JP6731826B2 - Extraction device, extraction method, and extraction program - Google Patents

Description

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

Conventionally, information such as a feature amount extracted from various information has been used. For example, the feature amount of the content of the search query (hereinafter, also simply referred to as a “query”) is extracted, the content having the feature amount that is highly identical to the feature amount is selected, and based on the related meta information that the content has. There is provided a technique of determining the degree of similarity, grouping and displaying.

JP, 2007-164633, A

"Word2vec", [online], [July 29, 2016 search], Internet <URL: https://code.google.com/p/word2vec/>

However, in the above-mentioned conventional technology, it is not always possible to extract information characteristic of a specific user. For example, it may not be possible to extract information characteristic of a specific user only by extracting information based on information of similar users.

The present application has been made in view of the above, and an object thereof is to provide an extraction device, an extraction method, and an extraction program that extract characteristic information for a specific user.

An extraction device according to the present application, an acquisition unit that acquires information regarding a search query used by a target user who performed a predetermined action, and information regarding a search query used by another user who did not perform the predetermined action, The information about the search query used by the target user is classified into the first cluster based on the similarity between the vectors corresponding to the information about the search query used by the target user, and the search queries used by the other users are classified. Based on a difference between the first cluster and the second cluster, and a classification unit that classifies information about a search query used by the other user into a second cluster based on a similarity between vectors corresponding to the information about An extraction unit that extracts a characteristic cluster, which is a cluster showing a characteristic behavior of the target user, from the first cluster.

According to one aspect of the embodiment, it is possible to extract information characteristic of a specific user.

FIG. 1 is a diagram illustrating an example of an extraction process according to the embodiment. FIG. 2 is a diagram illustrating a configuration example of the extraction device according to the embodiment. FIG. 3 is a diagram illustrating an example of the user information storage unit according to the embodiment. FIG. 4 is a diagram illustrating an example of the search history information storage unit according to the embodiment. FIG. 5 is a diagram illustrating an example of the vector information storage unit according to the embodiment. FIG. 6 is a diagram illustrating an example of the feature cluster information storage unit according to the embodiment. FIG. 7 is a diagram illustrating an example of the advertisement information storage unit according to the embodiment. FIG. 8 is a flowchart showing an extraction processing procedure according to the embodiment. FIG. 9 is a diagram illustrating an example of the advertisement distribution process according to the embodiment. FIG. 10 is a flowchart showing an advertisement distribution processing procedure according to the embodiment. FIG. 11 is a hardware configuration diagram illustrating an example of a computer that realizes the function of the extraction device.

Hereinafter, modes (hereinafter, referred to as “embodiments”) for carrying out an extraction device, an extraction method, and an extraction program according to the present application will be described in detail with reference to the drawings. Note that the extraction device, the extraction method, and the extraction program according to the present application are not limited by this embodiment. Also, in each of the following embodiments, the same parts are designated by the same reference numerals, and duplicated description will be omitted.

(Embodiment)
[1. Extraction process]
First, an example of the extraction process according to the embodiment will be described with reference to FIG. FIG. 1 is a diagram illustrating an example of an extraction process according to the embodiment. In FIG. 1, the extraction device 100 (see FIG. 2) is based on two groups: a group of target users who have performed a predetermined action on a plurality of users, and a group of other users who have not performed a predetermined action, The case where a feature cluster is extracted is shown. In the example of FIG. 1, the case where the predetermined action is the purchase of the personal computer of company A is shown. Moreover, in the example of FIG. 1, the case where the user's action information is the token used in the search by the user is shown. For example, the "token" here means a character string delimited by a predetermined delimiter included in the query used by the user. For example, "token" means each character string delimited by a space in the query used by the user. For example, if the search query is “personal computer performance”, the character string “personal computer” or the character string “performance” becomes a token.

For example, the user group GL1 (group A) in FIG. 1 includes users U1 and U2 who are users who purchased the personal computer of company A. In the example of FIG. 1, it is assumed that 1000 users are included in the user group GL1 which is a personal computer purchasing user group of the company A. Further, for example, the user group GL2 (group B) in FIG. 1 includes users U3 and U4 who are users who have not purchased the personal computer of company A. In the example of FIG. 1, it is assumed that 2000 users are included in the user group GL2, which is the personal computer non-purchasing user group of Company A.

Further, the extraction device 100 collects the tokens used by each user in the search in a predetermined period. Further, the predetermined period regarding the user who has performed the predetermined action may be a certain period (for example, one month) before the corresponding user performs the predetermined action, or the like. For example, the period for collecting tokens for the user U1 may be a period from the time when the user U1 purchased the personal computer A to one month ago. Further, the predetermined period regarding the user who did not perform the predetermined action may be a certain period (for example, one month) before performing the extraction process. For example, the period for collecting tokens for the user U3 may be a period from the time when the extraction device 100 performs the extraction process to one month ago. The above-mentioned predetermined period is an example, and may be changed to various periods according to the purpose and the like.

In the example shown in FIG. 1, the used token information TK1 indicates the token used by the user U1 in the search in a predetermined period. For example, the used token information TK1 includes tokens such as “personal computer”, “company B model number X”, “strength”, and “charge”. Further, in the example shown in FIG. 1, the used token information TK2 indicates the token used by the user U2 in the search in the predetermined period. For example, the used token information TK2 includes tokens such as “notebook PC”, “company B model number Y”, “light weight”, “price” and the like.

In the example shown in FIG. 1, the used token information TK3 indicates the token used by the user U3 in the search in a predetermined period. For example, the used token information TK3 includes tokens such as “smartphone”, “stock price”, “PC”, and “soccer”. Further, in the example shown in FIG. 1, the used token information TK4 indicates the token used by the user U4 in the search in the predetermined period. For example, the used token information TK4 includes tokens such as "job change", "mail", "transportation expense", "no-paso", and the like.

First, the extraction device 100 classifies tokens used by users included in group A into clusters (step S11). Specifically, the extraction device 100 classifies the tokens used by the users included in the group A into the first cluster based on the similarity of the tokens. In the example of FIG. 1, the extraction device 100 classifies the tokens used by the users included in the group A into the first cluster based on the similarity of the vector corresponding to each token.

For example, the extraction device 100 may use the vector of each token derived based on the technique related to “word2vec” in Non-Patent Document 1. Note that the extraction device 100 may use the vector of each token derived by any technique as long as the vector corresponding to each token can be derived. For example, the extraction device 100 may use the vector of each token that the vector of each token is more similar as the meaning of each token is closer. In this case, for example, the vector corresponding to the token "PC" and the vector corresponding to the token "PC" are similar.

Further, for example, the extraction device 100 classifies the tokens used by the users included in the group A into the first cluster by appropriately using various conventional techniques such as the vector of each token described above and Kmeans. For example, the extraction device 100 classifies the tokens used by the users included in the group A into the first cluster so that the number of clusters becomes a predetermined number (for example, 1000). Note that the extraction device 100 may use any clustering technique as long as the tokens can be classified into a plurality of clusters based on the above-described similarity of the vector of each token.

In the example of FIG. 1, the extraction device 100 classifies the tokens used by the users included in the group A into the first cluster, as shown in the first cluster group CL11. For example, in the cluster identified by the cluster ID “C11” in the first cluster group CL11 (hereinafter sometimes referred to as “cluster C11”), “PC”, “notebook PC”, “PC”, “ Indicates that a token such as "Nopaso" is included. For example, “personal computer”, “notebook PC”, “PC”, and “no-paso” are words indicating a common personal computer, and are therefore classified by the extraction device 100 into the cluster C11 as semantically similar tokens. Further, for example, in the cluster identified by the cluster ID “C12” in the first cluster group CL11 (hereinafter sometimes referred to as “cluster C12”), “B company model number X” and “B company model number Y” are included. ”, “ZA series”, and “ZB series” are included. For example, in the extraction device 100, “B company model number X” and “B company model number Y” are B company personal computers, and “ZA series” and “ZB series” are B company personal computer series. Is classified as a cluster C12 as a token similar to. Further, for example, in the cluster identified by the cluster ID “C13” in the first cluster group CL11 (hereinafter sometimes referred to as “cluster C13”), “price”, “cost”, “price”, Indicates that a token such as "transportation expenses" is included.

Further, the extraction device 100 classifies the tokens used by the users included in the group B into clusters (step S12). Specifically, the extraction device 100 classifies the tokens used by the users included in the group B into the second cluster based on the similarity of the tokens. In the example of FIG. 1, for simplification of description, the extraction device 100 classifies the tokens used by the users included in the group B into the second cluster similar to the first cluster. In the example of FIG. 1, the extraction device 100 classifies the tokens used by the users included in the group B into the second cluster, as shown in the second cluster group CL12. That is, in the example of FIG. 1, the extraction device 100 classifies the tokens used by the users included in the group B into the second cluster similar to the first cluster in the first cluster group CL11.

Then, the extraction device 100 extracts, from the first cluster, a characteristic cluster that is a cluster that shows a characteristic behavior of the target user, based on the difference between the first cluster and the second cluster (step S13). In the example of FIG. 1, the extraction device 100 extracts the characteristic cluster based on the information regarding the usage rate of the tokens belonging to each cluster shown in the first cluster group CL11 and the second cluster group CL12.

In the example of FIG. 1, as shown in the first cluster group CL11, the usage rate of the tokens included in the cluster C11 in the group A is 80%. That is, in the example of FIG. 1, the token included in the cluster C11 is used by 800 users out of 1000 users in the group A. It should be noted that the “usage rate” mentioned here indicates the proportion of users who have used the tokens belonging to the corresponding clusters. For example, a certain user may select “personal computer”, “notebook PC”, “PC”, or “no-personal computer” Even if a plurality of tokens such as “” are used, the user is totaled as “1 (user)”.

Further, in the example of FIG. 1, as shown in the first cluster group CL11, the usage rate of the tokens included in the cluster C12 in the group A is 60%. That is, in the example of FIG. 1, the token included in the cluster C12 is used by 600 users of the 1000 users in the group A. Further, in the example of FIG. 1, as shown in the first cluster group CL11, the usage rate of the tokens included in the cluster C13 in the group A is 20%. That is, in the example of FIG. 1, the token included in the cluster C13 is used by 200 users of the 1000 users in the group A.

Further, in the example of FIG. 1, as shown in the second cluster group CL12, the usage rate of the tokens included in the cluster C11 in the group B is 60%. That is, in the example of FIG. 1, the token included in the cluster C11 is used by 1200 of the 2000 users in group B. Further, in the example of FIG. 1, as shown in the second cluster group CL12, the usage rate of the tokens included in the cluster C12 in the group B is 5%. That is, in the example of FIG. 1, the token included in the cluster C12 is used by 100 users out of 2000 users in the group B. Further, in the example of FIG. 1, as shown in the second cluster group CL12, the usage rate of the tokens included in the cluster C13 in the group B is 20%. That is, in the example of FIG. 1, the token included in the cluster C13 is used by 400 users of the 2000 users in the group B.

The extraction device 100 uses the above-mentioned information about the usage rate to extract, as a feature cluster, a cluster whose evaluation value based on the difference between the usage rate in the group A and the usage rate in the group B is equal to or more than a predetermined threshold value. In the example of FIG. 1, the extraction device 100 extracts a cluster in which the ratio (evaluation value) of the usage rate in the group A to the usage rate in the group B is 3 or more as a feature cluster.

For example, in the extraction device 100, as for the cluster C11, the usage rate in the group A is 80% and the usage rate in the group B is 60%, and thus the evaluation value is “1.33 (=80/60)”. Therefore, the cluster C11 is not extracted as a feature cluster. Further, for example, in the extraction device 100, as for the cluster C13, the usage rate in the group A is 20% and the usage rate in the group B is 20%, so that the evaluation value is “1 (=20/20)”. Therefore, the cluster C13 is not extracted as a feature cluster.

Further, for example, in the extraction device 100, for the cluster C12, the usage rate in the group A is 60%, and the usage rate in the group B is 5%, so that the evaluation value is “12 (=60/5)”. Therefore, the cluster C13 is extracted as a feature cluster. In this way, the extraction device 100 extracts a cluster characteristic of the user who purchased the personal computer of the company A as a characteristic cluster. In the example of FIG. 1, the extraction device 100 extracts, as the characteristic cluster information storage unit 124, a cluster having an evaluation value of 3 or more, such as the cluster C12 or the cluster C21, as a characteristic cluster.

As described above, the extraction device 100 extracts the characteristic cluster, which is a cluster showing the characteristic behavior of the target user, from the first cluster based on the difference between the first cluster and the second cluster. Information characteristic of the user can be extracted. The extraction device 100 may provide the extracted feature cluster to an external information processing device. The extraction device 100 may also provide various services based on the extracted feature cluster. The details of the service provision based on the feature cluster extracted by the extraction device 100 will be described later. Further, in the above example, the case where the token included in the query is used as the action information has been described, but the extraction device 100 extracts any of the above feature clusters using any information as long as it is the action information of the user. You may perform the process. For example, the extraction device 100 may perform the above extraction process using any information as long as it is the behavior information of the user and a vector corresponding to the behavior information can be acquired.

For example, the extraction device 100 may extract the feature cluster based on the action information that accompanies the movement of the user, with the predetermined action of going to a certain place. For example, the extraction device 100 sets a user who has gone to the place A (for example, the sea, a restaurant, etc.) as a target user based on the action information that accompanies the movement of the user, and the target user has a predetermined period of time before going to the place A. The feature cluster may be extracted using the first cluster in which the behavior information is clustered. In this case, for example, the extraction device 100 may use a user who did not go to the place A as another user, cluster the behavior information of the other user for a predetermined period, and use the clustered information as the second cluster. In addition, for example, the extraction device 100 may use, as the user's action information, action information related to the user driving a car or the like. Note that the above is an example, and the extraction device 100 may perform the above-described feature cluster extraction processing by using any information as long as the information allows the above-described feature cluster extraction processing.

[2. Configuration of extraction device]
Next, the configuration of the extraction device 100 according to the embodiment will be described with reference to FIG. FIG. 2 is a diagram illustrating a configuration example of the extraction device according to the embodiment. As shown in FIG. 2, the extraction device 100 includes a communication unit 110, a storage unit 120, and a control unit 130. The extraction device 100 may include a display unit that displays various types of information and an input unit that inputs various types of information.

(Communication unit 110)
The communication unit 110 is realized by, for example, a NIC (Network Interface Card) or the like. The communication unit 110 is connected to a predetermined network in a wired or wireless manner, and transmits/receives information to/from an external information processing device such as the terminal device 10 (see FIG. 9).

(Storage unit 120)
The storage unit 120 is realized by, for example, a semiconductor memory device such as a RAM (Random Access Memory) or a flash memory (Flash Memory), or a storage device such as a hard disk or an optical disk. As shown in FIG. 2, the storage unit 120 according to the embodiment has a user information storage unit 121, a search history information storage unit 122, a vector information storage unit 123, a feature cluster information storage unit 124, and an advertisement information storage unit. 125 and.

(User information storage unit 121)
The user information storage unit 121 according to the embodiment stores various information regarding user attributes. For example, the user information storage unit 121 stores user attribute information. FIG. 3 is a diagram illustrating an example of the user information storage unit according to the embodiment. The user information storage unit 121 shown in FIG. 3 includes items such as “user ID”, “group (whether or not a company A personal computer is purchased)”, “age”, “sex”, “place of residence”, and “interest”.

The “user ID” indicates identification information for identifying the user. For example, the user identified by the user ID “U1” corresponds to the user U1 shown in the example of FIG. “Group (Purchase of PC of A company)” indicates a group to which the corresponding user belongs according to whether or not the PC of A company is purchased. The "age" indicates the age of the user identified by the user ID. The “age” may be a specific age of the user identified by the user ID, such as 35 years old. Further, "sex" indicates the sex of the user identified by the user ID. In addition, "residential location" indicates an area where the user identified by the user ID resides. Note that the "residential place" may be a specific address of the user identified by the user ID. Further, "interest" indicates a target of interest to the user identified by the user ID. A plurality of “interests” may be registered.

For example, in the example shown in FIG. 3, the user identified by the user ID “U1” is a user belonging to the group A and a user who purchased the personal computer of company A. The age of the user identified by the user ID “U1” is “30s” and the gender is “male”. Further, for example, in the example illustrated in FIG. 3, the user identified by the user ID “U1” indicates that the place of residence is “region A” and the target of interest is “sports”.

The user information storage unit 121 is not limited to the above, and may store various information according to the purpose. For example, the user information storage unit 121 may store information regarding the demographic attribute of the user and information regarding the psychographic attribute. For example, the user information storage unit 121 may store information such as name, family structure, income, and lifestyle.

(Search history information storage unit 122)
The search history information storage unit 122 according to the embodiment stores various kinds of information regarding searches. For example, in FIG. 4, the search history information storage unit 122 stores information about queries used in a predetermined search system. FIG. 4 shows an example of the search history information storage unit 122 according to the embodiment. The search history information storage unit 122 illustrated in FIG. 4 has items such as “search query ID”, “search query”, “input user”, and “date and time”. The "search query" includes items such as "token 1" and "token 2".

The “search query ID” indicates identification information for identifying the query. The "token 1" and "token 2" of the "search query" indicate the tokens included in the corresponding query. “Input user” indicates a user who has input (used) the corresponding query. Further, “date and time” indicates the date and time when the corresponding query was input (used).

In the example of FIG. 4, it is shown that the query identified by the search query ID “Q11” includes a token such as “personal computer” or “company B model number X”. It also indicates that the user who used the query identified by the search query ID “Q11” is the user U1. It also indicates that the search using the query identified by the search query ID “Q11” was performed on the date DA. In the example of FIG. 4, the date and time is represented by an abstract symbol such as the date and time DA, but a specific date and time such as “July 27, 2016, 21:28:54” may be used. Good.

The search history information storage unit 122 is not limited to the above, and may store various information according to the purpose.

(Vector information storage unit 123)
The vector information storage unit 123 according to the embodiment stores vector information of each word (token). FIG. 5 is a diagram illustrating an example of the vector information storage unit according to the embodiment. As shown in FIG. 5, the vector information storage unit 123 stores information about each word. Specifically, the vector information storage unit 123 stores information about the feature weight of each word learned by the distributed expression. As shown in FIG. 5, the vector information storage unit 123 has items such as “word ID”, “word (token)”, and “feature 1” to “feature 3” as vector information.

The “word ID” indicates identification information for identifying a word (token). "Word (token)" indicates a corresponding word (token). "Feature 1" to "feature 3" indicate the weight of each feature of the corresponding word (token).

For example, in the example shown in FIG. 5, the token “PC” identified by the word ID “W1” has a feature 1 weight of “1.5”, a feature 2 weight of “0”, and a feature 3 weight of “0”. 0.8" or the like. For example, when each word (token) is represented by an m-dimensional vector, the number of features is m, and the weights of the features 1 to m are stored.

The vector information storage unit 123 is not limited to the above, and may store various kinds of information according to the purpose, and any information may be stored as long as it is a vector expression of each word (token). ..

(Feature cluster information storage unit 124)
The feature cluster information storage unit 124 according to the embodiment stores various types of information regarding feature clusters. For example, in FIG. 6, the feature cluster information storage unit 124 stores information about feature clusters. FIG. 6 shows an example of the feature cluster information storage unit 124 according to the embodiment. The characteristic cluster information storage unit 124 illustrated in FIG. 6 has items such as “cluster ID”, “token list”, and “evaluation value (group A)”. The token list includes items such as "token 1" to "token 4".

“Cluster ID” indicates identification information for identifying a cluster. “Token 1” to “token 4” and the like in the “token list” indicate tokens belonging to the corresponding cluster. The “evaluation value (group A)” indicates the evaluation of the corresponding cluster.

The example of FIG. 6 shows that the cluster C12 identified by the cluster ID “C12” is extracted as a feature cluster. Further, it is indicated that the cluster C12 includes tokens such as "B company model number X", "B company model number Y", "ZA series", and "ZB series". The evaluation value of the group A of the cluster C12 is “12”.

The feature cluster information storage unit 124 is not limited to the above, and may store various information according to the purpose. For example, the feature cluster information storage unit 124 may store information regarding the date and time when the feature cluster was extracted.

(Advertisement information storage unit 125)
The advertisement information storage unit 125 according to the embodiment stores various information regarding advertisements. FIG. 7 shows an example of the advertisement information storage unit 125 according to the embodiment. The advertisement information storage unit 125 illustrated in FIG. 7 has items such as “advertisement ID”, “advertisement content”, and “product”.

The “advertisement ID” indicates identification information for identifying the advertisement. The “advertisement content” indicates an advertisement submitted by an advertiser, an agency, or the like. FIG. 7 shows an example in which conceptual information such as “advertisement A” is stored in “advertising content”, but actually, still images, moving images, text data, URLs, or storage locations of these are shown. File path name etc. are stored. In the following, the notation “advertiser” is a concept that includes not only the advertiser but also the agency. “Products and the like” indicate products and services targeted for advertisement. The information for identifying the product or service is not limited to the "product or the like", and information such as a "product ID" for identifying the product or the service may be stored.

For example, in the example illustrated in FIG. 7, the advertisement A identified by the advertisement ID “AC1” indicates that the advertisement is for the product “Company A PC XA”.

The advertisement information storage unit 125 is not limited to the above, and may store various information according to the purpose. For example, the advertisement information storage unit 125 may store information regarding a bid price and the like.

(Control unit 130)
Returning to the description of FIG. 2, the control unit 130 is a controller, and is stored in a storage device inside the extraction device 100 by, for example, a CPU (Central Processing Unit) or an MPU (Micro Processing Unit). It is realized by executing various programs (corresponding to an example of the extraction program) using the RAM as a work area. The control unit 130 is a controller, and is realized by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

As illustrated in FIG. 2, the control unit 130 includes an acquisition unit 131, a classification unit 132, an extraction unit 133, and a distribution unit 134, and realizes or executes the functions and actions of information processing described below. .. The internal configuration of the control unit 130 is not limited to the configuration shown in FIG. 2 and may be another configuration as long as it is a configuration for performing information processing described later.

(Acquisition unit 131)
For example, the acquisition unit 131 acquires various kinds of information. For example, the acquisition unit 131 acquires information about the user from the user information storage unit 121. Further, for example, the acquisition unit 131 acquires information regarding the search history from the search history information storage unit 122. Further, for example, the acquisition unit 131 acquires vector information of each word (token) from the vector information storage unit 123. Further, for example, the acquisition unit 131 acquires information about the feature cluster from the feature cluster information storage unit 124. Further, for example, the acquisition unit 131 acquires information regarding an advertisement from the advertisement information storage unit 125. For example, the acquisition unit 131 acquires information regarding the search query used by the target user who has performed the predetermined action and information regarding the search query used by the other user who has not performed the predetermined action. For example, the acquisition unit 131 acquires information regarding a search query used by another user having an attribute similar to the attribute of the target user. For example, the acquisition unit 131 acquires the token included in the search query used by the target user as the information related to the search query used by the target user, and the token used by the other user is used as the information related to the search query used by the other user. Get the token included in the search query. For example, the acquisition unit 131 acquires information about a search query used by a target user who purchased a predetermined product and information about a search query used by another user who did not purchase the predetermined product.

For example, the acquisition unit 131 acquires the action information of the target user who has performed the predetermined action and the action information of the other user who has not performed the predetermined action. For example, the acquisition unit 131 acquires the behavior information of the target user who purchased the predetermined product and the behavior information of another user who did not purchase the predetermined product. For example, the acquisition unit 131 acquires information regarding the search query used by the target user as the behavior information of the target user, and acquires information regarding the search query used by the other user as the behavior information of the other user. For example, the acquisition unit 131 acquires the token included in the search query used by the target user as the behavior information of the target user, and the token included in the search query used by the other user as the behavior information of the other user. get.

For example, the acquisition unit 131 acquires information regarding the query used by the target user for the search from the search history information storage unit 122. For example, the acquisition unit 131 acquires, from the search history information storage unit 122, information about the token included in the query used by the target user for the search. In the example of FIG. 1, for example, the acquisition unit 131 acquires the usage token information TK1 of the user U1 who is the target user from the search history information storage unit 122.

For example, the acquisition unit 131 acquires the behavior information of another user having an attribute similar to the attribute of the target user. For example, the acquisition unit 131 acquires, from the search history information storage unit 122, information about the token included in the query used by the users other than the target user for the search. In the example of FIG. 1, for example, the acquisition unit 131 acquires the usage token information TK3 of the user U3 who is a user other than the target user from the search history information storage unit 122. For example, the acquisition unit 131 acquires the use token information TK3 of the user U3 who is a user in his thirties who is similar to the attribute of the user U1 who is the target user. The acquisition unit 131 may acquire the vector information of each token from an external processing device.

(Classification unit 132)
The classification unit 132 classifies various information. The classifying unit 132 classifies the behavior information of the target user into the first cluster based on the similarity between the behavior information of the target user, and the behavior of the other user based on the similarity between the behavior information of other users. Classify the information into the second cluster. For example, the classification unit 132 classifies the behavior information of the target user into the first cluster based on the degree of similarity between the vectors corresponding to the behavior information of the target user, and divides between the vectors corresponding to the behavior information of other users. The behavior information of other users is classified into the second cluster based on the similarity. For example, the classification unit 132 classifies the tokens used by the target user into the first cluster based on the similarity between the tokens used by the target user, and based on the similarity between the tokens used by other users. Tokens used by other users are classified into the second cluster. For example, the classification unit 132 classifies the information regarding the search query used by the target user into the first cluster based on the similarity between the vectors corresponding to the information regarding each search query used by the target user, and is used by other users. The information about the search query used by another user is classified into the second cluster based on the similarity between the vectors corresponding to the information about each search query.

For example, the classification unit 132 generates information about the similarity of words (tokens). For example, the classification unit 132 learns the distributed expression based on the information about the token stored in the search history information storage unit 122, and generates the feature weight of each word (token) as vector information. For example, the classification unit 132 may perform learning of a distributed expression based on the technique related to “word2vec” in Non-Patent Document 1. The classifying unit 132 derives a word vector (for example, referred to as “word embedding” or the like) that is an m-dimensional real-valued vector for each word (token). For example, the real value of each dimension corresponds to the weight of each feature in the vector information. In this way, the classification unit 132 generates information about the vector of tokens by learning the distributed expression. When the extraction device 100 acquires the vector information of each token from an external processing device, the classification unit 132 does not have to generate the vector information.

In the example of FIG. 1, the classification unit 132 classifies the tokens used by the users included in the group A into clusters. Specifically, the classification unit 132 classifies the tokens used by the users included in the group A into the first cluster based on the similarity of the tokens. The classification unit 132 classifies the tokens used by the users included in the group A into the first cluster based on the similarity of the vector corresponding to each token.

For example, the classifying unit 132 classifies the tokens used by the users included in the group A into the first cluster based on the similarity of the vector of each token derived based on the technique regarding “word2vec” in Non-Patent Document 1. To do. Further, for example, the classifying unit 132 classifies the tokens used by the users included in the group A into the first cluster by appropriately using various conventional techniques such as the vector of each token described above and Kmeans. For example, the classification unit 132 classifies the tokens used by the users included in the group A into the first cluster so that the number of clusters becomes a predetermined number (for example, 1000).

In the example of FIG. 1, the classification unit 132 classifies the tokens used by the users included in the group A into the first cluster, as shown in the first cluster group CL11. For example, the classification unit 132 classifies tokens such as “personal computer”, “notebook PC”, “PC”, and “no-paso” as the cluster C11, as shown in the first cluster group CL11. In addition, for example, the classification unit 132 classifies tokens such as “company B model number X”, “company B model number Y”, “ZA series”, and “ZB series” as cluster C12 as shown in the first cluster group CL11. To do.

(Extractor 133)
The extraction unit 133 extracts various information. For example, the extraction unit 133 extracts, from the first cluster, a characteristic cluster that is a cluster that shows a characteristic behavior of the target user, based on the difference between the first cluster and the second cluster. For example, the extraction unit 133 sets the ratio of the users who have acted on the behavior information belonging to the first cluster among the target users and the ratio of the users who have acted on the behavior information belonging to the second cluster among other users. When the ratio is equal to or larger than the predetermined threshold value, the first cluster is extracted as the feature cluster. For example, the extraction unit 133, among the target users, the ratio of the users who searched using the tokens belonging to the first cluster, and the ratio of the users who searched using the tokens belonging to the second cluster among other users. When the ratio of and is greater than or equal to a predetermined threshold value, the first cluster is extracted as a feature cluster. For example, the extraction unit 133 uses the ratio of users who have performed a search using the information regarding the search query belonging to the first cluster among the target users and the search using the information regarding the search query belonging to the second cluster among the other users. When the ratio to the ratio of users who have performed the above is greater than or equal to a predetermined threshold value, the first cluster is extracted as a feature cluster.

In the example of FIG. 1, the extraction unit 133 extracts the feature cluster based on the information regarding the usage rate of the tokens belonging to each cluster shown in the first cluster group CL11 and the second cluster group CL12. For example, the extraction unit 133 extracts, as a feature cluster, a cluster whose evaluation value based on the difference between the usage rate in the group A and the usage rate in the group B is equal to or greater than a predetermined threshold. For example, the extraction unit 133 extracts, as a characteristic cluster, a cluster in which the ratio (evaluation value) of the usage rate in the group A to the usage rate in the group B is 3 or more. Further, for example, with respect to the cluster C12, the extraction unit 133 has a usage rate of 60% in the group A and a usage rate of 5% in the group B, and thus the evaluation value is “12 (=60/5)”. Therefore, the cluster C12 is extracted as a feature cluster. For example, the extraction unit 133 extracts a cluster characteristic of the user who purchased the personal computer of the company A as a characteristic cluster. In the example of FIG. 1, the extraction unit 133 extracts, as the characteristic cluster information storage unit 124, a cluster having an evaluation value of 3 or more, such as the cluster C12 or the cluster C21, as a characteristic cluster.

The extraction unit 133 also extracts an advertisement from the advertisement information storage unit 125. For example, when the extraction unit 133 acquires the query input by the user who is performing the search by the acquisition unit 131, the token in the query matches the token of the feature cluster stored in the feature cluster information storage unit 124. , Extract advertisements corresponding to feature clusters. Further, for example, when the acquisition unit 131 acquires the query input by the user who is performing the search, the extraction unit 133 matches the token in the query with the token of the feature cluster stored in the feature cluster information storage unit 124. If not, the advertisement is extracted based on a predetermined standard. For example, when the token in the query does not match the token of the feature cluster stored in the feature cluster information storage unit 124, the extraction unit 133 may extract the advertisement based on the bid price of each advertisement or information about the user.

(Distribution unit 134)
The distribution unit 134 transmits various types of information to an external information processing device. For example, the delivery unit 134 delivers the content related to the predetermined action to the user who has performed the action corresponding to the action information included in the feature cluster extracted by the extraction unit 133. For example, the delivery unit 134 delivers an advertisement regarding a predetermined action to the user who has performed the action corresponding to the action information included in the feature cluster extracted by the extraction unit 133. In the example of FIG. 1, for example, the distribution unit 134 distributes the advertisement of the personal computer of the company A to the user who searched using the token included in the feature cluster information storage unit 124 extracted by the extraction unit 133. Details of this point will be described later. For example, the distribution unit 134 distributes the advertisement extracted by the extraction unit 133. For example, the distribution unit 134 distributes the content regarding the predetermined action to the user who has performed the search using the information regarding the search query included in the feature cluster extracted by the extraction unit 133. For example, the delivery unit 134 delivers an advertisement regarding a predetermined action to a user who has performed a search using the information regarding the search query included in the feature cluster extracted by the extraction unit 133.

[3. Extraction process flow]
Next, the procedure of the extraction processing by the extraction device 100 according to the embodiment will be described with reference to FIG. FIG. 8 is a flowchart showing an extraction processing procedure according to the embodiment.

As illustrated in FIG. 8, the extraction device 100 acquires information regarding the search query used by the user for the search (step S101). For example, the extraction device 100 acquires the token included in the search query used by the target user as the behavior information of the target user, and the token included in the search query used by the other user as the behavior information of the other user. get. Then, for example, the extraction device 100 performs distributed expression learning or the like on the token acquired by the acquisition unit 131, and generates information about the vector of each token. When the vector information of each token is acquired from the external processing device, the extraction device 100 does not have to generate the vector information.

Then, the extraction device 100 classifies the tokens of group A (target user group) into the first cluster (step S102). In the example of FIG. 1, the extraction device 100 classifies the tokens used by the users included in the group A into the first cluster, as shown in the first cluster group CL11.

Further, the extraction device 100 classifies the tokens of the group B (non-target user group) into the second cluster (step S103). In the example of FIG. 1, the extraction device 100 classifies the tokens used by the users included in the group B into the second cluster, as shown in the second cluster group CL12.

Then, the extraction device 100 extracts a feature cluster based on the first cluster and the second cluster (step S104). In the example of FIG. 1, the extraction device 100 extracts, as the characteristic cluster information storage unit 124, a cluster having an evaluation value of 3 or more, such as the cluster C12 or the cluster C21, as a characteristic cluster.

[4. Advertising distribution processing]
Next, the advertisement distribution process by the extraction device 100 according to the embodiment will be described with reference to FIG. 9. FIG. 9 is a diagram illustrating an example of the advertisement distribution process according to the embodiment. The example of FIG. 9 shows a case where an advertisement based on a query input by the user U11 when performing a search is delivered.

First, the extraction device 100 acquires a query from the terminal device 10 used by the user U11 (step S21). In the example of FIG. 9, the extraction device 100 acquires the query “ZA series” from the terminal device 10. In the example illustrated in FIG. 9, the user U11 inputs the query “ZA series” in the search window in the page W11 that displays the search screen displayed on the screen of the terminal device 10, and presses the search button. Then, the terminal device 10 transmits the search query “ZA series” to the extraction device 100.

Then, the extraction device 100 extracts the advertisement based on the acquired search query “ZA series” (step S22). For example, the extraction device 100 extracts an advertisement depending on whether the token “ZA series” included in the acquired search query “ZA series” matches the token in the feature cluster information storage unit 124. In the example of FIG. 9, the extraction device 100 extracts the advertisement A regarding the personal computer XA of the company A because the token “ZA series” matches the token in the feature cluster information storage unit 124, as shown in the advertisement list AL11. ..

After that, the extraction device 100 delivers the extracted advertisement to the terminal device 10 used by the user U11 (step S23). In the example of FIG. 9, the extraction device 100 delivers the advertisement A regarding the personal computer XA of the company A to the terminal device 10 used by the user U11 (step S23).

As described above, the extraction device 100 can improve the advertising effect by distributing the advertisement regarding the personal computer of the company A to the user who has searched using the token characteristic of the target user. .. For example, when a token that is characteristic of a user who has purchased a personal computer of company A is used, the extraction device 100 distributes an advertisement relating to the personal computer of company A to the user who performed the search, thereby advertising. The effect can be improved.

[5. Flow of advertisement distribution processing]
Next, with reference to FIG. 10, a procedure of advertisement distribution processing by the extraction device 100 according to the embodiment will be described. FIG. 10 is a flowchart showing an advertisement distribution processing procedure according to the embodiment.

As illustrated in FIG. 10, the extraction device 100 acquires a search query (step S201). Then, when the token in the search query matches the token of the feature cluster (step S202: Yes), the extraction device 100 extracts the advertisement corresponding to the feature cluster (step S203).

On the other hand, when the token in the search query does not match the token of the feature cluster (step S202: No), the extraction device 100 extracts the advertisement based on a predetermined criterion (step S204). Then, the extraction device 100 delivers the extracted advertisement (step S205).

[6. effect〕
As described above, the extraction device 100 according to the embodiment has the acquisition unit 131, the classification unit 132, and the extraction unit 133. The acquisition unit 131 acquires information about a search query used by a target user who has performed a predetermined action and information about a search query used by another user who has not performed a predetermined action. The classifying unit 132 classifies the information about the search query used by the target user into the first cluster based on the similarity between the vectors corresponding to the information about each search query used by the target user, and uses each of the information used by other users. Information regarding a search query used by another user is classified into the second cluster based on the similarity between vectors corresponding to the information regarding a search query. The extraction unit 133 extracts, from the first cluster, a characteristic cluster that is a cluster that shows a characteristic behavior of the target user, based on the difference between the first cluster and the second cluster.

Thereby, the extraction device 100 according to the embodiment extracts a characteristic cluster, which is a cluster showing a characteristic behavior of the target user, from the first cluster based on the difference between the first cluster and the second cluster. Information characteristic of a particular user can be extracted.

Further, in the extraction device 100 according to the embodiment, the extraction unit 133 sets the ratio of the users who have performed the search using the information regarding the search query belonging to the first cluster among the target users and the second cluster among the other users. When the ratio to the ratio of users who have performed a search using the information about the belonging search query is equal to or greater than a predetermined threshold value, the first cluster is extracted as a feature cluster.

As a result, the extraction device 100 according to the embodiment relates to the ratio of users who have performed a search using the information about the search query belonging to the first cluster among the target users and the search query belonging to the second cluster among other users. When the ratio to the ratio of users who have searched using information is equal to or greater than a predetermined threshold value, the characteristic information of a specific user can be extracted by extracting the first cluster as a characteristic cluster.

Further, in the extraction device 100 according to the embodiment, the acquisition unit 131 acquires information regarding a search query used by another user having an attribute similar to the attribute of the target user.

Thereby, the extraction device 100 according to the embodiment can extract characteristic information of a specific user based on the difference from the user having an attribute similar to the attribute of the target user.

In addition, the extraction device 100 according to the embodiment has a distribution unit 134. The distribution unit 134 distributes the content regarding the predetermined action to the user who has performed the search using the information regarding the search query included in the feature cluster extracted by the extraction unit 133.

As a result, the extraction device 100 according to the embodiment distributes the content related to the predetermined action to the user who has performed the search using the information related to the search query included in the extracted feature cluster, and thereby the characteristic of the specific user is characteristic. Appropriate content can be delivered based on the information.

Further, in the extraction device 100 according to the embodiment, the distribution unit 134 distributes an advertisement regarding a predetermined action to a user who has performed a search using information about a search query included in the feature cluster extracted by the extraction unit 133.

As a result, the extraction device 100 according to the embodiment distributes an advertisement regarding a predetermined action to a user who has performed a search using the information regarding the search query included in the extracted feature cluster, and thus is characteristic for a specific user. Appropriate advertisements can be delivered based on the information.

Further, in the extraction device 100 according to the embodiment, the acquisition unit 131 acquires the token included in the search query used by the target user as the information regarding the search query used by the target user, and the search query used by another user. As information regarding the above, a token included in a search query used by another user is acquired. The classifying unit 132 classifies the tokens used by the target user into the first cluster based on the similarity between the tokens used by the target user, and determines other tokens based on the similarity between the tokens used by other users. The tokens used by the user are classified into the second cluster.

Thereby, the extraction device 100 according to the embodiment uses the token included in the search query used by the target user as the information regarding the search query used by the target user, and the information regarding the search query used by another user as By using the token included in the search query used by the user, it is possible to extract the token characteristic of the specific user.

In addition, in the extraction device 100 according to the embodiment, the extraction unit 133 determines the ratio of users who have searched using the tokens belonging to the first cluster among the target users and the tokens belonging to the second cluster among other users. If the ratio of the used search to the ratio of users is equal to or greater than a predetermined threshold, the first cluster is extracted as a feature cluster.

As a result, the extraction device 100 according to the embodiment performs a search using the token belonging to the second cluster among other users and the ratio of the users who performed the search using the token belonging to the first cluster among the target users. When the ratio to the ratio of users who have performed is a predetermined threshold value or more, by extracting the first cluster as a feature cluster, it is possible to extract a token characteristic of a specific user.

Further, in the extraction device 100 according to the embodiment, the acquisition unit 131 relates to information about a search query used by a target user who purchased a predetermined product and a search query used by another user who did not purchase the predetermined product. Get information and.

Thereby, the extraction device 100 according to the embodiment uses the information about the search query used by the target user who purchased the predetermined product and the information about the search query used by the other user who did not purchase the predetermined product. This makes it possible to extract information characteristic of the user who purchased the predetermined product.

[7. Hardware configuration]
The extraction device 100 according to the above-described embodiment is realized by, for example, a computer 1000 configured as shown in FIG. FIG. 11 is a hardware configuration diagram illustrating an example of a computer that realizes the function of the extraction device. The computer 1000 has a CPU 1100, a RAM 1200, a ROM 1300, an HDD 1400, a communication interface (I/F) 1500, an input/output interface (I/F) 1600, and a media interface (I/F) 1700.

The CPU 1100 operates based on a program stored in the ROM 1300 or the HDD 1400, and controls each part. The ROM 1300 stores a boot program executed by the CPU 1100 when the computer 1000 starts up, a program dependent 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. The communication interface 1500 receives data from another device via the predetermined network N, sends the data to the CPU 1100, and transmits the data generated by the CPU 1100 to the other device via the predetermined network N.

The CPU 1100 controls output devices such as a display and a printer and input devices such as a keyboard and a mouse via the input/output interface 1600. The CPU 1100 acquires data from the input device via the input/output interface 1600. Further, the CPU 1100 outputs the generated data to the output device via the input/output interface 1600.

The media interface 1700 reads a program or data stored in the recording medium 1800 and provides the program or data to the CPU 1100 via the RAM 1200. The CPU 1100 loads the program from the recording medium 1800 onto the RAM 1200 via the 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) and a PD (Phase change rewritable Disk), 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 extraction device 100 according to the embodiment, the CPU 1100 of the computer 1000 realizes the function of the control unit 130 by executing the program loaded on the RAM 1200. The CPU 1100 of the computer 1000 reads these programs from the recording medium 1800 and executes them, but as another example, these programs may be acquired from another device via a predetermined network.

As described above, some of the embodiments of the present application have been described in detail based on the drawings, but these are examples, and various modifications based on the knowledge of those skilled in the art, starting from the aspect described in the row of the disclosure of the invention, It is possible to implement the present invention in other forms with improvements.

[8. Other]
Further, of the processes described in the above embodiment, all or part of the processes described as being automatically performed may be manually performed, or the processes described as manually performed may be performed. All or part of the process can be automatically performed by a known method. In addition, the processing procedures, specific names, information including various data and parameters shown in the above-mentioned 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. Further, each component of each device shown in the drawings is functionally conceptual and does not necessarily have to be physically configured as shown. That is, the specific form of distribution/integration of each device is not limited to that shown in the figure, and all or part of the device may be functionally or physically distributed/arranged in arbitrary units according to various loads and usage conditions. It can be integrated and configured.

In addition, the above-described embodiments can be appropriately combined within the range in which the processing content is not inconsistent.

Further, the "section (module, unit)" described above can be read as "means" or "circuit". For example, the acquisition unit can be read as an acquisition unit or an acquisition circuit.

100 extraction device 121 user information storage unit 122 search history information storage unit 123 vector information storage unit 124 feature cluster information storage unit 125 advertisement information storage unit 130 control unit 131 acquisition unit 132 classification unit 133 extraction unit 134 distribution unit

Claims (10)

Information regarding the search query used by the target user who has performed a predetermined action among the user group stored in the user information storage unit and used by another user who has not performed the predetermined action among the user group An acquisition unit that acquires information about the search query from the search history information storage unit ,
The information about the search query used by the target user is classified into the first cluster based on the similarity between the vectors corresponding to the information about the search query used by the target user, and the search queries used by the other users are classified. A classifying unit that classifies information about a search query used by the other user into a second cluster based on the similarity between vectors corresponding to the information about
An extraction unit for extracting a characteristic cluster, which is a cluster showing a characteristic behavior of the target user, from the first cluster based on the difference between the first cluster and the second cluster;
An extraction device comprising: The extraction unit is
Among the target users, a ratio of users who performed a search using information about a search query belonging to the first cluster, and a search using information about a search query belonging to the second cluster among the other users was performed. The extraction device according to claim 1, wherein the first cluster is extracted as the feature cluster when the ratio to the user ratio is equal to or greater than a predetermined threshold value. The acquisition unit is
The extraction device according to claim 1 or 2, wherein information regarding a search query used by the other user having an attribute similar to the attribute of the target user is acquired. A distribution unit that distributes the content related to the predetermined action to the user who performed the search using the information related to the search query included in the feature cluster extracted by the extraction unit,
The extraction device according to claim 1, further comprising: The delivery unit is
The extraction device according to claim 4, wherein an advertisement regarding the predetermined action is delivered to a user who has performed a search using information regarding a search query included in the feature cluster extracted by the extraction unit. The acquisition unit is
As information on the search query used by the target user, a token included in the search query used by the target user is acquired, and as the information on the search query used by the other user, the search query used by the other user Get the token contained in
The classification unit is
The token used by the target user is classified into a first cluster based on the similarity between the tokens used by the target user, and the other user is based on the similarity between the tokens used by the other user. The extraction device according to any one of claims 1 to 5, characterized in that the token used by is classified into a second cluster. The extraction unit is
Ratio of the ratio of the users who searched using the tokens belonging to the first cluster to the target users and the ratio of the users who searched using the tokens belonging to the second cluster among the other users Is equal to or larger than a predetermined threshold, the first cluster is extracted as the feature cluster. The acquisition unit is
The information about the search query used by the target user who purchased a predetermined product and the information about the search query used by the other user who did not purchase the predetermined product are acquired. The extraction device according to any one of 1 to 7. A computer-implemented extraction method,
Information regarding the search query used by the target user who has performed a predetermined action among the user group stored in the user information storage unit and used by another user who has not performed the predetermined action among the user group An acquisition step of acquiring information about the search query from the search history information storage unit ,
The information about the search query used by the target user is classified into the first cluster based on the similarity between the vectors corresponding to the information about the search query used by the target user, and the search queries used by the other users are classified. A step of classifying information about a search query used by the other user into a second cluster based on a similarity between vectors corresponding to the information about
An extraction step of extracting, from the first cluster, a characteristic cluster that is a cluster showing a characteristic behavior of the target user, based on a difference between the first cluster and the second cluster;
An extraction method comprising: Information regarding the search query used by the target user who has performed a predetermined action among the user group stored in the user information storage unit and used by another user who has not performed the predetermined action among the user group An acquisition procedure for acquiring information about the search query from the search history information storage unit ,
The information about the search query used by the target user is classified into the first cluster based on the similarity between the vectors corresponding to the information about the search query used by the target user, and the search queries used by the other users are classified. A classification procedure for classifying information on a search query used by the other user into a second cluster based on the similarity between vectors corresponding to the information on
An extraction procedure for extracting a characteristic cluster, which is a cluster showing a characteristic behavior of the target user, from the first cluster based on the difference between the first cluster and the second cluster;
An extraction program characterized by causing a computer to execute.

Images