JP7260294B2 - Information processing device, information processing method and information processing program - Google Patents

Description

The present invention relates to an information processing device, an information processing method, and an information processing program.

Conventionally, with the product as the main key, multiple characteristic phrases exemplified by the product's attributes, specifications, position on the thesaurus, etc. are extracted from product reviews, and the characteristic phrases and one product are linked, A technology is known that executes a search process based on a characteristic phrase by specifying a product from the characteristic phrase (for example, Patent Literature 1).

Japanese Patent No. 6422544

The technique of Patent Literature 1 has a problem that it is impossible to execute a search process based on subjective expressions different from product attributes, specifications, positioning on the thesaurus, and the like.

SUMMARY OF THE INVENTION It is an object of the present invention to provide an information processing apparatus, an information processing method, and an information processing program that enable search processing based on subjective expressions.

In order to solve the above-described problems and achieve the object, an information processing apparatus includes: an input unit that receives input of search conditions including a search word; a search unit that specifies a subjective expression group to which the search word belongs based on the degree of similarity, and searches for sales objects included in the sales object group associated with the subjective expression group; and an output unit for outputting the sales target as a search result.

According to the present invention, it is possible to provide an information processing device, an information processing method, and an information processing program that enable search processing based on subjective expressions.

FIG. 1 is an explanatory diagram illustrating an example of an information processing apparatus according to an embodiment. FIG. 2 is a diagram showing the configuration of an information processing system including the information processing apparatus of FIG. FIG. 3 is a control block diagram of the information processing apparatus of FIG. 4 is a diagram showing an example of a display screen displayed by the input unit in FIG. 3. FIG. FIG. 5 is a diagram showing an example of a display screen generated by the output unit in FIG. 3 based on search results generated by the search unit. FIG. 6 is a diagram showing an example of the sale target database of FIG. FIG. 7 is a diagram showing an example of the subjective expression database of FIG. FIG. 8 is a diagram showing an example of the related information database of FIG. FIG. 9 is an explanatory diagram for explaining the degree of similarity, which is an index for integration of the subjective expression database of FIG. FIG. 10 is a flowchart of an information processing method according to the embodiment. FIG. 11 is a flow chart of a method for creating the subjective expression database and related information database of FIG. FIG. 12 is a hardware configuration diagram showing an example of a computer that implements the functions of the information processing apparatus.

EMBODIMENT OF THE INVENTION Below, embodiment which concerns on this invention is described in detail based on drawing. In addition, this invention is not limited by this embodiment. In addition, components in the embodiments include those that can be easily replaced by those skilled in the art, or those that are substantially the same. Furthermore, the components described below can be combined as appropriate.

[1. Example of information processing device 10]
FIG. 1 is an explanatory diagram illustrating an example of an information processing apparatus according to an embodiment. As shown in FIG. 1, the information processing apparatus 10, which is an example of the information processing apparatus according to the embodiment, searches for pre-registered sales in response to an input of a search condition 42 from a user terminal 40 used by a user 41. Among the targets 24 , sales targets 48 that match the search conditions 42 are output as search results 47 .

Here, the search conditions 42 include search words 43 used when searching for sales targets 48 . The search condition 42 may further include category designation information 45 that designates the category (genre) of the search range of the sale target 48 . Search word 43 includes subjective expression 44 . The subjective expression 44 included in the search word 43 is, like the subjective expression 32 described later, for example, a metaphorical expression that metaphorically expresses the sales target 48 or an expression using adjectives, and is based on the human sense of the user 41. It is due to the elements.

The pre-registered sales target 24 is a product or service that can be purchased by the user 41 through the use of the information processing apparatus 10 via the user terminal 40 . In the information processing device 10, the sales target 24 is included in the subjective expression 32 included in the sales target information 27 (see FIG. 6) about the sale target 24, and the subjective expression group 31 generated by integrating the subjective expression 32. Based on this, the sales objects 24 are grouped into a sales object group 35 and managed.

As shown in FIG. 1 , information processing apparatus 10 first receives an input of search word 43 and identifies subjective expression group 31 corresponding to subjective expression 44 included in search word 43 . The information processing apparatus 10, for example, based on the similarity 33 between the subjective expression 44 and the subjective expression 32 classified into the subjective expression group 31, selects the subjective expression group 31 corresponding to the subjective expression 44. Identify. Here, the degree of similarity 33 is a parameter that increases as the properties of expressions are closer and the properties of objects to be represented are closer.

Then, the information processing apparatus 10 creates a sales target group 35, which is a collection of sales targets 24 having sales target information 27 including any of the subjective expressions 32 associated with the identified subjective expression group 31. The sales object 48 included in the searched sales object group 35 is retrieved as the search result 47 . For example, the information processing apparatus 10 classifies a plurality of sales object groups 35 for each subjective expression group 31, and based on related information 34 that associates the subjective expression group 31 and a plurality of sales object groups 35, The sales target 48 included in the sales target group 35 associated with the identified subjective expression group 31 is retrieved as the search result 47 .

Further, as shown in FIG. 1, the information processing apparatus 10 accepts the input of the category designation information 45, and the search range based on the search word 43 is designated by the category designation information 45 among the sales targets 24 registered in advance. It may be limited to those to which the category information 26 corresponding to the selected category is added. The detailed configuration and action of the information processing apparatus 10 will be described later.

The user terminal 40 includes a mobile phone including a high-performance mobile phone (so-called smart phone), a tablet terminal, a laptop or desktop PC (Personal Computer), a PDA (Personal Digital Assistant) which is a mobile information terminal, and eyeglasses. It is an information processing terminal exemplified by a wearable device such as a type or a watch type.

The user terminal 40 has a function of receiving an operation input by the user 41 . The user terminal 40 also has a function for using the information processing device 10, for example, a function of acquiring various information from the information processing device 10 and displaying it on the display unit of the user terminal 40, and It has a function of generating various kinds of information in accordance with an operation input from and transmitting the information to the information processing apparatus 10 . The user terminal 40 performs these various functions by executing software or applications for using the information processing device 10 or by executing an Internet browser function for using the information processing device 10. come true.

Specifically, the user terminal 40 accesses the information processing device 10, displays the input format of each information of the search word 43 and the category designation information 45 received from the information processing device 10, and allows the user 41 to input this information. By transmitting each information of the search word 43 and the category designation information 45 inputted in the input format to the information processing device 10, the information of the search result 47 processed by the information processing device 10 is received and displayed.

The user 41 is an example of a purchaser who purchases the sales target 48 included in the sales target 24 registered in the information processing device 10 by using the information processing device 10 via the user terminal 40. Various procedures as a buyer in electronic commerce such as search, selection, settlement, evaluation, etc. of the sale object 48 to be purchased can be performed.

[2. Example of information processing system 1]
Next, the configuration of the information processing system 1 including the information processing device 10 will be described with reference to FIG. FIG. 2 is a diagram showing the configuration of an information processing system 1 including the information processing apparatus 10 of FIG. The information processing system 1 includes an information processing device 10 and a user terminal 40, as shown in FIG. In the information processing system 1, the information processing device 10 and the user terminal 40 are connected via a communication network 50 so as to be able to communicate information by wire or wirelessly. The information processing system 1 is not limited to the configuration including the number of devices and terminals shown in FIG. good.

[3. Configuration of Information Processing Device 10]
Next, the configuration of the information processing device 10 will be described with reference to FIG. FIG. 3 is a control block diagram of the information processing apparatus 10 of FIG. The information processing apparatus 10, as shown in FIG. 3, has a processing section 10A, a storage section 10B, and an information communication interface 10C.

(Regarding the processing unit 10A)
The processing unit 10A is a controller, and for example, various programs stored in the storage unit 10B, which is a storage device inside the information processing apparatus 10, by a CPU (Central Processing Unit), MPU (Micro Processing Unit), or the like. (corresponding to an example of an information processing program) is implemented by being executed using a RAM (Random Access Memory) as a work area. Also, the processing unit 10A is, for example, a controller, and is implemented by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array). The processing unit 10A is connected to an information communication interface 10C that receives input of information from the user terminal 40 and outputs information to the user terminal 40. FIG.

As shown in FIG. 3, the processing unit 10A is electrically connected to the storage unit 10B and the information communication interface 10C so as to be able to communicate information with each other, and functions as a control unit that controls these components. That is, the processing unit 10A functions as a control unit together with the storage unit 10B, and causes the information processing apparatus 10 to execute the information processing method according to the embodiment of the present invention.

As shown in FIG. 3, the processing unit 10A includes an input unit 11, a search unit 12, an output unit 13, an acquisition unit 14, an extraction unit 15, an integration unit 16, and a related information generation unit 17. have. The input unit 11 receives various information input from external terminals including the user terminal 40 via the communication network 50 via the information communication interface 10C. Specifically, the input unit 11 receives input of a search condition 42 including a search word 43 input by the user 41 at the user terminal 40 . The input unit 11 may further receive input of category designation information 45 included in the search condition 42 . The input unit 11 also receives various types of information on the sales target 24 necessary for creating a sales target database 21, which is a database on the sales target 24 to be searched by the search unit 12, which will be described later.

The input unit 11 creates a display screen 51 (see FIG. 4), transmits it to the user terminal 40 via the information communication interface 10C and the communication network 50, and displays the user 41 on the display unit of the user terminal 40. By displaying it visibly, the user is instructed to transmit the search condition 42 including the search word 43 and the category designation information 45 to the information processing apparatus 10 based on the input on the display screen 51 displayed on the user terminal 40 . Prompt 41. Note that the display screen 51 will be described later.

The search unit 12 selects the sales target 24 registered in advance in the sales target database 21 as a search target, which is a range to be searched based on the search word 43 input by the input unit 11 .

Specifically, the search unit 12 searches for sales targets based on the subjective expression 44 included in the search word 43 from among a plurality of sales target groups 35 classified based on the subjective expressions 32 included in the sale target information 27 . The group 35 is searched, and the sales target 48 included in the searched sales target group 35 is retrieved as the search result 47 .

More specifically, the search unit 12 first uses the subjective expression database 22, which is a database in which subjective expressions 32 and subjective expression groups 31 are associated with each other, to search for subjective expressions 44 included in the search word 43. Identify the subjective expression group 31 corresponding to . The search unit 12 identifies the subjective expression group 31 corresponding to the subjective expression 44, for example, based on the similarity 33 between the subjective expression 44 and the subjective expression 32 classified into the subjective expression group 31. can do. In addition, the search unit 12 inputs the subjective expression 44 included in the search word 43 as a query into a predetermined response model that executes processing using the subjective expression database 22. Based on the response result obtained, Subjective expression group 31 corresponding to subjective expression 44 can be identified.

The search unit 12 then uses the related information database 23, which is a database related to related information 34 that associates a subjective expression group 31 and a plurality of sales target groups 35, to identify the subjective expression group identified in the above process. A sales object group 35 that is a collection of sales objects 24 having sales object information 27 including any of the subjective expressions 32 associated with 31 is retrieved.

The search unit 12 further attaches the category information 26 corresponding to the category designated by the category designation information 45, that is, the sales target 24 belonging to the category designated by the category designation information 45 whose input is received by the input unit 11. It is also possible to search for the sale target 24 as a search range.

Based on the degree of similarity 33 between the subjective expression 44 included in the search word 43 and the subjective expression 32 included in the sale target information 27 of the sale target 24 included in the sale target group 35, the search unit 12 The objects for sale 48 that result in the retrieved search results 47 may be permuted. Specifically, the search unit 12 retrieves the sales target 48 whose sales target information 27 includes the subjective expression 32 having a higher similarity 33 to the subjective expression 44 included in the search word 43 whose input is received by the input unit 11. A higher order can be assigned by .

The output unit 13 outputs each information generated by the processing unit 10A and transmits the information to the user terminal 40 via the information communication interface 10C and the communication network 50. FIG. The output unit 13 outputs the sale target 48 retrieved by the retrieval unit 12 as a retrieval result 47 . Specifically, the output unit 13 generates and outputs a display screen 55 (see FIG. 5) based on the search results 47 when outputting the search results 47 . Note that the display screen 55 will be described later.

Further, when the search unit 12 has assigned the sales targets 48 with a permutation, the output unit 13 may arrange the sales targets 48 according to the permutation and output the search results 47 . More specifically, the output unit 13 generates and outputs a display screen 55 in which the sales targets 48 with higher permutations are arranged higher than the sales targets 48 with lower permutations.

The acquisition unit 14 acquires the sales target information 27 (see FIG. 6). The acquisition unit 14 acquires, for example, a description 28 (see FIG. 6) of the sale target 24 registered in advance in the sale target database 21 and a review 29 of the sale target 24 posted by the user who purchased the sale target 24. (see FIG. 6) is obtained as the sales target information 27.

The extraction unit 15 extracts subjective expressions 32, i.e., metaphorical expressions or adjectives that express the sales target 24 from the sales target information 27 acquired by the acquisition unit 14. Extract expressions based on sensory elements.

The integration unit 16 integrates the subjective expressions 32 extracted by the extraction unit 15 to generate a subjective expression group 31 . The integration unit 16 integrates the subjective expressions 32 to generate a subjective expression group 31 based on, for example, the degree of similarity 33 between the subjective expressions 32 extracted by the extraction unit 15 . Specifically, the integration unit 16 uses a machine learning or clustering technique to group together subjective expressions 32 that have a large degree of similarity 33, that is, have similar properties of expressions and have similar properties of objects to be expressed. together as a subjective expression group 31. The integration unit 16 thereby generates a subjective expression database 22 (see FIG. 7), which will be described later.

The related information generating unit 17 is an aggregate of the subjective expression group 31 generated by the integrating unit 16 and the sales target 24 which is the target of the sales target information 27 including the subjective expression 32 integrated into the subjective expression group 31. Related information 34 associated with a certain sales target group 35 is generated. The related information generation unit 17 thereby generates a related information database 23 (see FIG. 8), which will be described later.

Each unit included in the processing unit 10A, that is, the input unit 11, the search unit 12, the output unit 13, the acquisition unit 14, the extraction unit 15, the integration unit 16, and the related information generation unit 17, all of which the processing unit 10A performs information processing. It is a functional part realized by executing a program. Detailed functions of the input unit 11, the search unit 12, and the output unit 13 will be described later in the detailed description of the information processing method. Detailed functions of the acquisition unit 14, the extraction unit 15, the integration unit 16, and the related information generation unit 17 will be described later in the detailed description of the database creation method.

(Regarding storage unit 10B)
The storage unit 10B is realized by, for example, a semiconductor memory device such as a ROM (Read Only Memory), a RAM (Random Access Memory), a flash memory, or a storage device such as a hard disk or an optical disk. The storage unit 10B stores and saves a sales target database 21, a subjective expression database 22, and a related information database 23. FIG.

(Regarding information communication interface 10C)
The information communication interface 10C is a wired or wireless information communication interface, and connects the processing unit 10A and the user terminal 40 via the communication network 50 by wire or wirelessly so that information can be communicated with each other. The information communication interface 10C receives each information input and transmitted from the user terminal 40, for example, the search condition 42 including the search word 43 and the category designation information 45, and transmits the information to the input unit 11. . The information communication interface 10C also displays a display screen 51 for prompting the input of search conditions 42 including search words 43 and category designation information 45, and a display screen 55 based on search results 47 generated by the processing unit 10A. It is received from the processing unit 10A and transmitted to the user terminal 40. FIG.

(Regarding display screen 51)
FIG. 4 is a diagram showing an example of the display screen 51 displayed by the input unit 11 of FIG. As shown in FIG. 4, the display screen 51 includes a search word input section 52 for accepting input of a search word 43 by entering characters, and a category designation section for accepting input of category designation information 45 by checking each category. 53, the search word 43 input to the search word input unit 52 and the category designation information 45 input to the category designation unit 53 are sent to the information processing apparatus 10, and a button for inputting a search process. and a search button 54 . It is preferable that the categories displayed in the category designating portion 53 have a format corresponding to the category information 26 attached to each sale target 24 in the sale target database 21, which will be described later. As shown in FIG. 4, the display screen 51 is in a state in which "like a spring breeze" has been input as the search word 43 and the category "wine" has been input as the category designation information 45. FIG.

(Regarding the display screen 55)
FIG. 5 is a diagram showing an example of a display screen 55 generated by the output unit 13 of FIG. 3 based on the search results 47 generated by the search unit 12. As shown in FIG. As shown in FIG. 5, the display screen 55 is based on the search result 47 when "like spring breeze" is entered as the search word 43 and the category "wine" is entered as the category designation information 45. The sales target information 27 (see FIG. 6) includes the subjective expression 44 "like a spring breeze" included in the search word 43, and the sales targets 48 belonging to the category "wine" are "wine A" and "wine B", "Wine C", etc. are listed and displayed in the search item display portion 56 in the form of a list. By selecting each display of the sale target 48 “wine A”, “wine B” and “wine C” displayed in the search item display section 56, the details of the sale target 48 can be browsed or the sale target 48 can be displayed. It is a button that allows you to display a screen that allows you to perform purchase procedures for

(Regarding sales target database 21)
FIG. 6 is a diagram showing an example of the sale target database 21 of FIG. The sales target database 21, as shown in FIG. 6, is a database relating to all sales targets 24 for which the input unit 11 has received input of various kinds of information. All of the sales targets 24 registered in the sales target database 21 are assigned a sales target ID (Identifier) 25 in the sales target database 21 shown in FIG. Corresponding category information 26 is attached, and sale target information 27 about the sale target 24 is associated.

For example, in the sales object database 21 shown in FIG. 6, the description 28 included in the sales object information 27 of the sales object 24 whose sales object ID 25 is "Wine00001" includes the subjective expression 32 "like a spring breeze." found to be included. In the sales target database 21 shown in FIG. 6, the review 29 included in the sales target information 27 of the sale target 24 whose sales target ID 25 is "Wine00002" includes the subjective expression 32 "like a spring breeze." found to be included. From this, the sales target 24 whose sales target ID 25 is "Wine00001" includes the subjective expression 32 "like a spring breeze" in the description 28, and the sales target 24 whose review 29 includes the subjective expression 32 "like a spring breeze". The target ID 25 for sale 24 is "Wine00002". to be output to the output unit 13 as .

For the sales target ID 25, for example, a JAN (Japanese Article Number) code based on Japanese standards is preferably used. The sale item information 27 about the sale item 24 includes, for example, but is not limited to, a description 28 about the sale item 24 and a review 29 about the sale item 24, and other information about the sale item 24. may contain The description 28 is registered by the seller or affiliate of the seller who sells the sale item 24 in question. The reviews 29 are posted by the user who purchased the sale target 24, and each review 29 is associated with a review ID, posting date and time, poster ID, and content.

(Regarding subjective expression database 22)
FIG. 7 is a diagram showing an example of the subjective expression database 22 of FIG. As shown in FIG. 7, the subjective expression database 22 contains all the subjective expressions 32 extracted from the sales target information 27 by the extracting unit 15 and the subjective expressions 32 generated by integrating the subjective expressions 32 by the integrating unit 16. It is a database that associates groups 31 with each other. All subjective expressions 32 belong to at least one subjective expression group 31 and may belong to a plurality of subjective expression groups 31 in the subjective expression database 22 .

For example, the subjective expression group 31 whose subjective expression group ID is represented by S1 is the subjective expression 32 in the subjective expression database 22 shown in FIG. Like a refreshing spring” etc. are integrated and belong.

FIG. 9 is an explanatory diagram for explaining the similarity 33, which is an index of integration of the subjective expression database 22 of FIG. The similarity 33 table shown in FIG. 9 represents the similarity 33 of various subjective expressions 32 based on the subjective expression 32 "like a spring breeze". In the similarity 33 table shown in FIG. 9, the similarity 33 of the same subjective expression 32 "like spring breeze" is 100. In addition, the similarity 33 table shown in FIG. Like" is expressed with a degree of similarity 33 of 99, for example.

In addition, in the table of similarity 33 shown in FIG. Like a refreshing spring” is expressed with a similarity of 33, for example, 50.

Also, in the table of similarity 33 shown in FIG. The similarity 33 of the expression 32, ``image of blooming cherry blossoms'', is 30, for example. In addition, in the table of the degree of similarity 33 shown in FIG. 9, although there are elements reminiscent of "spring" that play a part of the main meaning of "like the spring wind", there is still the possibility of reminiscent of seasons other than spring. The similarity 33 of the subjective expression 32 "flower scent" is 20, for example.

In addition, in the similarity 33 table shown in FIG. 9, the subjective expression 32 is associated with "summer," which is clearly different from "spring," which is part of the main meaning of "like the spring breeze." The similarity 33 of "like fresh green in early summer" is 0. In addition, in the table of similarity 33 shown in FIG. , the similarity 33 is also represented as 0.

In addition, in the table of the similarity 33 shown in FIG. 9, the subjective expressions 32 with the similarity 33 of 50 or more are integrated into one subjective expression group 31 with the subjective expression group ID S1. It is

In addition, in the table of the similarity 33 shown in FIG. 9, regarding the subjective expression 32 with the similarity 33 represented by 30, the subjective expression group 31 to which "like the spring breeze" belongs is integrated into another subjective expression group 31 represented by S2. In addition, in the table of similarity 33 shown in FIG. Subjective expression group IDs having a relatively close relationship are integrated into another subjective expression group 31 represented by S3.

In addition, in the table of the similarity 33 shown in FIG. 9, the subjective expression 32 "like the fresh green of early summer" with the similarity 33 represented by 0 is a subjective expression to which "like the spring breeze" belongs. Subjective expression group IDs that have only the common point of being expressions using seasons with group 31 are integrated into another subjective expression group 31 represented by S4.

In addition, in the table of the similarity 33 shown in FIG. 9, the subjective expressions 32 "like silk" and "like velvet" with the similarity 33 being 0 are replaced with "like spring breeze". is integrated into completely different subjective expression groups 31 represented by T1 and T2, respectively.

In this embodiment, as described above, for all the subjective expressions 32, based on the characters of the expression, the meaning of the expression, the category of the expression, etc., the closeness of the nature of the expression and the closeness of the nature of the object to be expressed are evaluated. It is evaluated numerically based on a predetermined standard and adopted as an index for integration of the subjective expression database 22 as a degree of similarity 33 based on the predetermined standard. Then, all subjective expressions 32 are integrated into subjective expression groups 31 based on similarities 33 .

(Regarding the related information database 23)
FIG. 8 is a diagram showing an example of the related information database 23 of FIG. As shown in FIG. It is a database related to all related information 34 associated with a sales target group 35 that is a collection of targets 24 .

Specifically, in the related information database 23, each subjective expression 32 integrated into the subjective expression group 31 is associated with the sales target 24 that is the target of the sales target information 27 including the subjective expression 32. Just as expressions 32 are aggregated into subjective expression group 31 , bundles of sale objects 24 are aggregated to form sale object group 35 .

For example, the subjective expression group 31 whose subjective expression group ID is represented by S1 converts the subjective expression 32 integrated with the subjective expression group 31 into the sales target information 27 in the related information database 23 shown in FIG. A sale target group 35 is associated which includes "Wine A", "Wine B", and "Wine C" which are sale targets 24 containing.

[4. Example of information processing]
The operation of the information processing apparatus 10 according to the embodiment of the invention will be described below. FIG. 10 is a flowchart of an information processing method according to the embodiment. An information processing method according to an embodiment of the present invention executed by the information processing device 10 will be described with reference to FIG. do. The information processing method according to the embodiment is executed by the input unit 11, the search unit 12 and the output unit 13 of the processing unit 10A of the information processing apparatus 10, and as shown in FIG. 10, an input step S11, a search step S12, and an output step S13.

The input step S11 is a step in which the input unit 11 receives input of search conditions 42 including search words 43 input by the user 41 to the user terminal 40 . In the input step S<b>11 , the input unit 11 may further receive input of category designation information 45 included in the search condition 42 .

Specifically, in the input step S11, the input unit 11 first creates the display screen 51 shown in FIG. Search words 43 and categories are transmitted to the terminal 40 and displayed on the display unit of the user terminal 40 so that the user 41 can visually recognize them. The user 41 is urged to send the search condition 42 including the designation information 45 to the information processing apparatus 10 . In the input step S11, the input unit 11 thereafter receives the input information to the display screen 51 via the information communication interface 10C and the communication network 50, thereby obtaining the search conditions 42 including the search word 43 and the category designation information 45. accepts the input of

In the search step S12, the search unit 12 selects the search words 43 included in the search word 43 received in the input step S11 among the plurality of sale target groups 35 classified based on the subjective expression 32 included in the sale target information 27. This is a step of retrieving the sales object group 35 based on the subjective expression 44 and retrieving the sales object 48 included in the searched sales object group 35 as the search result 47 .

More specifically, in the search step S12 , the search unit 12 first uses the subjective expression database 22 to identify the subjective expression group 31 corresponding to the subjective expression 44 included in the search word 43 .

In the search step S12, the search unit 12, for example, based on the similarity 33 between the subjective expression 44 and the subjective expression 32 classified into the subjective expression group 31, the subjective expression corresponding to the subjective expression 44 Expression groups 31 can be specified. In the search step S12, when the search unit 12 identifies the subjective expression group 31 corresponding to the subjective expression 44 based on the similarity 33, first, the subjective expression 44 and each subjective expression 32 are searched for expressions characters, meanings of expressions, categories of expressions, etc., and based on the characters, meanings of expressions, categories of expressions, etc. of these expressions, the characteristics of the expressions of the subjective expressions 44 and the respective subjective expressions 32 are determined. A similarity 33 is derived based on a predetermined standard by evaluating the closeness, the closeness of the properties of the objects to be represented, numerically based on a predetermined standard. In the search step S12, the search unit 12 then selects a combination with a high derived similarity 33 for the subjective expression 44, for example, a combination whose average value of the similarity 33 is equal to or greater than a predetermined set value and the lowest similarity 33 The subjective expression group 31 to which the subjective expression 32 whose value is equal to or greater than the predetermined set value belongs is specified as the subjective expression group 31 corresponding to the subjective expression 44 . Note that in the search step S12, the search unit 12 may specify a plurality of subjective expression groups 31 to which the search word 43 belongs.

Further, in the search step S12, the search unit 12 inputs the subjective expression 44 included in the search word 43 as a query into a predetermined response model that executes processing using the subjective expression database 22. Based on the results, subjective expression group 31 corresponding to subjective expression 44 can be identified. In the search step S12, when the search unit 12 uses a predetermined response model to identify the subjective expression group 31 corresponding to the subjective expression 44, the response model whose response accuracy is improved as needed by machine learning is used. is preferred.

In the search step S12, the search unit 12 next searches for each subjective expression 32 associated with the subjective expression group 31 identified in the above process, based on the related information 34 registered in advance in the related information database 23. A sales object group 35 that is a collection of sales objects 24 having sales object information 27 including any of them is determined as a sales object group 35 that matches the search word 43 . In the search step S12, the search unit 12 then searches for the sale target 24 included in the sale target group 35 that matches the search word 43 determined in the above process.

In the search step S12, the search unit 12 then adds category information 26 corresponding to the category designated by the category designation information 45 among the sales objects 24 included in the sales object group 35 matching the search word 43. The limited sale target 24 is retrieved as a sale target 48 that becomes a search result 47 matching the search word 43 and the category designation information 45 .

In the search step S12, the search unit 12 further sorts the subjective expressions 44 included in the search word 43 and the subjective expressions 32 included in the sale target information 27 of the sale target 48 in descending order of similarity 33, respectively. may be permuted with respect to the sale object 48 of .

The output step S<b>13 is a step in which the output unit 13 outputs the sales target 48 searched in the search step S<b>12 as the search result 47 .

Specifically, in the output step S13, the output unit 13 first displays the display screen 55 shown in FIG. Generate. In the output step S13, the output unit 13 next transmits the display screen 55 generated by the above-described processing to the user terminal 40 via the information communication interface 10C and the communication network 50. The user 41 displays it on the display so that it can be visually recognized.

(Method of creating subjective expression database 22 and related information database 23)
Next, a method for creating the subjective expression database 22 and the related information database 23 will be described with reference to FIG. FIG. 11 is a flow chart of a method of creating the subjective expression database 22 and the related information database 23 of FIG. The method of creating the subjective expression database 22 and the related information database 23 is executed by the acquiring unit 14, the extracting unit 15, the integrating unit 16, and the related information generating unit 17 of the processing unit 10A of the information processing device 10, as shown in FIG. , an acquisition step S21, an extraction step S22, an integration step S23, and a related information generation unit S24.

Acquisition step S<b>21 is a step in which the acquisition unit 14 acquires the sale target information 27 about the sale target 24 registered in the sale target database 21 in advance. In the acquisition step S21, the acquiring unit 14 specifically acquires at least one of the description 28 of the sale target 24 and the review 29 of the sale target 24 posted by the user who purchased the sale target 24. Acquired as target information 27 .

In the extracting step S22, the extracting unit 15 extracts the subjective expression 32, that is, a metaphorical expression or an adjective expression about the sales target 24 from the sales target information 27 acquired by the acquisition unit 14 in the acquisition step S21. is a step of extracting an expression based on human sensory elements by the user 41 .

The integration step S23 is a step in which the integration unit 16 integrates the subjective expressions 32 extracted by the extraction unit 15 in the extraction step S22 to generate the subjective expression group 31 . Specifically, in the integrating step S23, the integrating unit 16 integrates the subjective expressions 32 based on the similarity 33 between the subjective expressions 32 extracted by the extracting unit 15 in the extracting step S22 to obtain subjective expressions. Generate group 31 . More specifically, in the integration step S23, the integration unit 16 uses a machine learning or clustering technique to determine the subjective expressions 32 that have a high degree of similarity 33, that is, that the properties of the expressions are similar and the properties of the objects to be represented are similar. By grouping them into a subjective expression group 31, the subjective expression database 22 shown in FIG. 7 is generated.

The related information generating unit S24 generates sales target information 27 including the subjective expression group 31 generated by the integrating unit 16 in the integration step S23 and the subjective expression 32 integrated into the subjective expression group 31. In this step, the related information database 23 shown in FIG. 8 is generated by generating the related information 34 in which the related information 34 is associated with the sales target group 35 which is a collection of the sales targets 24 which are the targets of the above.

In this way, the information processing apparatus 10 can generate the subjective expression database 22 and the related information database 23 that the information processing apparatus 10 employs when executing the information processing method according to the embodiment.

[5. Hardware configuration]
The information processing apparatus 10 according to the embodiments described above is implemented by a computer 1000 configured as shown in FIG. 12, for example. FIG. 12 is a hardware configuration diagram showing an example of a computer 1000 that implements the functions of the information processing apparatus 10. As shown in FIG. Computer 1000 has CPU 1100 , RAM 1200 , ROM 1300 , HDD 1400 , communication interface (I/F) 1500 , input/output interface (I/F) 1600 and media interface (I/F) 1700 .

The CPU 1100 operates based on programs stored in the ROM 1300 or HDD 1400 and controls each section. The ROM 1300 stores a boot program executed by the CPU 1100 when the computer 1000 is started, programs depending on the hardware of the computer 1000, and the like.

The HDD 1400 stores programs executed by the CPU 1100, data used by the programs, and the like. Communication interface 1500 receives data from other devices via communication network 500 and sends the data to CPU 1100 , and transmits data generated by CPU 1100 to other devices via communication network 500 .

The CPU 1100 controls output devices such as displays and printers, and input devices such as keyboards and mice, through an input/output interface 1600 . CPU 1100 acquires data from an input device via input/output interface 1600 . Also, CPU 1100 outputs the generated data to an output device via input/output interface 1600 .

Media interface 1700 reads programs or data stored in recording medium 1800 and provides them to CPU 1100 via RAM 1200 . CPU 1100 loads such a program from recording medium 1800 onto RAM 1200 via media interface 1700, and executes the loaded program. The recording medium 1800 is, for example, an optical recording medium such as a DVD (Digital Versatile Disc) or a PD (Phase change rewritable disc), a magneto-optical recording medium such as an MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory. etc.

For example, when the computer 1000 functions as the information processing apparatus 10 according to the embodiment, the CPU 1100 of the computer 1000 implements the functions of the processing section 10A by executing a program loaded onto the RAM 1200 . In addition, the data in storage unit 10B is stored in HDD 1400 . CPU 1100 of computer 1000 reads these programs from recording medium 1800 and executes them, but as another example, these programs may be obtained from other devices via communication network 500 .

[6. others〕
Further, among the processes described in the above embodiments, all or part of the processes described as being automatically performed can be manually performed, or the processes described as being performed manually can be performed manually. All or part of this can also be done automatically by known methods. In addition, information including processing procedures, specific names, various data and parameters shown in the above documents and drawings can be arbitrarily changed unless otherwise specified. For example, the various information shown in each drawing is not limited to the illustrated information.

Also, each component of each device illustrated is functionally conceptual, and does not necessarily need to be physically configured as illustrated. In other words, the specific form of distribution and integration of each device is not limited to the one shown in the figure, and all or part of them can be functionally or physically distributed and integrated in arbitrary units according to various loads and usage conditions. Can be integrated and configured. For example, information such as the sales target database 21, the subjective expression database 22, and the related information database 23 stored in the storage unit 10B shown in FIG. may be stored in a predetermined storage device provided in the

Further, in the above embodiment, the information processing apparatus 10 executes the information processing method according to the embodiment by performing the input step S11, the search step S12, and the output step S13, for example. However, the information processing apparatus 10 described above may be separated into an input device that performs the input step S11, a search device that performs the search step S12, and an output device that performs the output step S13. The processing related to the information processing method by the information processing apparatus 10 is implemented by an information processing system having an input device, a search device, and an output device.

Further, in the above-described embodiment, the information processing apparatus 10 performs, for example, the acquisition step S21, the extraction step S22, the integration step S23, and the related information generation unit S24, so that the subjective expression database 22 and the related information An example of creating the database 23 has been shown. However, the information processing apparatus 10 described above includes an acquisition device that performs the acquisition step S21, an extraction device that performs the extraction step S22, an integration device that performs the integration step S23, and a related information generation device that performs the related information generation unit S24. may be separated. Then, the processing related to the method of creating the subjective expression database 22 and the related information database 23 by the information processing device 10 described above is performed by an information processing device having an acquisition device, an extraction device, an integration device, and a related information generation device. realized by the system.

Also, the above-described embodiments and modifications can be appropriately combined within a range that does not contradict the processing contents.

[7. effect〕
As described above, the information processing apparatus 10 according to the embodiment has the input unit 11, the search unit 12, and the output unit 13. Input unit 11 receives search word 43 . The search unit 12 searches a plurality of sales target groups 35 classified based on the subjective expressions 32 included in the sales target information 27, and searches for sales based on the subjective expressions 44 included in the search words 43 received by the input unit 11. A subject group 35 is searched. The output unit 13 outputs the sale target 48 retrieved by the retrieval unit 12 as a retrieval result 47 .

As described above, the information processing apparatus 10 according to the embodiment enables the search unit 12 to perform search processing based on the subjective expression 44 included in the search word 43 . More specifically, the information processing apparatus 10 according to the embodiment uses a metaphorical expression that metaphorically expresses the sales target 48 desired by the user 41, an expression using adjectives, or a subjective expression by the user 41 based on human sensory factors. Sale items 24 that match search words 43 that include generic expressions 44 may be retrieved. Therefore, the information processing apparatus 10 according to the embodiment enables the user 41 to search for the sale target 24 using the search word 43 including the subjective expression 44 .

Conventionally, the search for the sales target 24 with the search word 43 including the subjective expression 44 requires the user 41 to understand the search to some extent, for example, it is necessary to replace the specific sales target 24 with words that directly describe the target sale 24 . literacy was required. However, conventionally, the sales target 24 was the primary key for searching the sales target 24, but the information processing apparatus 10 according to the embodiment has a search configuration in which the subjective expression 32 is used as the primary key. Even if the person 41 does not have a certain degree of literacy regarding such searches, it is possible to search for the sales target 24 with search words 43 including subjective expressions 44 .

Further, the search unit 12 searches for a sales target group 35 based on the subjective expression 44 included in the search word 43 from among the plurality of sales target groups 35 classified according to the subjective expression group 31 .

As described above, the information processing apparatus 10 according to the embodiment further uses the correspondence between the sales target group 35 and the subjective expression group 31 to identify sales targets that match the search word 43 including the subjective expression 44 . 24 can be searched, fluctuations in the subjective expressions 44 included in the search words 43 can be dealt with, and search for the sales target 24 by the search words 43 including stable subjective expressions 44 can be provided. making it possible.

In addition, the search unit 12 identifies the subjective expression group 31 corresponding to the subjective expression 44 included in the search word 43 , thereby searching for sales that include any of the subjective expressions 32 included in the subjective expression group 31 . A sales object group 35 that is a collection of sales objects 24 having object information 27 is searched.

As described above, the information processing apparatus 10 according to the embodiment further has a search configuration in which the subjective expression group 31 obtained by integrating the subjective expressions 32 is used as a primary key. Fluctuations in the expression 44 can be dealt with, and it is possible to provide a search for the sales target 24 using a search word 43 containing a more stable subjective expression 44 .

In addition, the search unit 12 uses the similarity 33 between the subjective expression 44 included in the search word 43 and the subjective expression group 31 classified into the subjective expression group 31, based on the degree of similarity 33. A subjective expression group 31 corresponding to the subjective expression 44 is identified. Alternatively, the search unit 12 associates subjective expressions 44 included in the search word 43 with the subjective expression group 31 obtained when the subjective expression 44 included in the search word 43 is input as a query into a predetermined response model. It is specified as subjective expression group 31 .

In this way, the information processing apparatus 10 according to the embodiment, based on the similarity 33 or the learned response model, finds the subjective expression group 31 corresponding to the subjective expression 44 included in the search word 43 with higher accuracy. The ability to identify them makes it possible to provide searches for sale items 24 by search words 43 that contain more precise subjective expressions 44 .

Further, based on the degree of similarity 33 between the subjective expression 44 included in the search word 43 and the subjective expression 32 included in the sale target information 27 of the sale target 24 included in the sale target group 35, the search unit 12 The searched sales objects 48 are permuted, and the output unit 13 arranges the sales objects 48 according to the permutation and outputs search results 47 .

In this way, in the information processing apparatus 10 according to the embodiment, the search unit 12 attaches a higher permutation to the sales target 48 highly matching the subjective expression 44 included in the search word 43, and outputs the result. Since the unit 13 outputs the sales target 48 having high relevance to the subjective expression 44 included in the search word 43 with higher permutation so as to rank higher, the possibility that the user 41 is searching more is increased. It is possible to recognize a sales target 48 with high probability at a higher rank.

The information processing apparatus 10 according to the embodiment further includes an acquisition unit 14 , an extraction unit 15 , an integration unit 16 and a related information generation unit 17 . The acquisition unit 14 acquires the sale target information 27 . The extraction unit 15 extracts subjective expressions 32 from the sales target information 27 acquired by the acquisition unit 14 . The integration unit 16 integrates the subjective expressions 32 extracted by the extraction unit 15 to generate a subjective expression group 31 . The related information generating unit 17 is an aggregate of the subjective expression group 31 generated by the integrating unit 16 and the sales target 24 which is the target of the sales target information 27 including the subjective expression 32 integrated into the subjective expression group 31. Related information 34 associated with a certain sales target group 35 is generated. Based on the related information 34 , the search unit 12 then searches for the sale target 24 included in the sale target group 35 associated with the subjective expression 44 included in the search word 43 .

As described above, the information processing apparatus 10 according to the embodiment can generate the subjective expression database 22 and the related information database 23 that are employed when the information processing apparatus 10 executes the information processing method according to the embodiment. Therefore, the subjective expression database 22 and the related information database 23 are automatically updated as appropriate based on the sales target 24 newly registered in the sales target database 21, and the user 41 can use the subjective expression 44 in the latest state. allows searching for sale items 24 with search words 43 including

Also, the integration unit 16 integrates the subjective expressions 32 based on the degree of similarity 33 between the subjective expressions 32 extracted by the extraction unit 15 to generate a subjective expression group 31 .

In this way, the information processing apparatus 10 according to the embodiment divides the subjective expressions 32 whose similarities 33 between the subjective expressions 32 are large, that is, the properties of the expressions are close to each other and the properties of the objects to be expressed are close to each other, into one set. Since the subjective expression group 31 can be integrated into the subjective expression group 31, it is possible to generate the subjective expression database 22 suitable for searching with the search word 43 including the subjective expression 44 for the user 41. .

The input unit 11 also receives an input of a metaphorical expression or an adjective expression as the subjective expression 44 .

As described above, the information processing apparatus 10 according to the embodiment can receive an input of a metaphorical expression or an adjective expression as the subjective expression 44, so that the user 41 can input the subjective expression more intuitively and in consideration of trends. Allows searching for sale items 24 with search words 43 containing 44.

In addition, the input unit 11 further receives input of search conditions 42 including category designation information 45 that designates the category of the search range of the sale target 24, and the search unit 12 belongs to the category designated by the category designation information 45. Search for sale object 24 .

As described above, the information processing apparatus 10 according to the embodiment allows the search unit 12 to perform search processing in which the search range is limited to the sales targets 24 belonging to the category designated by the category designation information 45. makes it possible to search for sales objects 24 with search words 43 including subjective expressions 44 within the range of categories designated by category designation information 45 .

Also, the information processing method according to the embodiment has an input step S11, a search step S12, and an output step S13. The input step S<b>11 is a step of receiving the search word 43 . In the search step S12, among a plurality of sales target groups 35 classified based on the subjective expressions 32 included in the sales target information 27, the subjective expressions 44 included in the search word 43 received in the input step S11 are searched. This is a step of searching for a sales target group 35 based on The output step S13 is a step of outputting the sale target 48 retrieved in the retrieval step S12 as the retrieval result 47. FIG.

Thus, the information processing method according to the embodiment enables search processing based on the subjective expression 44 included in the search word 43 in the search step S12. More specifically, the information processing method according to the embodiment is a metaphorical expression that metaphorically expresses the sales target 48 desired by the user 41, an expression using adjectives, or a subjective Items for sale 24 matching search words 43 containing expressions 44 may be retrieved. Therefore, the information processing method according to the embodiment enables the user 41 to search for the sale target 24 using the search word 43 including the subjective expression 44 .

Further, the information processing program according to the embodiment is for causing the computer 1000 to execute an input procedure, a search procedure, and an output procedure. The input procedure is a procedure for accepting the search word 43 . In the search procedure, among a plurality of sales object groups 35 classified based on the subjective expressions 32 contained in the sales object information 27, sales are performed based on the subjective expressions 44 included in the search words 43 received in the input procedure. It is a procedure for searching the target group 35 . The output procedure is a procedure for outputting the sale target 48 searched by the search procedure as the search result 47 .

Thus, the information processing program according to the embodiment enables the computer 1000 to perform search processing based on the subjective expressions 44 included in the search words 43 in the search procedure. More specifically, the information processing program according to the embodiment is a metaphorical expression that metaphorically expresses the sales target 48 desired by the user 41, an expression using adjectives, or a subjective Offers 24 matching search words 43 containing expressions 44 may be retrieved. Therefore, the information processing program according to the embodiment enables the user 41 to search for the sale target 24 using the search word 43 including the subjective expression 44 .

As described above, the embodiments of the present application have been described in detail with reference to the drawings, but these are examples, and various modifications and improvements, including the aspects described in the section of the disclosure of the invention, can be made based on the knowledge of those skilled in the art. It is possible to implement the invention in other forms.

Also, the above-mentioned "section, module, unit" can be read as "means" or "circuit". For example, the search unit can be read as search means or a search circuit.

1 information processing system 10 information processing device 10A processing unit 10B storage unit 10C information communication interface 11 input unit 12 search unit 13 output unit 14 acquisition unit 15 extraction unit 16 integration unit 17 related information generation unit 21 sales target database 22 subjective expression database 23 Related information database 24, 48 Sales target 26 Category information 27 Sales target information 28 Description 29 Review 31 Subjective expression group 32, 44 Subjective expression 33 Similarity 34 Related information 35 Sales target group 40 User terminal 41 User 42 Search Conditions 43 Search words 45 Category designation information 47 Search results

Claims (11)

an input unit that accepts a search word;
an acquisition unit for acquiring sales target information;
an extraction unit for extracting a subjective expression from the sales target information acquired by the acquisition unit;
an integrating unit configured to generate a subjective expression group by integrating, among the subjective expressions extracted by the extracting unit, the subjective expressions having similar characteristics of the subject represented by the subjective expressions;
The subjective expression group generated by the integration unit is associated with a sales target group that is a collection of sales targets that are targets of the sales target information including the subjective expression integrated into the subjective expression group. a related information generation unit that generates related information;
subjective expressions included in the search word received by the input unit, and the subjective a search unit that searches for sales targets included in the sales target group associated with the subjective expression group identified based on the degree of similarity with the subjective expression group;
an output unit that outputs the sales target searched by the search unit as a search result;
has
The acquisition unit acquires, as the sale target information, at least one of a description of the sale target registered in advance in a predetermined database and a review posted by a user who purchased the sale target.
An information processing device characterized by: The search unit searches for a sales target group based on the subjective expression included in the search word, among the plurality of sales target groups classified by subjective expression group.
The information processing apparatus according to claim 1, characterized by: The search unit specifies a subjective expression group corresponding to the subjective expression included in the search word, and has the sales target information including any of the subjective expressions included in the subjective expression group. searching for the sale target group, which is a collection of sale targets;
3. The information processing apparatus according to claim 2, characterized by: The search unit searches the subjective expressions included in the search word based on the degree of similarity between the subjective expressions included in the search word and the subjective expressions classified into the subjective expression group. identify the subjective representation group corresponding to
4. The information processing apparatus according to claim 3, characterized by: The search unit retrieves a subjective expression group obtained when the subjective expression included in the search word is input as a query into a predetermined response model as a subjective expression corresponding to the subjective expression included in the search word. identify as a group,
4. The information processing apparatus according to claim 3, characterized by: The search unit searches based on the degree of similarity between the subjective expression included in the search word and the subjective expression included in the sales target information of the sales target included in the sales target group. Permuting the objects for sale,
The output unit arranges the sales targets according to the permutation and outputs the search results.
6. The information processing apparatus according to any one of claims 1 to 5, characterized by: The integration unit generates a subjective expression group by integrating the subjective expressions based on the similarity between the subjective expressions extracted by the extraction unit.
7. The information processing apparatus according to any one of claims 1 to 6, characterized by: The input unit receives an input of a metaphorical expression or an adjective expression as the subjective expression;
8. The information processing apparatus according to any one of claims 1 to 7, characterized by: The input unit further receives input of search conditions including category designation information designating a category of the search range of the sale target,
The search unit further searches for the sales target belonging to the category specified by the category specifying information.
9. The information processing apparatus according to any one of claims 1 to 8, characterized by: A computer-implemented information processing method comprising:
an input step that accepts a search word;
an acquisition step of acquiring information to be sold;
an extraction step of extracting a subjective expression from the sales target information acquired by the acquisition step;
an integration step of generating a subjective expression group by integrating the subjective expressions extracted by the extracting step, the subjective expressions having similar characteristics of the subject represented by the subjective expressions;
The group of subjective expressions generated by the integration step is associated with a group of sales objects, which is a collection of sales objects that are objects of the sales object information including the subjective expressions integrated into the subjective expression group. a related information generating step for generating related information;
Based on the relevant information, subjective expressions included in the search word received in the input step and subjective expressions included in the search word received in the input step, among a plurality of sales target groups classified based on the subjective expressions included in the sales target information a search step of searching for sales objects included in the sales object group associated with the subjective expression group identified based on the degree of similarity with the subjective expression group;
an output step of outputting the sales target searched in the search step as a search result;
has
The obtaining step obtains, as the sales target information, at least one of a description of the sale target registered in advance in a predetermined database and a review posted by a user who purchased the sale target.
An information processing method characterized by: to the computer,
an input procedure for accepting search words;
Acquisition procedure for acquiring sales target information;
an extraction procedure for extracting a subjective expression from the sales target information acquired by the acquisition procedure;
an integration procedure for generating a subjective expression group by integrating the subjective expressions extracted by the extraction procedure, the subjective expressions having similar characteristics of the subject represented by the subjective expressions;
The subjective expression group generated by the integration procedure is associated with a sales object group that is a collection of sales objects that are objects of the sales object information including the subjective expression integrated into the subjective expression group. a related information generation procedure for generating related information;
Based on the relevant information, subjective expressions included in the search word accepted in the input procedure, and the subjective a search procedure for searching sales targets included in the sales target group associated with the subjective expression group identified based on the degree of similarity with the subjective expression group;
an output step of outputting the sales target searched by the search step as a search result;
and
The acquisition step acquires at least one of a description of a sale target registered in advance in a predetermined database and a review posted by a user who purchased the sale target as the sale target information.
Information processing program for.

Images