 determines whether or not there is any room type that has not yet been selected among the multiple predetermined room types (step S512). If there is any room type that has not yet been selected (step S512: YES), the process proceeds to step S503, and the generation unit 114 selects one of the room types that have not yet been selected. On the other hand, if there is no room type that has not yet been selected (step S512: NO), the generation unit 114 associates the determined parameters with the user ID of the target user and stores them in the model DB 14d (step S513), and the learning process ends. At this time, the generation unit 114 stores parameters for each combination of room type and feature item in association with that combination.

As described above, according to this embodiment, the information providing server 1 displays floor plan images for each of the multiple properties. The information providing server 1 also acquires evaluation information indicating user evaluations of the displayed floor plan images for each of the multiple properties. The information providing server 1 also acquires feature information for each of the multiple properties, the feature information being generated based on the floor plan images and indicating the features of the rooms in the property. The information providing server 1 also generates a property classifier based on the acquired evaluation information and the acquired feature information. The information providing server 1 also determines the properties to be presented to the user based on the generated property classifier. Therefore, information on properties preferred by the user can be provided even if the user does not input the conditions of the property as understood from the floor plan.

Here, the information providing server 1 may display the floor plan image on the screen of a touch panel that has a display having a screen and a sensor that detects a touch position on the screen. Furthermore, when movement of the contact position in a predetermined direction is detected on the screen on which the floor plan image is displayed, the information providing server 1 may acquire evaluation information indicating a predetermined evaluation. In this case, the evaluation can be input with a simple operation.

In addition, the information providing server 1 may generate a property classifier for each user. In this case, personalized property information can be provided.

Further, the information providing server 1 may determine the property to be presented to the target user based on a property classifier generated for another user having the same attributes as the target user. In this case, even if a property classifier with appropriate accuracy has not been generated for the user to whom the property is presented, it is possible to appropriately determine the property to be presented to the user.

Further, the information providing server 1 may further acquire partial evaluation information indicating the user's evaluation of a portion specified by the user in the displayed floor plan image for at least one of the plurality of properties. . In addition, the information providing server 1 uses the acquired partial evaluation information for a room corresponding to a specified part of the rooms in the property, and uses the obtained partial evaluation information to A property classifier may be generated using the overall evaluation of the acquired evaluation information for a room different from the room. In this case, even if a specific rating can be given to a property as a whole, if the user wants to give a different rating to some rooms in the property, each rating can be classified into the property classification. This can be reflected in the creation of vessels.

The information providing server 1 may also redisplay the floor plan image for at least one of the multiple properties for which a floor plan image has been displayed a predetermined number of days after the floor plan image was displayed. The information providing server 1 may further acquire value information indicating a reevaluation from the user of the redisplayed floor plan image. The information providing server 1 may also generate a property classifier using the reevaluation information acquired later for a property for which a floor plan image has been redisplayed among the multiple properties, and using the acquired evaluation information for a property other than the reevaluation information. In this case, a property classifier can be generated that reflects the user's preferences, which change over time or as the user views information on various properties.

Further, the information providing server 1 may specify an area corresponding to a room from the floor plan image. Further, the information providing server 1 may specify the characteristics of the room based on the specified area. Here, the information providing server 1 may specify the characteristics of the room based on the characteristics of the specified area. In this case, evaluations of visual features obtained from floor plan images of rooms within the property can be reflected in the property classifier.

The information providing server 1 also provides characteristic information indicating at least one of the area of the room, the length of the side of a figure corresponding to the range occupied by the room in the property, the interior angle of the figure, and the number of vertices of the figure. You may obtain it. In this case, it is possible to provide property information according to the user's preferences regarding room characteristics that can be specified from the floor plan.

In addition, the information providing server 1 has a certain feature in which the proportion of the characteristic to which a rating of "like" is given is greater than the proportion of the feature to which a rating of "dislike" is given. A property classifier may be generated that identifies a property that the user prefers from among properties that have rooms. In this case, a property that has a room that has at least features that the user prefers can be identified as a property that the user prefers.

In addition, the information providing server 1 stores properties that the user prefers in a plurality of probability distributions corresponding to a plurality of evaluations, and each of the characteristic information from which evaluation information indicating the evaluation corresponding to the probability distribution is acquired is set as a probability variable. A property classifier may be generated that identifies properties based on a plurality of probability distributions. In this case, a property classifier is generated that identifies properties preferred by the user based on the occupancy rate or appearance probability of the feature to which the evaluation is given in each of the plurality of evaluations.

Further, the information providing server 1 identifies the type of room from the floor plan image for each room included in the property, and indicates, for each of the plurality of room models, the characteristics of the type of room corresponding to the room model. A room model may be generated based on the feature information and evaluation information. In this case, the user's preferences can be reflected in the property classifier for each room type.

1 Information providing server 2 User terminal 11 System control unit 12 System bus 13 Input/output interface 14 Storage unit 14a Member DB
14b Property DB
14c Evaluation information DB
14d Model DB
15 Communication unit 111 Display control unit 112 Evaluation information acquisition unit 113 Characteristic information acquisition unit 114 Generation unit 115 Determination unit NW Network S Property information providing system

Claims (7)

Characteristic information generated based on a floor plan image of a property and indicating the characteristics of a room in the property is stored in a plurality of storage means in association with property identification information for identifying the property. Display control means for displaying the floor plan image on the screen of a display device that has a display unit having a screen and a detection unit that detects a contact position on the screen for each of the properties;
an evaluation information acquisition means for acquiring evaluation information indicating an evaluation of the displayed floor plan image for each of the plurality of properties, the evaluation information acquisition means for acquiring evaluation information indicating an evaluation of the displayed floor plan image; If movement of the contact position in the direction is detected, the evaluation information indicating a positive evaluation of the displayed floor plan image is acquired, and the movement of the contact position in a predetermined second direction different from the first direction is obtained. evaluation information acquisition means for acquiring the evaluation information indicating a negative evaluation of the displayed floor plan image when a movement of the contact position is detected;
For each of the plurality of properties, evaluation information storage control causes the acquired evaluation information to be stored in the storage means in association with user identification information that identifies a user of the display device and the property identification information that identifies the property. means and
An information processing device comprising: The evaluation information acquisition means
When a movement of the contact position in the first direction in a predetermined first manner is detected, the evaluation information indicating the positive evaluation for the entire property for which the floor plan image is displayed on the screen is acquired;
When the movement of the contact position in the first manner in the second direction is detected, the evaluation information indicating the negative evaluation for the entire property for which the floor plan image is displayed on the screen is transmitted. Acquired,
When movement of the contact position in a predetermined second manner in a predetermined third direction and that is different from the movement in the first manner is detected, the displayed floor plan image obtaining the evaluation information indicating the positive evaluation for the room at the contact position among at least one room indicated by;
When the movement of the second aspect of the contact position in a predetermined fourth direction different from the third direction is detected, the evaluation information indicating the negative evaluation of a room at the contact position among at least one room shown by the displayed floor plan image is acquired;
The information processing device according to claim 1, characterized in that when the evaluation information indicating the evaluation for the room at the contact position is acquired by the evaluation information acquisition means, the evaluation information storage control means associates the acquired evaluation information with the user identification information, the property identification information, and room identification information that identifies the room at the contact position and stores it in the storage means. The storage means stores, for each of at least one room in the property, the characteristic information indicating the characteristics of the room in association with the property identification information and room identification information identifying the room. The information processing device according to claim 1 or 2. positive evaluation information acquisition means for acquiring the evaluation information indicating the positive evaluation from among the evaluation information stored in the storage means;
a negative evaluation information acquiring means for acquiring the evaluation information indicating the negative evaluation from among the evaluation information stored in the storage means;
4. The information processing apparatus according to claim 1, further comprising: The acquired evaluation information indicates the positive evaluation based on the evaluation information acquired by the positive evaluation information acquisition means and the characteristic information associated with the property identification information associated with the evaluation information. a positive evaluation model generating means for generating a positive evaluation model showing a probability distribution using the characteristic information about the rooms in the property as a random variable;
a negative evaluation model generating means for generating a negative evaluation model showing a probability distribution in which the characteristic information for a room in a property for which the acquired evaluation information indicates the negative evaluation is a random variable, based on the evaluation information acquired by the negative evaluation information acquiring means and the characteristic information associated with the property identification information associated with the evaluation information;
The information processing device according to claim 4, further comprising: In an information processing method performed by a computer,
a display control step of displaying the floor plan image on a screen of a display device having a display unit with a screen and a detection unit for detecting a contact position on the screen, for each of a plurality of properties stored in a storage means in association with property identification information that identifies the property, the feature information being generated based on a floor plan image of the property, the feature information indicating features of the rooms in the property;
an evaluation information acquisition step of acquiring evaluation information indicating an evaluation of the displayed floor plan image for each of the plurality of properties, in which, when movement of the contact position in a predetermined first direction is detected on the screen on which the floor plan image is displayed, the evaluation information indicating a positive evaluation of the displayed floor plan image is acquired, and, when movement of the contact position in a predetermined second direction different from the first direction is detected, the evaluation information indicating a negative evaluation of the displayed floor plan image is acquired;
For each of the plurality of properties, evaluation information storage control causes the acquired evaluation information to be stored in the storage means in association with user identification information that identifies a user of the display device and the property identification information that identifies the property. step and
An information processing method characterized by comprising: computer,
a display control means for displaying the floor plan image on a screen of a display device having a display unit with a screen and a detection unit for detecting a contact position on the screen, the display unit having feature information generated based on a floor plan image of a property, the feature information indicating features of a room within the property being stored in a storage means in association with property identification information that identifies the property;
an evaluation information acquisition means for acquiring evaluation information indicating an evaluation of the displayed floor plan image for each of the plurality of properties, the evaluation information acquisition means for acquiring evaluation information indicating an evaluation of the displayed floor plan image; If movement of the contact position in the direction is detected, the evaluation information indicating a positive evaluation of the displayed floor plan image is acquired, and the movement of the contact position in a predetermined second direction different from the first direction is obtained. evaluation information acquisition means for acquiring the evaluation information indicating a negative evaluation of the displayed floor plan image when a movement of the contact position is detected;
an evaluation information storage control means for storing, in the storage means, the acquired evaluation information for each of the plurality of properties in association with user identification information for identifying a user of the display device and the property identification information for identifying the property;
An information processing program characterized by causing the program to function as follows.

Images