JP7083369B2 - Attribute ratio estimation system - Google Patents

Description

The present invention relates to a technique for analyzing the movement information of a person, and more specifically, to an attribute ratio estimation system capable of estimating the attributes of a large number of people who have stayed in a predetermined place for a predetermined period. be.

In recent years, big data has attracted a great deal of attention in the market. The Ministry of Internal Affairs and Communications has also positioned 2017 as the "first year of big data utilization," and said, "It will lead to the realization of solutions for solving new problems that could not be imagined in the past, and in the realization of those solutions, in different areas. "Realization of innovation in collaboration with players", and we have high expectations for the utilization of big data.

Big data is not only a large amount (Volume), but also various types (Variity), data generation is extremely fast (Velocity), and economic value (Value) is generated by analysis and utilization. , And is said to have so-called "4 Vs". Therefore, unlike conventional databases, it is not easy to handle, and various efforts are being made to realize useful analysis and utilization.

In addition, the Ministry of Internal Affairs and Communications has four types of big data: "open data provided by the national and local governments", "data that digitizes and structures the implicit knowledge (knowledge) possessed by companies", and "M2M by companies". It is classified into "streaming data discharged from Machine to Machine" and "personal data related to individual attributes". Of these, regarding personal data, "personal information" defined by law after establishing a system of "anonymously processed information" in the "Revised Personal Information Protection Law" in order to improve the environment that contributes to the proper utilization of big data. In addition, personal data includes personal flow information, product information, etc. that have been processed so that a specific individual cannot be identified.

On the other hand, recently, access points (hereinafter, simply referred to as "wireless access points") that provide a public wireless LAN (Local Area Network) have been placed in various places in the city. When this wireless access point detects a terminal device (for example, a smartphone) carried by a person staying in the vicinity, it stores the MAC (Media Access Control) address of the terminal device together with the detected time. In other words, it is possible to obtain the approximate number of people who stayed around the wireless access point during a predetermined time period, and it is possible to obtain information such as fluctuations in the number of residents during the day and night and the ratio of the number of residents on weekdays and holidays. As a result, the behavior history of these residents can also be estimated.

However, for marketing conducted by companies and businesses, information on the number of residents is not so useful, and those who are interested in companies (that is, those who are likely to be customers), in other words, specific ones. Information on the person who has the attribute (action history, etc.) is required. However, since it is difficult to obtain "personal information" defined by law, it is conceivable to utilize "anonymously processed information" defined by the Ministry of Internal Affairs and Communications. Naturally, in order to utilize anonymously processed information, it is necessary to acquire a certain amount of information from the user, and various acquisition methods have been proposed so far. For example, Patent Document 1 proposes a technique for registering user attribute information (such as the date of birth and gender of a user) to the extent that an individual is not specified when using an application downloaded to a terminal device.

Japanese Unexamined Patent Publication No. 2019-036895

In Patent Document 1, by downloading a predetermined application to a terminal device in advance, a peripheral wireless access point automatically authenticates the user, that is, the user connects to a public wireless LAN without any special operation. It is a technology that can be done. That is, the technique disclosed in Patent Document 1 is not a technique for effectively analyzing and utilizing the registered attribute information, although the user registers predetermined attribute information in order to download the application.

An object of the present invention is to solve a conventional problem, that is, to provide an attribute ratio estimation system capable of effectively analyzing and utilizing attribute information input by a user.

The present invention focuses on the point that an attribute is given to information acquired by a wireless access point (information having no attribute) by using attribute information input by a specific user as a clue, which has not been conventionally performed. It is an invention made based on the idea.

The attribute ratio estimation system of the present invention includes a first storage means, a user terminal device carried by the user, and an attribute ratio calculation means. Of these, the first storage means is a means for storing user information including user attributes, and the attribute ratio calculation means is a means for calculating a "user attribute ratio". Here, the user attribute ratio is the ratio of attributes related to a plurality of users who have been around the access point within a predetermined period. When the user terminal device connects to the access point, the user terminal device transmits "user connection information" including access point identification information (information identifying the access point) and the connection time to the first storage means, and the first storage means is the user. Store connection information and user information in association with each other. Then, the attribute ratio calculating means reads the user connection information within a predetermined period from the first storage means, and calculates the user attribute ratio based on the user information related to the user connection information.

The attribute ratio estimation system of the present invention may further include a second storage means and a means for estimating the number of people staying in the spotlight. This second storage means is a means for storing "detection information" including the terminal device detected by the access point and the detection time, and the means for estimating the number of notable residents is a means for estimating the number of "attention residents". be. Here, the notable resident is a resident who has a specific attribute among the resident who was in the vicinity of the access point within a predetermined period. Further, the means for estimating the number of people of interest read the detection information within a predetermined period from the second storage means, and estimate the number of people of interest based on the number of the detected information and the user attribute ratio.

The attribute ratio estimation system of the present invention may further include an attribute input means for the operator to input a desired attribute. In this case, the attention resident number estimation means estimates the number of attention resident having the attribute input by the attribute input means.

The attribute ratio estimation system of the present invention may further include a period setting means for setting a predetermined period by the operator. In this case, the attribute ratio calculating means calculates the user attribute ratio for each predetermined period set by the period setting means, and the attention resident number estimation means estimates the number of attention resident for each predetermined period.

In the attribute ratio estimation system of the present invention, the user terminal device may be equipped with a "connected application". This connection application connects the user terminal device to the detected access point, and when the user terminal device is connected to the access point, the user connection information is transmitted to the first storage means.

The attribute ratio estimation system of the present invention has the following effects.
(1) Only by registering the attribute information of the user to the extent that it does not identify an individual, that is, without handling the personal information defined by law, the person of which attribute stayed in the predetermined place for a predetermined period and for how long. Can be grasped.
(2) Since the position of the wireless access point is known and the time when the wireless access point is detected is also known, it is necessary to grasp how many people with a specific attribute stayed at a predetermined place for a predetermined period. Can be done. This makes it possible to estimate how much the person with the attribute desired by the company or the like fluctuates during the day and night, how much it differs between weekdays and holidays, and the behavior history of these residents.
(3) By estimating the attributes of residents, the distribution and retail industry and the real estate industry can be used for predicting customer behavior, and the transportation industry can be used for congestion mitigation measures such as MaaS (Mobility as a Service) and autonomous driving services. can do.

The block diagram which shows the main structure of the attribute ratio estimation system of this invention. The flow diagram which shows the main processing of a user terminal apparatus. A model diagram schematically showing user connection information stored by the first storage means. The model figure which shows typically the 2nd storage means which stores the detection information of a plurality of user terminal devices and general terminal devices detected by a wireless access point. A model diagram schematically showing the detection information stored by the second storage means. A model diagram schematically showing 4 users and 12 general residents staying around the wireless access point AP.

An example of the attribute ratio estimation system of the present invention will be described with reference to the figure.

FIG. 1 is a block diagram showing a main configuration of the attribute ratio estimation system 100 of the present invention. As shown in this figure, the attribute ratio estimation system 100 of the present invention is configured to include the user terminal device 101, the attribute ratio calculation means 102, and the first storage means 106, and further includes the attention resident number estimation means 103 and the attribute input means. It can also be configured to include 104, a period setting means 105, a second storage means 107, and the like.

The user terminal device 101 constituting the attribute ratio estimation system 100 can use a portable terminal device such as a smartphone or a tablet computer (iPad (registered trademark) or the like). Further, the attribute ratio calculating means 102, the number of notable resident estimation means 103, the attribute input means 104, and the period setting means 105 can be manufactured as dedicated ones, or a general-purpose computer device can be used. This computer device is equipped with a processor such as a CPU and a memory such as a ROM or RAM, and may also include an input means such as a mouse or a keyboard and a display, and may be configured by, for example, a personal computer (PC) or a server. Can be done.

The first storage means 106 and the second storage means 107 can use the storage device of the computer device or can be constructed on the database server. In particular, the first storage means 106 can also use the storage device of the user terminal device 101. When constructing on a database server, it can be placed on a local network (LAN: Local Area Network), or it can be a cloud server that saves via the Internet (that is, wireless communication).

Hereinafter, each of the main means constituting the attribute ratio estimation system 100 will be described in detail.

(User terminal device)
As described above, the user terminal device 101 is a portable terminal device such as a smartphone, and when connected to the wireless access point AP, information related to the connection (hereinafter referred to as "user connection information") is transmitted to the first storage means 106. It has a function. For convenience, the portable terminal device having such a function is referred to as a "user terminal device 101", and the general portable terminal device not having the function is referred to as a "general terminal device TD". ". Further, the person who carries the user terminal device 101 is particularly referred to as a "user".

FIG. 2 is a flow chart showing the main processing of the user terminal device 101. In this figure, the processing of the user terminal device 101 is shown on the left side, and the processing of the first storage means 106 is shown on the right side. As shown in FIG. 2, the user downloads a predetermined application (Step 11 in FIG. 2) or implements a predetermined program to make his / her own terminal device a user terminal device 101.

Then, by operating a predetermined application or program (hereinafter referred to as "connected application"), information including user attributes such as date of birth and gender (hereinafter referred to as "user information") is registered (Fig.). Step 12 of 2). User information can include various attributes such as date of birth, gender, occupation, family structure, etc., but it is best to keep the information to the extent that it does not fall under the "personal information" defined by law. desirable.

The user information registered by the user is stored in the first storage means 106 (Step 13 in FIG. 2). At this time, the first storage means 106 stores (associates) the user information with the information identifying the user (hereinafter referred to as "user ID (IDentification)"). The process up to this point is the process to be performed in advance, and the process after that is the process in the operation stage.

When a user carrying the user terminal device 101 visits the vicinity of a wireless access point AP such as Wi-Fi (registered trademark), the user terminal device 101 detects the wireless access point AP. Then, the connection application connects the user terminal device 101 to the detected wireless access point AP (Step 14 in FIG. 2). At this time, the specification may be such that the user manually performs an authentication operation to connect to the wireless access point AP, or the user may automatically connect to the wireless access point AP by automatic authentication processing. Further, when a plurality of wireless access point APs are detected, the specifications can be set so that the user selects a desired wireless access point AP, or the specifications are such that the wireless access point AP is connected to the selected wireless access point AP under predetermined conditions. You can also do it.

When the user terminal device 101 is connected to the wireless access point AP, the connection application transmits "user connection information" to the first storage means 106 (Step 15 in FIG. 2).
The user connection information includes at least the connection time and information for identifying the wireless access point AP (hereinafter, referred to as "access point identification information"). Then, the first storage means 106 that has received the user connection information stores (associates) the user connection information with the user ID (or user information) (Step 16 in FIG. 2).

FIG. 3 is a model diagram schematically showing the user connection information stored by the first storage means 106. In this figure, user connection information about six users whose user IDs are 215, 374, 495, 678, 784, and 835, respectively, is stored. In this example, the wireless access point AP is specified by SSID (Service Set Identity) and BSSID (Basic Service Set Identifier), that is, the combination of SSID and BSSID is used as the access point identification information, but the SSID is not limited to this. Can be used as access point identification information, and various other information can be used as access point identification information.

For example, in FIG. 3, a user whose user ID is 215 at 10:00 on April 1 (hereinafter, simply referred to as “user ID 215” or the like) has an SSID of AAA and a BSSID of 111 (hereinafter, simply referred to as “AAA111” or the like). It can be seen that the wireless access point AP (shown) is being visited. It can also be seen that at the same time, the user ID 374, the user ID 678, and the user ID 784 are visiting the wireless access point AP of the AAA111.

(Attribute ratio calculation means)
The attribute ratio calculation means 102 calculates the "user attribute ratio" based on the user ID (or user information) read from the first storage means 106 and the user connection information. Here, the user attribute ratio is the ratio of the attributes of a plurality of users who have been around the predetermined wireless access point AP within a predetermined period.

For example, in the case of FIG. 3, four users (user ID 215 and user ID 374, user ID 678, user ID 784) are visiting the wireless access point AP of AAA111 during the period from 9:50 to 10:10 on April 1. .. If the user ID 215, the user ID 374, and the user ID 678 are "female" and the user ID 784 is "male", the attribute ratio calculation means 102 calculates the female ratio (user attribute ratio) as 75%. Alternatively, if ID 215 is "teens", user ID 374 and user ID 678 are "20s", and user ID 784 is "30s", the attribute ratio calculation means 102 sets the ratio of 20s (user attribute ratio) to 50%. Calculated as. As described above, since the first storage means 106 stores (associates) the user ID (or user information) with the user connection information, the attribute ratio calculation means 102 calculates the user attribute ratio as described above. You can do it.

Further, the attribute ratio calculating means 102 can also calculate a predetermined user attribute ratio (for example, a ratio of women or a ratio in their twenties), or calculate a user attribute ratio input by the attribute input means 104. You can also. Further, the attribute ratio calculating means 102 can repeatedly calculate the user attribute ratio every predetermined unit period (for example, 20 minutes or one week), or repeats every unit period set by the period setting means 105. It is also possible to calculate the user attribute ratio.

(Second memory means)
The second storage means 107 stores "detection information". Here, the detection information includes information for identifying the terminal device detected by the wireless access point AP (hereinafter, referred to as “terminal device ID”) and the detection time thereof. The second storage means 107 stores the detection information about all the terminal devices detected by the wireless access point AP, that is, the "user terminal device 101" and the "general terminal device TD" detected by the wireless access point AP. Store the detection information. For example, in FIG. 4, the wireless access point AP of AAA111 detects four user terminal devices 101 (users (user ID 215 and user ID 374, user ID 678, user ID 784) and two general terminal devices TD). The second storage means 107 stores these detection information.

FIG. 5 is a model diagram schematically showing the detection information stored by the second storage means 106. In this figure, it can be seen that 20 user IDs are detected by the wireless access point AP of AAA111 around 10 o'clock on April 1. In this way, the wireless access point AP acquires the detection time (detection information) together with the detected terminal device ID (for example, MAC address), and the second storage means 107 receiving the detection information is a terminal as shown in FIG. The device ID and its detection time are associated (linked) and stored.

(Means for estimating the number of notable residents)
The attention resident number estimation means 103 calculates the "attention resident number" based on the user attribute ratio calculated by the attribute ratio calculation means 102 and the detection information read from the second storage means 107. Here, the number of notable residents is a resident having a specific attribute among a plurality of "residents" who were in the vicinity of a predetermined wireless access point AP within a predetermined period (hereinafter referred to as "attention resident"). The number of people.

For example, in the case of FIG. 6, 16 people are staying around the wireless access point AP of AAA111, 4 of them are user UAs, and the remaining 12 people are people other than user UAs (hereinafter, "general resident UG"). ".). As shown in FIG. 5, these 16 people are staying around the wireless access point AP at 10:00 on April 1. If the ratio of females (user attribute ratio) of the four user UAs among the residents is 75%, the means of estimating the number of residents of interest 103 will be used during the period from 9:50 to 10:10 on April 1. The number of female residents (attention residents) (number of attention residents) is calculated as 12 (16 people × 75%). Alternatively, if the ratio of the four user UAs in their twenties (user attribute ratio) is 50%, the means of estimating the number of resident residents 103 is in their twenties during the period from 9:50 to 10:10 on April 1. The number of residents (attention residents) (number of attention residents) is calculated as 8 (16 people × 50%).

Further, the attention resident number estimation means 103 can also calculate the number of attention resident based on a predetermined user attribute ratio (for example, the ratio of women or the ratio of people in their twenties), as in the attribute ratio calculation means 102. However, it is also possible to calculate the number of people of interest based on the user attribute ratio input by the attribute input means 104. Further, the attention resident number estimation means 103 can repeatedly calculate the number of attention resident every predetermined unit period (for example, 20 minutes, one week, etc.) like the attribute ratio calculation means 102, and can set the period. It is also possible to repeatedly calculate the number of people of interest for each unit period set by the means 105.

The attribute ratio estimation system of the present invention can be used in various fields such as congestion mitigation measures, automatic driving services, and customer behavior prediction.

100 Attribute ratio estimation system 101 (attribute ratio estimation system) User terminal device 102 (attribute ratio estimation system) Attribute ratio calculation means 103 (attribute ratio estimation system) Attention resident number estimation means 104 (attribute ratio estimation system) Attribute input means (of the system) 105 (of the attribute ratio estimation system) Period setting means 106 (of the attribute ratio estimation system) First storage means 107 (of the attribute ratio estimation system) Second storage means AP wireless access point TD General terminal device UA User UG General resident

Claims (5)

Based on the number of "users" who carry a "user terminal device" equipped with a connected application, specific attributes among the "general resident" carrying a "general terminal device" not equipped with the connected application and the user. It is a system that estimates the number of people involved in
A first storage means for storing "user information" including the user's attributes,
An attribute ratio calculation means for calculating a "user attribute ratio" indicating the ratio of the number of people related to a specific attribute among the plurality of users who were around the access point within a predetermined period.
A second storage means for storing "detection information" including information for specifying the user terminal device detected by the access point, information for specifying the general terminal device, and detection time.
A means for estimating the number of notable resident who estimates the number of "attention resident" having a specific attribute among the user and the general resident who were around the access point within the predetermined period is provided.
When the user terminal device is connected to the access point, the connection application transmits "user connection information" including an access point identification information for identifying the access point and a connection time to the first storage means.
The first storage means stores the user connection information in association with the user information.
The attribute ratio calculating means reads the user connection information within the predetermined period from the first storage means, and calculates the user attribute ratio based on the user information related to the user connection information.
The means for estimating the number of notable residents reads the detection information within the predetermined period from the second storage means, and of the total number of the users and the general residents related to the detection information and the user attribute ratio. Estimate the number of notable residents based on the product,
An attribute ratio estimation system characterized by this. Further equipped with an attribute input means for the operator to input a desired attribute,
The attribute ratio calculating means calculates the user attribute ratio of the user related to the attribute input by the attribute input means.
The attribute ratio estimation system according to claim 1. The attention resident number estimation means estimates the number of attention resident having the attribute input by the attribute input means.
2. The attribute ratio estimation system according to claim 2. Further provided with a period setting means for the operator to set the predetermined period,
The attribute ratio calculating means calculates the user attribute ratio for each predetermined period set by the period setting means.
The attribute ratio estimation system according to any one of claims 1 to 3, wherein the attribute ratio estimation system is characterized. The attention resident number estimation means estimates the number of attention resident for each predetermined period set by the period setting means.
The attribute ratio estimation system according to claim 4, wherein the attribute ratio estimation system is characterized in that.

Images