JP6679683B2 - Information processing apparatus, 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, a technique of extracting information characteristic of a specific user is known. Specifically, 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 each search query used by the target user, and the search queries of other users are classified. Information regarding a search query used by another user is classified into the second cluster based on the degree of similarity between vectors corresponding to the information regarding. Then, based on the difference between the first cluster and the second cluster, a technique has been proposed for extracting a characteristic cluster, which is a cluster showing a characteristic behavior of the target user, from the first cluster.

Japanese Unexamined Patent Publication No. 2018-60469

  However, the above-mentioned conventional technique cannot always improve the accuracy of analysis of the user's search trend. Specifically, the meaning of a word may depend on the context in which it is used. Therefore, even if the string used in the context is separated from the context and extracted, it is difficult to interpret the original meaning of the string (that is, the meaning that the string was used in the context). There is. In view of the above, in the above-mentioned conventional technique, it is difficult to interpret the original meaning of the search query because the similarity of the search query as a character string is only considered. Therefore, in the above-mentioned conventional technique, it is difficult to analyze the search trend of the user in consideration of the context of the user who inputs the search query. Therefore, the above-mentioned conventional technique cannot always improve the accuracy of analysis of the search trend of the user.

  The present application has been made in view of the above, and an object thereof is to provide an information processing apparatus, an information processing method, and an information processing program that can improve the accuracy of analysis of a user's search trend.

  The information processing apparatus according to the present application uses the first learning model in which the features of the plurality of search queries are learned, assuming that the plurality of search queries input by the same user within a predetermined time have similar features. And a service providing unit that extracts a similar query that is a search query having characteristics similar to a predetermined search query.

  According to the aspect of the embodiment, it is possible to improve the accuracy of analysis of the search trend of the user.

FIG. 1 is a diagram illustrating an example of information processing according to the embodiment. FIG. 2 is a diagram illustrating an example of a generation process of the first learning model according to the embodiment. FIG. 3 is a diagram illustrating an example of a generation process of the first learning model according to the embodiment. FIG. 4 is a diagram illustrating an example of the second learning model generation process according to the embodiment. FIG. 5 is a diagram illustrating a configuration example of the information processing system according to the embodiment. FIG. 6 is a diagram illustrating a configuration example of the information processing device according to the embodiment. FIG. 7 is a diagram illustrating an example of the query information storage unit according to the embodiment. FIG. 8 is a diagram illustrating an example of the vector information storage unit according to the embodiment. FIG. 9 is a diagram illustrating an example of the classification definition storage unit according to the embodiment. FIG. 10 is a diagram illustrating an example of the category information storage unit according to the embodiment. FIG. 11 is a diagram illustrating an example of the model information storage unit according to the embodiment. FIG. 12 is a diagram illustrating an example of the first learning model according to the embodiment. FIG. 13 is a diagram illustrating an example of the second learning model according to the embodiment. FIG. 14 is a flowchart showing a procedure for generating the first learning model according to the embodiment. FIG. 15 is a flowchart showing the procedure for generating the second learning model according to the embodiment. FIG. 16 is a flowchart showing the information processing procedure according to the embodiment. FIG. 17 is a hardware configuration diagram illustrating an example of a computer that realizes the functions of the information processing device.

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

[1. Example of information processing]
First, an example of information processing according to the embodiment will be described with reference to FIG. FIG. 1 is a diagram illustrating an example of information processing according to the embodiment. The information processing shown in FIG. 1 is performed by the user terminal 10, the client terminal 20, the search server 50, and the information processing apparatus 100.

[Configuration of information processing system]
Prior to the description of FIG. 1, the configuration of the information processing system 1 will be described with reference to FIG. FIG. 5 is a diagram illustrating a configuration example of the information processing system according to the embodiment. As shown in FIG. 5, the information processing system 1 includes a user terminal 10, a client terminal 20, a search server 50, and an information processing device 100. The user terminal 10, the client terminal 20, the search server 50, and the information processing apparatus 100 are communicably connected to each other via a predetermined network N by wire or wirelessly. Note that the information processing system 1 illustrated in FIG. 5 includes an arbitrary number of user terminals 10, an arbitrary number of client terminals 20, an arbitrary number of search servers 50, and an arbitrary number of information processing apparatuses 100. Good.

  The user terminal 10 is an information processing device used by a user who uses the search service. The user terminal 10 is realized by, for example, a smartphone, a tablet terminal, a notebook PC (Personal Computer), a mobile phone, a PDA (Personal Digital Assistant), or the like. In the following, the user terminal 10 may be identified with the user. That is, in the following, the user can be read as the user terminal 10.

  In the following, the user identified by the user ID “U1” may be referred to as “user U1”. Thus, in the following, when "user U * (* is an arbitrary numerical value)" is described, it means that the user is the user specified by the user ID "U *". For example, when described as “user U2”, the user is the user specified by the user ID “U2”.

  Moreover, below, the user terminal 10 is demonstrated as the user terminals 10-1 and 10-2 according to the user who uses the user terminal 10. For example, the user terminal 10-1 is the user terminal 10 used by the user U1. Further, for example, the user terminal 10-2 is the user terminal 10 used by the user U2. Further, in the following, the user terminals 10-1 and 10-2 will be referred to as the user terminal 10 in the case of being described without particular distinction.

  The user terminal 10 transmits the search query input by the user to the search server 50. Specifically, the user terminal 10 acquires a search page including a search box for inputting a search query from the search server 50 according to an operation by the user. Subsequently, the user terminal 10, when an operation of transmitting a search query is performed following the operation of the user inputting a character in the search box, sets the character input in the search box as a search query via the search page. It transmits to the search server 50. For example, when the user presses the send button of the search query or the enter key is pressed after the user inputs characters in the search box, the user terminal 10 displays the search page via the search page. The characters entered in the search box are transmitted to the search server 50 as a search query.

  The client terminal 20 is an information processing device used by a client who is an advertiser. The client terminal 20 is realized by, for example, a smartphone, a tablet terminal, a notebook PC, a mobile phone, a PDA, or the like. In the following, the client terminal 20 may be identified with the client (advertiser). That is, in the following, the client (advertiser) can be read as the client terminal 20.

  The client terminal 20 transmits an analysis query, which is a keyword related to the analysis theme for analyzing the search trend of the user, to the information processing device 100. Specifically, the client terminal 20 transmits a plurality of analysis queries to the information processing device 100. For example, assume that a client who is an advertiser desires "bone health-related" as an analysis theme of a search query. Further, more specifically, it is assumed that the advertiser client intends “bone health”, “bone strength”, “calcium small fish”, etc. as keywords regarding the analysis theme. In this case, the client terminal 20 transmits a plurality of analysis queries such as “bone health”, “bone strength”, and “calcium small fish” to the information processing apparatus 100 according to the operation of the client who is the advertiser. In addition, the client terminal 20 transmits information about a classification axis (category) that classifies a similar query, which is a search query similar to the analysis query, to the information processing apparatus 100. For example, the client terminal 20 transmits the analysis query and the information about the classification axis (category) to the information processing apparatus 100 through the user interface for using the search query analysis service provided by the information processing apparatus 100.

  The search server 50 is a server device that provides a search service. Specifically, when the search server 50 receives the search query from the user terminal 10, the search server 50 selects the content that is output as the search result, which is the content according to the received search query. Then, the search server 50 delivers the search result page including the selected content to the user terminal 10. Here, the content distributed by the search server 50 is not limited to the web page displayed by the web browser. For example, the content distributed by the search server 50 may be content displayed by a dedicated application installed in the user terminal 10. The contents distributed by the search server 50 are music contents, images (including not only still images but also moving images) contents, and text contents (including news articles and articles posted on SNS (Social Networking Service)). , Any combination of images and text, game contents, and the like.

  The search server 50 also stores information about the search query input by the user. Specifically, the search server 50 stores information regarding the search history of the user. For example, when the search server 50 receives a search query from the user terminal 10, the search server 50 registers the received search query, the user ID that identifies the user who is the sender of the search query, and the transmission date and time of the search query in association with each other. . The search server 50 transmits information regarding the search query input by the user to the information processing apparatus 100 in response to a request from the information processing apparatus 100.

  The information processing device 100 is a server device that provides a search query analysis service to a client who is an advertiser. Specifically, the information processing apparatus 100 receives, from the client terminal 20, an analysis query that is a keyword related to an analysis theme that analyzes the search trend of the user. The information processing apparatus 100 may accept a plurality of analysis queries regarding analysis themes. For example, assume that a client who is an advertiser desires "bone health-related" as an analysis theme of a search query. Then, the information processing apparatus 100 receives a plurality of analysis queries such as “bone health”, “bone strength”, “calcium small fish” from the client terminal 20. As a result, the information processing apparatus 100 can extract more similar queries than when receiving one analysis query at a time. That is, the information processing apparatus 100 can extract all the similar queries according to the analysis theme of the client who is the advertiser. Then, the information processing apparatus 100 extracts the search query corresponding to the accepted analysis theme. For example, the information processing apparatus 100 extracts a similar query that is a search query having characteristics similar to the analysis query, using a first learning model described below. Here, the first learning model is learned by assuming that a plurality of search queries input by the same user within a predetermined time period have similar characteristics, so that a predetermined search query can be changed to a predetermined search query It is a learning model that predicts feature information. In the following, the first learning model will be appropriately referred to as the first model (or the first model M1).

  Further, the information processing apparatus 100 receives from the client terminal 20 a classification axis (category) that classifies a similar query that is a search query similar to the analysis query. The information processing apparatus 100 may accept a plurality of classification axes (categories) for one set of similar queries to be classified. Subsequently, the information processing apparatus 100 classifies the similar queries into categories corresponding to the accepted classification axis. For example, the information processing apparatus 100 classifies the similar queries into the category corresponding to the accepted classification axis using the second learning model described later. Here, the second learning model is a learning model generated using the first learning model, and is a learning model that predicts a category to which a predetermined search query belongs from a predetermined search query. In the following, the second learning model will be appropriately referred to as the second model (or the second model M2).

  In addition, the information processing device 100 generates a first learning model. Specifically, the information processing apparatus 100 learns the first learning model so that the distributed expressions of the plurality of search queries input by the same user within a predetermined time are similar to each other so that the predetermined search query A first learning model that predicts characteristic information of a predetermined search query is generated. Here, the information processing apparatus 100 learns the entire character string input in the search box each time the user makes a search as a search query input by the user. For example, when the search query including a plurality of character strings such as “Roppongi pasta” is input to the search box in one search by the user U1, the information processing apparatus 100 performs one search for the entire “Roppongi pasta”. Learn as a query. In addition, the information processing apparatus 100 may execute a plurality of search queries for which the same user has input each search query within a predetermined time (for example, within 2 minutes) at a predetermined time by the same user. Learn as multiple search queries entered in. That is, the information processing apparatus 100 learns that a plurality of search queries input by the same user in succession in a short time have similar characteristics. It should be noted that when a predetermined search query is input and a predetermined search time is exceeded (for example, 2 minutes or more) before another search query is input, the information processing apparatus 100 determines that the predetermined search query has been input. Learn as having different characteristics from other search queries.

  Generally, when a searcher performs a search, the searcher's intention is the result of multiple searches using different search queries, rather than the case where the searcher reaches the information intended by the searcher once. It is thought that there are more cases where information is reached. That is, since it is considered that the user performs the search a plurality of times with a certain intention, it is presumed that the search queries continuously input within a predetermined time have similar search intentions. Therefore, the information processing apparatus 100 according to the present invention trains the first learning model assuming that a plurality of search queries continuously input by the same user within a predetermined time have similar characteristics. Specifically, the information processing apparatus 100 acquires information regarding the search query input by the user from the search server 50. Then, the information processing apparatus 100 extracts a plurality of search queries input from the search server 50 by the same user within a predetermined time. Subsequently, the information processing apparatus 100 learns that the extracted plurality of search queries have similar features, thereby generating a first learning model that predicts the feature information of the predetermined search query from the predetermined search query. . For example, the information processing apparatus 100 trains the first learning model so that the extracted distributed expressions of the plurality of search queries are similar to each other, so that the distributed expression (the vector that includes the characteristic information of the predetermined search query from the predetermined search query) ) Is output to generate a first learning model.

  More specifically, the information processing apparatus 100 uses a technology of DSSM (Deep Structured Sematic Model) that uses LSTM (Long Short-Term Memory), which is a type of RNN (Recurrent Neural Network), for distributed representation generation. A first learning model that outputs a distributed expression (vector) is generated from the search query. For example, in the information processing apparatus 100, as the correct answer data of the first learning model, a pair of search queries input by the same user within a predetermined time have similar characteristics, and a distributed expression of the predetermined search query ( A vector) and a distributed expression (vector) of another search query that forms a pair with a predetermined search query are learned so that they exist near each other in the distributed expression space. Note that learning so that two vectors are close to each other in the distributed expression space can be rephrased as learning so that the two vectors are similar in the distributed expression space. As described above, since the information processing apparatus 100 generates the first learning model by using the DSSM technique in which the LSTM is used for the distributed expression generation, the first learning model generated by the information processing apparatus 100 is an arbitrary character. Distributed representations (vectors) can be generated from the columns. That is, the information processing apparatus 100 not only performs a distributed expression (vector) of a known character string (for example, a search query learned during learning) but also an unknown character string (for example, a search query that has not been learned during learning). Can also generate a distributed representation (vector).

  The information processing device 100 also generates a second learning model. Specifically, the information processing apparatus 100 uses the first learning model to generate a second learning model that predicts a category to which a predetermined search query belongs from a predetermined search query. More specifically, when the information processing apparatus 100 generates the first learning model, the information processing apparatus 100 acquires the generated first learning model (model data MDT1 of the first learning model M1). When the information processing apparatus 100 acquires the first model M1, it uses the acquired first model M1 to generate a second learning model M2. The information processing apparatus 100 re-learns the first model M1 to generate a second model M2 having a connection coefficient that is a weight of the learning model different from that of the first model M1. For example, when the search query is input to the learning model, the information processing apparatus 100 learns so that the classification result of the distributed expression output by the learning model corresponds to the category to which the search query belongs, thereby performing a predetermined search. A second model M2 that predicts a category to which a predetermined search query belongs is generated from the query.

  From here, the flow of information processing is demonstrated using FIG. FIG. 1 is a diagram illustrating an example of information processing according to the embodiment. In the example shown in FIG. 1, the analysis theme desired by the client who is the advertiser is “bone health related”. For example, a client who is an advertiser is a business operator who operates a health-related business. Further, the client as the advertiser desires to analyze the search trend of the user who is interested in "bone health" in developing the advertisement regarding the company's product. The information processing apparatus 100 receives, from the client terminal 20, the analysis query “bone health” and a classification axis (category) including four small classifications (step S1). Note that the information processing apparatus 100, instead of accepting the classification axis (category) specified by the client who is the advertiser, selects the client who is the advertiser from the categories shown in the classification definition storage unit 123 shown in FIG. The major classification selected by may be accepted as the classification axis. In the example illustrated in FIG. 1, the client who is the advertiser selects the large category “health system” identified by the large category ID “CAT2” from the categories shown in the category definition storage unit 123 described later. Then, the information processing apparatus 100 receives, as the classification axis, the large classification “health system” selected by the client who is the advertiser.

  Subsequently, when the information processing apparatus 100 receives the analysis query, the information processing apparatus 100 inputs the received analysis query “bone health” into the first model M1 (step S2). The information processing apparatus 100 preliminarily generates the first model M1 in which the features of the plurality of search queries are learned, assuming that the plurality of search queries input by the same user within a predetermined time have similar features. It is assumed that When the analysis query is input to the first model M1, the information processing apparatus 100 outputs the distributed expression (vector) of the analysis query “bone health” from the first model M1 (step S3).

  Subsequently, when the information processing apparatus 100 outputs the distributed expression (vector) of the analysis query “bone health”, the search query in which the similarity between the distributed expressions (vectors) exceeds a predetermined threshold is set as the analysis query “bone health”. It is extracted as a similar query (step S4). For example, the information processing apparatus 100 extracts a distributed expression (vector) located near the output distributed expression (vector) of the analysis query “bone health” in the distributed expression space. Subsequently, the information processing apparatus 100 calculates the similarity between the distributed expression (vector) of the analysis query “bone health” and the distributed expression (vector) located in the vicinity. For example, the information processing apparatus 100 calculates the cosine similarity between distributed expressions (vectors). Subsequently, the information processing apparatus 100 extracts a distributed expression (vector) whose cosine similarity exceeds a predetermined threshold as a distributed expression (vector) similar to the distributed expression (vector) of the analysis query “bone health”. Note that the information processing apparatus 100 calculates the similarity between distributed expressions (vectors) based on any index as long as it is an index applicable as a distance measure between vectors, not limited to the cosine similarity. Good. For example, the information processing apparatus 100 calculates a value of a predetermined distance function such as a Euclidean distance between distributed expressions (vectors), a distance in a non-Euclidean space such as a hyperbolic space, a Manhattan distance, a Mahalanobis distance, or the like. Subsequently, the information processing apparatus 100 analyzes the distributed expression (vector) in which the value of the predetermined distance function (that is, the distance in the distributed expression space) between the distributed expressions (vectors) is less than a predetermined threshold value in the analysis query “bone health” It may be extracted as a distributed expression (vector) similar to the distributed expression (vector). Subsequently, when the information processing apparatus 100 extracts a similar distributed expression (vector), the information processing apparatus 100 refers to the vector information storage unit 122 described later and specifies a search query corresponding to the extracted distributed expression (vector). Then, the information processing apparatus 100 extracts the specified search query as a similar query similar to the analysis query.

  Subsequently, when the information processing device 100 extracts the similar query, the information processing device 100 inputs the extracted similar query to the second model M2 (step S5). When the similar query is input to the second model M2, the information processing apparatus 100 outputs the category to which the similar query belongs from the second model M2 (step S6). The information processing apparatus 100 receives the classification axis (category) specified by the client who is the advertiser, and then learns the second model M2 so as to classify the search queries according to the received classification axis (category). You may let me. Alternatively, the information processing apparatus 100 trains the second model M2 in advance so as to classify the search queries according to the categories shown in the classification definition storage unit 123 described later. Then, the information processing apparatus 100 causes the client, which is the advertiser, to select a classification axis from the categories shown in the classification definition storage unit 123, and the classification axis (category) selected by the client, which is the advertiser, is selected from the client terminal 20. You may accept.

  Then, the information processing apparatus 100 transmits the classification result of the query according to the classification axis (category) to the client terminal 20 (step S7). For example, the information processing apparatus 100 provides the configuration of the search query searched as the search query regarding the analysis theme “bone health related” as the classification result of the query according to the classification axis (category). For example, as a breakdown of the search queries extracted as similar queries similar to the analysis query, the information processing apparatus 100 has a search query rate of “drink” of 40%, a search query rate of “food” of 30%, and “exercise”. The analysis result shows that the ratio of the search query regarding “” is 20% and the ratio of the search query regarding “supplement” is 10%.

  As described above, the information processing apparatus 100 assumes that the plurality of search queries input by the same user within a predetermined time period have similar features and learns the features of the plurality of search queries. Is used to extract a similar query that is a search query having characteristics similar to a predetermined search query. As will be described later, the first learning model makes it possible to properly interpret the meaning of the search query based on the context of the user who inputs the search query. Therefore, the information processing apparatus 100 can analyze the search trend of the user by using the first learning model in consideration of the context of the user who inputs the search query. Therefore, the information processing apparatus 100 can improve the accuracy of analysis of the search trend of the user.

[Generation processing of first learning model]
Next, the flow of generation processing of the first learning model will be described with reference to FIG. FIG. 2 is a diagram illustrating an example of a generation process of the first learning model according to the embodiment. In the example illustrated in FIG. 2, the information processing apparatus 100 is composed of a search query Q11 “Roppongi pasta” and a search query Q12 “Roppongi Italian” that are continuously input by the same user U1 within a predetermined time. The search query is extracted (step S11).

  Subsequently, the information processing apparatus 100 inputs the extracted search query Q11 into the first model M1 and outputs a vector BQV11 that is a distributed expression of the search query Q11. Here, the vector BQV11 is a distributed expression of the search query Q11 just output from the output layer of the first model M1, and represents a distributed expression before feedback (before learning) to the first model M1. Further, the information processing apparatus 100 inputs the extracted search query Q12 into the first model M1 and outputs the vector BQV12 that is a distributed expression of the search query Q12. Here, the vector BQV12 is a distributed expression of the search query Q12 just output from the output layer of the first model M1, and represents a distributed expression before feedback is given to the first model M1 (before learning). In this way, the information processing apparatus 100 outputs the vector BQV11, which is the distributed expression of the search query Q11, and the vector BQV12, which is the distributed expression of the search query Q12 (step S12).

  Subsequently, the information processing apparatus 100 includes a pair of searches including a search query Q11 (“Roppongi pasta”) and a search query Q12 (“Roppongi Italian”) continuously input by the same user U1 within a predetermined time. Since the query is presumed to be a search query input with a predetermined search intention (for example, a search intention of “searching for a restaurant at a certain place”), the search query Q11 is assumed to have similar features to each other. The first model M1 is trained so that the distributed expression (vector QV11) of (1) and the distributed expression (vector QV12) of the search query Q12 paired with the search query Q11 are similar in the distributed expression space. For example, the magnitude of the angle formed by the vector BQV11, which is the distributed expression of the search query Q11 before feedback is applied to the first model M1 (before learning), and the vector BQV12, which is the distributed expression of the search query Q12, is Θ. Further, the angle between the vector QV11, which is the distributed expression of the search query Q11 after feedback is applied to the first model M1 (after learning), and the vector QV12, which is the distributed expression of the search query Q12, is Φ. At this time, the information processing apparatus 100 trains the first model M1 so that Φ becomes smaller than Θ. For example, the information processing apparatus 100 calculates the value of the cosine similarity between the vector BQV11 and the vector BQV12. The information processing apparatus 100 also calculates the value of the cosine similarity between the vector QV11 and the vector QV12. Then, the information processing apparatus 100 makes the value of the cosine similarity of the vector QV11 and the vector QV12 larger than the value of the cosine similarity of the vector BQV11 and the vector BQV12 (so that the value approaches 1). The model M1 is trained. As described above, the information processing apparatus 100 learns the first model M1 so that two vectors, which are a pair of distributed expressions corresponding to a pair of search queries, are similar in the distributed expression space, and thus the information from the search query is distributed. A first model M1 that outputs an expression (vector) is generated (step S13). Note that the information processing apparatus 100 calculates the similarity between distributed expressions (vectors) based on any index as long as it is an index applicable as a distance measure between vectors, not limited to the cosine similarity. Good. Moreover, the information processing apparatus 100 may train the first model M1 based on any index as long as it is an index applicable as a distance measure between vectors. For example, the information processing apparatus 100 calculates a value of a predetermined distance function such as a Euclidean distance between distributed expressions (vectors), a distance in a non-Euclidean space such as a hyperbolic space, a Manhattan distance, a Mahalanobis distance, or the like. Subsequently, the information processing apparatus 100 may train the first model M1 so that the value of the predetermined distance function between the distributed expressions (vectors) (that is, the distance in the distributed expression space) becomes small.

  Next, the flow of the generation process of the first learning model will be described in more detail with reference to FIG. In the description of FIG. 3, the portions overlapping the description of FIG. 2 will be omitted as appropriate. FIG. 3 is a diagram illustrating a generation process of the first learning model according to the embodiment. In the example shown in FIG. 3, the distributed representation (vector) output by the first model M1 generated by the information processing apparatus 100 is mapped in the distributed representation space. The information processing apparatus 100 trains the first model M1 so that a distributed expression of a predetermined search query and a distributed expression of another search query paired with the predetermined search query are closely mapped in the distributed expression space. To do.

  In the example illustrated in the upper part of FIG. 3, the information processing apparatus 100 has a search query Q11 (“Roppongi pasta”), which is four search queries continuously input by the same user U1 within a predetermined time, and a search query. Q12 (“Roppongi Italian”), search query Q13 (“Akasaka pasta”), and search query Q14 (“Azabu pasta”) are extracted. The information processing apparatus 100 extracts four search queries whose time intervals at which each search query is input by the same user U1 are within a predetermined time. The information processing apparatus 100 extracts a plurality of search queries in which a time interval in which a pair of search queries to be described later are input by the same user U1 is within a predetermined time. When the information processing apparatus 100 arranges the search queries in the order in which they are input, the information processing apparatus 100 extracts the four search queries input in the order of the search query Q11, the search query Q12, the search query Q13, and the search query Q14. When extracting four search queries, the information processing apparatus 100 defines two search queries that are adjacent in time series as a pair of search queries, and is a pair of three search queries (search query Q11, search query Q12). , (Search query Q12, search query Q13), (search query Q13, search query Q14) are extracted (step S21-1). The information processing apparatus 100 may extract a plurality of search queries in which all the search queries have been input by the same user U1 within a predetermined time. Then, the information processing apparatus 100 selects two search queries from the plurality of extracted search queries regardless of whether or not they are adjacent in time series and sets the selected two search queries as a pair of search queries. May be extracted as

  Then, the information processing apparatus 100 inputs the extracted search query Q1k (k = 1, 2, 3, 4) into the first model M1 to search for the search query Q1k (k = 1, 2, 3, 4). A vector BQV1k (k = 1, 2, 3, 4) that is a distributed expression is output. Here, the vector BQV1k (k = 1, 2, 3, 4) is a distributed representation of the search query Q1k (k = 1, 2, 3, 4) just output from the output layer of the first model M1. , A distributed expression before feedback (before learning) to the first model M1 is shown (step S22-1).

  Subsequently, the information processing apparatus 100 determines that a pair of search queries continuously input by the same user U1 within a predetermined time period has a predetermined search intention (for example, “food and drink at a certain place (in Minato-ku, Tokyo)”). It is estimated that the search queries have been input with the "search intent to search for a store"), and therefore the distributed representation (vector QV11) of the search query Q11 and the search query Q11 are considered to have similar features to each other. The first model M1 is trained so that the distributed expression (vector QV12) of the search query Q12 becomes similar in the distributed expression space. Further, the information processing apparatus 100 determines that the distributed representation of the search query Q12 (vector QV12) and the distributed representation of the search query Q13 (vector QV13) paired with the search query Q12 are similar in the distributed representation space. 1 Learn the model M1. In addition, the information processing apparatus 100 determines that the distributed expression (vector QV13) of the search query Q13 and the distributed expression (vector QV14) of the search query Q14 paired with the search query Q13 are similar in the distributed expression space. 1 Learn the model M1. As described above, the information processing apparatus 100 learns the first model M1 so that two vectors, which are a pair of distributed expressions corresponding to a pair of search queries, are similar in the distributed expression space, and thus the information from the search query is distributed. A first model M1 that outputs an expression (vector) is generated (step S23-1).

  As a result of the information processing shown in the upper part of FIG. 3, the vector QV1k (k = 1, 2, 3, 4), which is the distributed expression of the search query Q1k (k = 1, 2, 3, 4), is close to the distributed expression space. It is shown that the position is mapped as the cluster CL11. For example, the search query Q1k (k = 1, 2, 3, 4) is a set of search queries searched by the user U1 under the search intent of “search for a restaurant in a certain place (near Minato-ku, Tokyo)”. Is estimated to be That is, the search query Q1k (k = 1, 2, 3, 4) is a search query that is searched under the search intent of “searching for a restaurant in a certain place (near Minato-ku, Tokyo)”. , Are presumed to be search queries having similar features to each other. Here, the information processing apparatus 100 receives the position of the cluster CL11 when a predetermined search query input with the search intention of “searching for a restaurant in a certain place (near Minato-ku, Tokyo)” is input to the first model. It is possible to output a distributed expression that is mapped to. As a result, for example, the information processing apparatus 100 extracts a search query corresponding to the distributed expression mapped to the position of the cluster CL11 to search for “a restaurant in a certain place (Near Minato-ku, Tokyo)”. It is possible to extract a search query according to the intention. Therefore, the information processing apparatus 100 can appropriately interpret the meaning of the search query.

  In the example illustrated in the lower part of FIG. 3, the information processing apparatus 100 has a search query Q21 (“refrigerator 400L”), which is three search queries continuously input by the same user U2 within a predetermined time, and a search query. Q22 (“refrigerator medium size”) and search query Q23 (“refrigerator medium size recommended”) are extracted. When the information processing apparatus 100 arranges the search queries in the input order, the information processing apparatus 100 extracts the three search queries input in the order of the search query Q21, the search query Q22, and the search query Q23. When the information processing apparatus 100 extracts the three search queries, the two search queries that are adjacent in time series are regarded as a pair of search queries and are a pair of two search queries (search query Q21, search query Q22). , (Search query Q22, search query Q23) are extracted (step S21-2).

  Subsequently, the information processing apparatus 100 inputs the extracted search query Q2m (m = 1, 2, 3) into the first model M1 and is a distributed expression of the search query Q2m (m = 1, 2, 3). The vector BQV2m (m = 1, 2, 3) is output. Here, the vector BQV2m (m = 1, 2, 3) is a distributed representation of the search query Q2m (m = 1, 2, 3) just output from the output layer of the first model M1, and is the first model. A distributed expression before feedback is given to M1 (before learning) is shown (step S22-2).

  Subsequently, the information processing apparatus 100 has a predetermined search intention (for example, a search intention of “searching a medium-sized refrigerator”) for a pair of search queries continuously input by the same user U2 within a predetermined time. Since it is estimated that the search queries are input, it is assumed that the search query Q21 has a distributed expression (vector QV21) and the search query Q22 has a distributed expression (vector) that has characteristics similar to each other. QV22) trains the first model M1 so that it is similar to QV22) in the distributed representation space. Further, the information processing apparatus 100 determines that the distributed representation of the search query Q22 (vector QV22) and the distributed representation of the search query Q23 (vector QV23) paired with the search query Q22 are similar in the distributed representation space. 1 Learn the model M1. In this way, the information processing apparatus 100 learns the first model M1 so that two vectors, which are a pair of distributed expressions corresponding to a pair of search queries, are similar in the distributed expression space, and thus the information from the search query is distributed. A first model M1 that outputs an expression (vector) is generated (step S23-2).

  As a result of the information processing shown in the lower part of FIG. 3, the vector QV2m (m = 1, 2, 3), which is the distributed expression of the search query Q2m (m = 1, 2, 3), is located near the cluster CL21 in the distributed expression space. Is mapped as. For example, the search query Q2m (m = 1, 2, 3) is presumed to be a set of search queries searched by the user U2 under the search intention of “searching for a medium-sized refrigerator”. That is, Q2m (m = 1, 2, 3) is a search query having similar characteristics in that it is a search query searched under the search intention of “searching for a medium-sized refrigerator”. Presumed. Here, the information processing apparatus 100, when a predetermined search query input with the search intention of “search for a medium-sized refrigerator” is input to the first model, a distributed expression that is mapped to the position of the cluster CL21. Can be output. Thereby, for example, the information processing apparatus 100 extracts the search query corresponding to the distributed expression mapped to the position of the cluster CL21, thereby extracting the search query according to the search intention of “inspecting the medium-sized refrigerator”. be able to. Therefore, the information processing apparatus 100 can appropriately interpret the meaning of the search query.

  In addition, the information processing apparatus 100 according to the present invention is a search query having mutually different characteristics in that the plurality of randomly extracted search queries are search queries searched under different search intentions. The first model M1 is learned as if it exists. Specifically, in the information processing apparatus 100, the distributed expression of the predetermined search query and the distributed expression of the search query randomly extracted irrespective of the predetermined search query are mapped far in the distributed expression space. Thus, the training of the first model M1 is performed. In the example illustrated in FIG. 3, the information processing apparatus 100 is assumed to extract the search query Q21 when the search query is randomly extracted regardless of the search query Q11. In this case, in the information processing apparatus 100, the distributed representation (vector QV11) of the search query Q11 and the distributed representation (vector QV21) of the search query Q21 that is randomly extracted irrespective of the search query Q11 are on the distributed representation space. The first model M1 is trained so as to be mapped far. As a result, the vector QV1k, which is a distributed expression of the search query Q1k (k = 1, 2, 3, 4) searched under the search intent of “searching for a restaurant in a certain place (near Minato-ku, Tokyo)”. It is a distributed representation of a cluster CL11 including (k = 1, 2, 3, 4) and a search query Q2m (m = 1, 2, 3) searched under the search intention of “searching for a medium-sized refrigerator”. The cluster CL21 including the vector QV2m (m = 1, 2, 3) is mapped far in the distributed representation space. That is, the information processing apparatus 100 according to the present invention distributes the distributed expressions of the search queries with different search intentions by learning the first model M1 so that the distributed expressions of the plurality of randomly extracted search queries are different. It is possible to output to a distant position in the expression space.

  As a result of the distributed representation (vector) output by the first model M1 generated by the information processing apparatus 100 being mapped in the distributed representation space, in addition to the cluster CL11 and the cluster CL21 described above, the same user can specify a predetermined value. A cluster CL12 or a cluster CL22, which is a set of distributed expressions (vectors) of a plurality of search queries input within the time period, is generated.

  As described above, the information processing device 100 acquires the search query input by the user. In addition, the information processing apparatus 100 learns that a plurality of search queries input by the same user within a predetermined time have similar characteristics among the acquired search queries, and thus the predetermined search query is performed. Generate a first model that predicts the feature information of the search query. That is, the information processing apparatus 100 according to the present invention is similar to each other in that the plurality of search queries continuously input within a predetermined time are search queries searched under a predetermined search intention. The first model is trained by regarding it as a search query having a characteristic. Specifically, the information processing apparatus 100 learns the first model so that the distributed expressions of the plurality of search queries input by the same user within a predetermined time are similar to each other, and thereby the information processing apparatus 100 determines a predetermined search query from a predetermined search query. A first model that outputs a distributed expression including the characteristic information of the search query is generated. That is, the information processing apparatus 100 according to the present invention learns the first model M1 so that the distributed expressions of a plurality of search queries continuously input within a predetermined time are similar to each other so that a predetermined search intention is obtained. The distributed expression of the search query searched below can be output to a close position in the distributed expression space. This allows the information processing apparatus 100 to output (interpret) the meaning (search intent) of the search query according to the context of the user who inputs the search query. Therefore, the information processing apparatus 100 can appropriately interpret the meaning of the search query. Furthermore, the information processing apparatus 100 extracts a search query corresponding to a distributed expression that is mapped in the vicinity of the distributed expression that includes the characteristic information of the predetermined search query, so that the predetermined search query is searched according to the search intention. Search queries can be extracted. That is, the information processing apparatus 100 enables the user's search trend to be analyzed in consideration of the search intention and context of the user who inputs the search query. Therefore, the information processing apparatus 100 can improve the accuracy of analysis of the search trend of the user. Further, the first model M1 generated by the information processing device 100 can be made to function as a part of the search system. Alternatively, the information processing apparatus 100 distributes the search query output by the first model M1 as input information to another system (for example, a search engine) that uses the feature information of the search query predicted by the first model M1. You can also provide an expression. Accordingly, the search system can select the content output as the search result based on the characteristic information of the search query predicted by the first model M1. That is, the search system can select the content output as the search result in consideration of the search intention and context of the user who inputs the search query. Further, the search system calculates the similarity between the distributed expression of the character string included in the content output as the search result and the distributed expression of the search query, based on the characteristic information of the search query predicted by the first model M1. It will be possible. Then, the search system can determine the display order of the content output as the search result based on the calculated similarity. That is, the search system can determine the display order of the content output as the search result in consideration of the search intention and the context of the user who inputs the search query. Therefore, the information processing apparatus 100 can improve usability in the search service.

[Generation process of second learning model]
Next, the flow of the second learning model generation process will be described with reference to FIG. FIG. 4 is a diagram illustrating an example of the second learning model generation process according to the embodiment. In the following, the second learning model will be appropriately referred to as the second model (or the second model M2). In the example illustrated in the upper part of FIG. 4, the information processing apparatus 100 has a search query Q11 (“Roppongi pasta”), which is four search queries continuously input by the same user U1 within a predetermined time, and a search query. Q12 (“Roppongi Italian”), search query Q13 (“Akasaka pasta”), and search query Q14 (“Azabu pasta”) are extracted. The information processing apparatus 100 extracts a plurality of search queries in which the time intervals at which the search queries are input by the same user U1 are within a predetermined time. Further, the information processing apparatus 100 extracts a plurality of search queries in which the time intervals at which the same user U1 inputs each search query pair are within a predetermined time. Here, it is assumed that the four search queries are search queries in which the search query Q11, the search query Q12, the search query Q13, and the search query Q14 are input by the user U1 within a predetermined time. When extracting four search queries, the information processing apparatus 100 defines two search queries that are adjacent in time series as a pair of search queries, and is a pair of three search queries (search query Q11, search query Q12). , (Search query Q12, search query Q13), (search query Q13, search query Q14) are extracted. When the information processing apparatus 100 extracts three pairs of search queries, the extracted search query Q1k (k = 1, 2, 3, 4) is input to the first model M1 (step S31). The information processing apparatus 100 may extract a plurality of search queries in which all the search queries have been input by the same user U1 within a predetermined time. Then, the information processing apparatus 100 selects two search queries from the plurality of extracted search queries regardless of whether or not they are adjacent in time series and sets the selected two search queries as a pair of search queries. May be extracted as

  Subsequently, the information processing apparatus 100 uses the vector BQV1k (k = 1, 2, 3, 4), which is the distributed expression of the search query Q1k (k = 1, 2, 3, 4), as the output data of the first model M1. Output (step S32). Here, the vector BQV1k (k = 1, 2, 3, 4) is a distributed representation of the search query Q1k (k = 1, 2, 3, 4) just output from the output layer of the first model M1. , A distributed expression before feedback is applied to the first model M1 (before learning).

  Here, the search query Q1k (k = 1, 2, 3, 4) continuously input by the same user U1 within a predetermined time is, for example, “some place (near Minato-ku, Tokyo)” by the user U1. It is presumed to be a set of search queries searched under the search intention of "searching for restaurants in." That is, the search query Q1k (k = 1, 2, 3, 4) is a search query that is searched under the search intent of “searching for a restaurant in a certain place (near Minato-ku, Tokyo)”. , Are presumed to be search queries having similar features to each other. Therefore, the information processing apparatus 100 generates a first model that predicts the characteristic information of the predetermined search query from the predetermined search query by learning that the successively input search queries have similar characteristics. (Step S33). Specifically, the information processing apparatus 100 learns that the distributed expressions of continuously input search queries are similar to each other, thereby predicting the distributed expression of the predetermined search query from the predetermined search query. Generate M1. For example, the information processing apparatus 100 determines that the distributed representation of the search query Q11 (vector QV11) and the distributed representation of the search query Q12 (vector QV12) paired with the search query Q11 are similar in the distributed representation space. 1 Learn the model M1. Further, the information processing apparatus 100 determines that the distributed representation of the search query Q12 (vector QV12) and the distributed representation of the search query Q13 (vector QV13) paired with the search query Q12 are similar in the distributed representation space. 1 Learn the model M1. In addition, the information processing apparatus 100 determines that the distributed expression (vector QV13) of the search query Q13 and the distributed expression (vector QV14) of the search query Q14 paired with the search query Q13 are similar in the distributed expression space. 1 Learn the model M1.

  On the right side of the upper part of FIG. 4, the variance of the search query Q1k (k = 1, 2, 3, 4) input by the same user U1 within a predetermined time as the output result of the learned first model M1. It is shown that the expression vector QV1k (k = 1, 2, 3, 4) is mapped as the cluster CL11 in the distributed expression space. In this way, the information processing apparatus 100 generates the first learning model M1 in which the features of the plurality of search queries input by the same user within a predetermined time are learned.

  After generating the first model M1, the information processing apparatus 100 acquires the generated first model M1 (model data MDT1 of the first model M1). When the information processing apparatus 100 acquires the first model M1, it uses the acquired first model M1 to generate a second learning model M2. Specifically, the information processing apparatus 100 re-learns the first model M1 to generate a second model M2 having a connection coefficient that is a weight of the learning model different from that of the first model M1. More specifically, the information processing apparatus 100 uses the first model M1 to generate a second learning model M2 that predicts a category to which a predetermined search query belongs from a predetermined search query (step S34).

  In the example illustrated in the lower part of FIG. 4, when the search query is input to the second model M2, the information processing apparatus 100 CAT11 (“search for a restaurant”), CAT12 (“search for a product”), CAT13 ( A second model M2 that predicts which one of the four categories “reserve a restaurant”) and CAT14 (“purchase a product”) is generated. Specifically, when the search query is input to the second model M2 as the input information, the information processing apparatus 100 outputs the second model M2 that outputs the probability that the search query belongs to the category for each category as the output information. To generate. For example, the information processing apparatus 100 learns a set of a search query and a category (one of CAT11 to CAT14) to which the search query belongs as the correct answer data of the second model M2.

  The fact that the search query belongs to CAT11 (“search for restaurants”) indicates that the search query is a search query that was input with the intention of searching for restaurants. Belonging to CAT12 (“search for product”) indicates that the search query is a search query input with the intention of searching for a product. In addition, the fact that the search query belongs to CAT13 (“reserve a restaurant”) indicates that the search query is a search query that is input with the intention of reserving a restaurant. Further, the fact that the search query belongs to CAT14 (“purchase the product”) indicates that the search query is a search query input with the intention of purchasing the product.

  Specifically, when the search query is input to the learning model, the information processing apparatus 100 learns so that the classification result of the distributed expression output by the learning model corresponds to the category to which the search query belongs, A second model M2 that predicts a category to which the predetermined search query belongs is generated from the predetermined search query. Then, the information processing apparatus 100 outputs the probability that the search query belongs to the category as the output information for each of the categories CAT11 to CAT14 when the search query is input to the second model M2 as the input information, for example. Generate M2.

  For example, when the search query Q11 (“Roppongi pasta”) is input to the second model M2 as input information (step S35), the information processing apparatus 100 distributes the search query Q11 (“Roppongi pasta”) as output information. The vector BQV11, which is the expression, is output. Here, the vector BQV11 is a distributed expression of the search query Q11 just output from the output layer of the second model M2, and represents a distributed expression before feedback is applied to the second model M2 (before learning). Here, it is assumed that the correct category to which the search query Q11 (“Roppongi pasta”) belongs is CAT11 (“Find a restaurant”). In this case, the information processing apparatus 100 sets the probability that the vector BQV11, which is the distributed expression of the output search query Q11 (“Roppongi pasta”), is classified into CAT11 (“Search for restaurants”) exceeds a predetermined threshold. To train the second model M2. The information processing apparatus 100 uses the correct answer data prepared in advance to train the second model. The information processing apparatus 100 may generate correct answer data of the second model M2. Then, the information processing apparatus 100 may learn the second model M2 using the generated correct answer data. Specifically, the information processing apparatus 100 determines the correct category to which the search query belongs, based on the post-search behavior of the user who searched for the search query. More specifically, the information processing apparatus 100 associates, with respect to the user who has searched for a predetermined search query, a predetermined action in which the ratio of users who have performed a predetermined action after the search exceeds a predetermined threshold, to the correct category. Determined as the action to be taken. For example, as a ratio of users who searched for the search query Q11 (“Roppongi pasta”) to take a predetermined action after the search, a ratio of users who took action to search for a restaurant was 90%, and an action to search for a product after the search. It is assumed that the ratio of users who have caused the problem is 0%, the ratio of users who have made an action to reserve a restaurant after the search is 10%, and the ratio of users who have made an action to purchase a product after the search is 0%. In this case, the information processing apparatus 100 uses the search query Q11 (“Roppongi pasta”) to search for an action because the proportion of users who have taken the action to search for the restaurant exceeds a predetermined threshold value (for example, 90%). Determined as an action corresponding to the correct category. Then, since the information processing apparatus 100 determines that the action corresponding to the correct answer category is the action of searching for a restaurant, the correct answer category to which the search query Q11 (“Roppongi pasta”) belongs is CAT11 (“search for a restaurant”). To decide.

  For example, when the search query Q11 (“Roppongi pasta”) is input to the second model M2 before learning, the information processing apparatus 100 classifies the vector BQV11, which is a distributed expression, into CAT11 (“Find a restaurant”). The probability of being classified as 80%, the probability of being classified as CAT12 (“Search for a product”) is 0%, the probability of being classified as CAT13 (“Reserve a restaurant”) is 20%, and CAT14 (“Purchase a product”). It is assumed that the probability of being classified as) is output as 0%. In this case, the information processing apparatus 100 learns the second model M2 so that the probability that the vector BQV11, which is a distributed expression, is classified into CAT11 (“search for a restaurant”) exceeds a predetermined threshold value (for example, 90%). Let In addition, the information processing apparatus 100 performs learning so that the probability that the vector BQV11, which is a distributed expression, is classified into CAT11 (“search for a restaurant”) exceeds a predetermined threshold value (for example, 90%), The second model M2 is trained so that the probability that the vector BQV11, which is a distributed expression, is classified into another category CAT13 (“reserve a restaurant”) is reduced to 10%.

  As described above, when the predetermined search query is input as the input information, the information processing apparatus 100 sets the probability that the distributed expression of the predetermined search query as the output information is classified into the correct category to exceed the predetermined threshold. Train two models. Then, when the predetermined search query is input as the input information, the information processing apparatus 100 sets the category in which the probability that the distributed expression of the predetermined search query belongs to the category exceeds the predetermined threshold to the category of the predetermined search query. Output as. For example, when the search query Q11 (“Roppongi pasta”) is input to the learned second model M2 as input information, the information processing apparatus 100 is a vector BQV11 that is a distributed expression of the search query Q11 (“Roppongi pasta”). Belongs to the category CAT11 (“search for restaurants”) exceeds 90%, the category to which the search query belongs is output as output information as CAT11 (“searches for restaurants”) (step S36). As described above, the information processing apparatus 100 learns the set of the search query and the correct category of the search query to generate the second model that predicts the category of the predetermined search query from the predetermined search query (step S37). ).

  In general, a user is considered to perform a search a plurality of times with a certain intention, and therefore it is assumed that search queries continuously input within a predetermined time have similar search intentions. Therefore, the information processing apparatus 100 according to the present invention is similar to each other in that a plurality of search queries continuously input within a predetermined time period are search queries searched under a predetermined search intention. The first model M1 is trained on the assumption that the search query has a characteristic that Accordingly, the information processing apparatus 100 can cause the first model M1 to learn the characteristics of the search query in consideration of the search intention. Then, the information processing apparatus 100 utilizes the first model M1 that has learned the characteristics of the search query in consideration of the search intention, and efficiently uses the second model that predicts the category of the predetermined search query from the predetermined search query. Can be generated. As a result, the information processing apparatus 100 enables the search query to be classified into categories in consideration of the search intention of the user who inputs the search query. Further, conventionally, it is necessary to prepare a sufficient amount of correct answer data in order to classify search queries into categories and obtain high classification accuracy. However, since the search queries themselves are diverse and have a long tail property, it is very troublesome and difficult to label the correct answer categories corresponding to a large number of search queries. Here, the information processing apparatus 100 learns the second model that predicts the category of the search query by using the user's search intention (the context of the user who inputs the search query) as a kind of correct answer, instead of labeling the correct answer category. Can be made. With this, the information processing apparatus 100 can train the second model without manually labeling the correct category of the search query. That is, the information processing apparatus 100 can obtain sufficient classification accuracy even when the correct answer data is small. In addition, the information processing apparatus 100 can obtain higher classification accuracy when there are many correct data. Therefore, the information processing device 100 can improve the classification accuracy of the search query.

[2. Configuration of information processing device]
Next, the configuration of the information processing apparatus 100 according to the embodiment will be described with reference to FIG. FIG. 6 is a diagram illustrating a configuration example of the information processing device 100 according to the embodiment. As shown in FIG. 6, the information processing device 100 includes a communication unit 110, a storage unit 120, and a control unit 130. The information processing apparatus 100 includes an input unit (for example, a keyboard and a mouse) that receives various operations from an administrator of the information processing apparatus 100 and a display unit (for example, a liquid crystal display) for displaying various information. You may have.

(Communication unit 110)
The communication unit 110 is realized by, for example, a NIC (Network Interface Card) or the like. Then, the communication unit 110 is connected to the network by wire or wirelessly, and transmits and receives information between the user terminal 10 and the search server 50, for example.

(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. 6, the storage unit 120 includes a query information storage unit 121, a vector information storage unit 122, a classification definition storage unit 123, a category information storage unit 124, and a model information storage unit 125.

(Query information storage unit 121)
The query information storage unit 121 stores various information regarding the search query input by the user. FIG. 7 shows an example of the query information storage unit according to the embodiment. In the example shown in FIG. 7, the query information storage unit 121 has items such as “user ID”, “date and time”, “search query”, and “search query ID”.

  The “user ID” indicates identification information for identifying the user who has input the search query. The “date and time” indicates the date and time when the search server accepted the search query from the user. "Search query" indicates a search query input by the user. The “search query ID” indicates identification information for identifying the search query input by the user.

  In the example shown in the first record of FIG. 7, the search query (search query Q11) identified by the search query ID “Q11” corresponds to the search query Q11 shown in FIG. The user ID “U1” indicates that the user who inputs the search query Q11 is the user (user U1) identified by the user ID “U1”. The date and time "2018/9/1 PM 17:00" indicates that the date and time when the search server receives the search query Q11 from the user U1 is 17:00 pm on September 1, 2018. The search query "Roppongi pasta" indicates the search query Q11 input by the user U1. Specifically, the search query “Roppongi pasta” indicates that it is a character string in which “Roppongi” indicating a place name and “pasta” indicating the type of food are separated by a space, which is a delimiter.

(Vector information storage unit 122)
The vector information storage unit 122 stores various kinds of information regarding vectors, which are distributed expressions of search queries. FIG. 8 shows an example of the vector information storage unit according to the embodiment. In the example shown in FIG. 8, the vector information storage unit 122 has items such as “vector ID”, “search query ID”, and “vector information”.

  “Vector ID” indicates identification information for identifying a vector that is a distributed expression of a search query. The “search query ID” indicates identification information for identifying the search query corresponding to the vector. “Vector information” indicates an N-dimensional vector that is a distributed expression of a search query. The vector that is the distributed expression of the search query is, for example, a 128-dimensional vector.

  In the example shown in the first record of FIG. 8, the vector identified by the vector ID “QV11” (vector QV11) corresponds to the vector QV11 that is the distributed expression of the search query Q11 shown in FIG. The search query (search query Q11) identified by the search query ID “Q11” indicates that the search query corresponding to the vector QV11 is the search query Q11. The vector information “QVDT11” indicates an N-dimensional vector that is a distributed expression of the search query Q11.

(Classification definition storage unit 123)
The classification definition storage unit 123 stores various kinds of information regarding the definition of the category into which the search query is classified. FIG. 9 shows an example of the classification definition storage unit according to the embodiment. In the example shown in FIG. 9, the classification definition storage unit 123 has items such as “large classification ID”, “large classification”, “small classification ID”, and “small classification”.

  The “major classification” indicates the major classification of the categories into which the search query is classified. The “major classification ID” indicates identification information for identifying the major classification. In the example shown in FIG. 9, the large classification “purchasing behavior type” corresponds to the large classification described in the example shown in the lower part of FIG. The large classification "purchasing behavior system" indicates a large classification of categories into which the search query is classified based on the purchasing behavior of the user. In the example shown in FIG. 9, the large classification “purchasing behavior type” further has four small classifications. The large classification ID “CAT1” indicates identification information for identifying the large classification “purchasing behavior type”.

  “Minor classification” indicates a minor classification of the category into which the search query is classified. The “small classification ID” indicates identification information for identifying the small classification. In the example shown in FIG. 9, the small category “search for restaurants” belongs to the large category “purchase behavior system”, and the search query classified into the small categories is input by the user with the intention of searching for restaurants. Indicates that the search query was performed. The small classification ID “CAT11” indicates identification information for identifying the small classification “search for restaurants”.

  The small category “search for products” is a category belonging to the large category “purchasing behavior”, and indicates that the search queries classified into the small categories are search queries input by the user with the intention of searching for products. . The small classification ID “CAT12” indicates identification information for identifying the small classification “search for products”.

  The small category “reserve a restaurant” is a category belonging to the large category “purchasing behavior type”, and the search query classified into the small category is a search query input by the user with the intention of reserving a restaurant. Indicates that. The small classification ID “CAT13” indicates identification information for identifying the small classification “reserve a restaurant”.

  The small category “Purchase goods” is a category that belongs to the large category “Purchasing behavior type”, and the search queries classified into the small categories are search queries input by the user with the intention of purchasing the product. Show. The small classification ID “CAT14” indicates identification information for identifying the small classification “purchase a product”.

(Category information storage unit 124)
The category information storage unit 124 stores various kinds of information regarding the category to which the search query belongs. Specifically, the category information storage unit 124 stores various kinds of information regarding the categories output by the second learning model when the search query is input to the learned second learning model. FIG. 10 shows an example of the category information storage unit according to the embodiment. In the example illustrated in FIG. 10, the category information storage unit 124 has items such as “search query ID”, “large classification ID”, “small classification ID”, and “probability (%)”.

  The “search query ID” indicates identification information for identifying the search query input by the user. In the example shown in FIG. 10, the search query (search query Q11) identified by the search query ID “Q11” corresponds to the search query Q11 shown in FIG.

  The “major classification ID” indicates identification information for identifying the major classification. The “small classification ID” indicates identification information for identifying the small classification. The “probability (%)” indicates the probability of each subclass output by the second learning model when a search query is input to the already learned second learning model. In the example shown in FIG. 10, the probability (%) “90” indicates that the probability that the search query Q11 is classified into the category CAT11 is 90%.

(Model information storage unit 125)
The model information storage unit 125 stores various kinds of information regarding the learning model generated by the information processing device 100. FIG. 11 shows an example of the model information storage unit according to the embodiment. In the example shown in FIG. 11, the model information storage unit 125 has items such as “model ID” and “model data”.

  The “model ID” indicates identification information for identifying the learning model generated by the information processing device 100. “Model data” indicates model data of a learning model generated by the information processing device 100. For example, “model data” stores data for converting a search query into a distributed expression.

  In the example shown in the first record of FIG. 11, the learning model identified by the model ID “M1” corresponds to the first model M1 shown in FIG. The model data “MDT1” indicates the model data (model data MDT1) of the first model M1 generated by the information processing device 100.

  The model data MDT1 includes an input layer to which a search query is input, an output layer, a first element belonging to any layer from the input layer to the output layer and other than the output layer, the first element, and the first element. A second element whose value is calculated based on the weight of one element, and a distributed expression of the search query input to the input layer is output from the output layer according to the search query input to the input layer. Thus, the information processing apparatus 100 may be made to function.

  Here, it is assumed that the model data MDT1 is realized by the regression model represented by “y = a1 * x1 + a2 * x2 + ... + ai * xi”. In this case, the first element included in the model data MDT1 corresponds to the input data (xi) such as x1 and x2. The weight of the first element corresponds to the coefficient ai corresponding to xi. Here, the regression model can be regarded as a simple perceptron having an input layer and an output layer. When each model is regarded as a simple perceptron, the first element can be regarded as a node included in the input layer, and the second element can be regarded as a node included in the output layer.

  It is also assumed that the model data MDT1 is realized by a neural network having one or a plurality of intermediate layers such as DNN (Deep Neural Network). In this case, the first element included in the model data MDT1 corresponds to any node included in the input layer or the intermediate layer. The second element corresponds to the node at the next stage, which is the node to which the value is transmitted from the node corresponding to the first element. The weight of the first element corresponds to the connection coefficient, which is a weight considered for the value transmitted from the node corresponding to the first element to the node corresponding to the second element.

  The information processing apparatus 100 calculates the distributed expression using a model having an arbitrary structure such as the regression model and the neural network described above. Specifically, the model data MDT1 is set with a coefficient so as to output a distributed expression when a search query is input. The information processing apparatus 100 uses such model data MDT1 to calculate the distributed representation.

  In the above example, the model data MDT1 is the model (hereinafter, referred to as model X1) that outputs the distributed expression of the search query when the search query is input. However, the model data MDT1 according to the embodiment may be a model generated based on a result obtained by repeating input / output of data to / from the model X1. For example, the model data MDT1 may be a model (hereinafter, referred to as model Y1) learned by inputting the distributed expression output by the model X1 when the search query is input. Alternatively, the model data MDT1 may be a model learned so that the search query is input and the output value of the model Y1 is output.

  Further, when the information processing apparatus 100 performs the estimation process using GAN (Generative Adversarial Networks), the model data MDT1 may be a model forming a part of GAN.

  In the example shown in the second record of FIG. 11, the learning model identified by the model ID “M2” corresponds to the second model M2 shown in FIG. Further, the model data “MDT2” indicates model data (model data MDT2) of the second model M2 generated by the information processing device 100.

  The model data MDT2 includes an input layer to which a search query is input, an output layer, a first element belonging to any layer from the input layer to the output layer and other than the output layer, the first element, and the first element. A second element whose value is calculated based on the weight of one element; and a probability that the search query input to the input layer belongs to each category according to the search query input to the input layer The information processing device 100 may be caused to function so as to output from.

  Here, it is assumed that the model data MDT2 is realized by a regression model represented by “y = a1 * x1 + a2 * x2 + ... + ai * xi”. In this case, the first element included in the model data MDT2 corresponds to the input data (xi) such as x1 and x2. The weight of the first element corresponds to the coefficient ai corresponding to xi. Here, the regression model can be regarded as a simple perceptron having an input layer and an output layer. When each model is regarded as a simple perceptron, the first element can be regarded as a node included in the input layer, and the second element can be regarded as a node included in the output layer.

  Further, it is assumed that the model data MDT2 is realized by a neural network having one or more intermediate layers such as DNN (Deep Neural Network). In this case, the first element included in the model data MDT2 corresponds to any node included in the input layer or the intermediate layer. The second element corresponds to the node at the next stage, which is the node to which the value is transmitted from the node corresponding to the first element. The weight of the first element corresponds to the connection coefficient, which is a weight considered for the value transmitted from the node corresponding to the first element to the node corresponding to the second element.

  The information processing apparatus 100 calculates the probability that the search query belongs to each category using a model having an arbitrary structure such as the regression model and the neural network described above. Specifically, the model data MDT2 is set with a coefficient so as to output the probability that the search query belongs to each category when the search query is input. The information processing apparatus 100 uses such model data MDT2 to calculate the probability that the search query belongs to each category.

  In the above example, the model data MDT2 is the model (hereinafter, referred to as model X2) that outputs the distributed expression of the search query when the search query is input. However, the model data MDT2 according to the embodiment may be a model generated based on a result obtained by repeating input / output of data to / from the model X2. For example, the model data MDT2 may be a model (hereinafter, referred to as model Y2) learned by inputting the distributed expression output by the model X2 when the search query is input. Alternatively, the model data MDT2 may be a model learned so that the search query is input and the output value of the model Y2 is output.

  When the information processing apparatus 100 performs the estimation process using GAN (Generative Adversarial Networks), the model data MDT2 may be a model forming a part of GAN.

(Control unit 130)
Returning to the description of FIG. 6, the control unit 130 is a controller, and is stored in a storage device inside the information processing 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 a generation program) in 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).

  In addition, the control unit 130 performs information processing in accordance with the first model M1 (model data MDT1) stored in the model information storage unit 125, in response to the search query input in the input layer, in each layer other than the output layer. The computer is made to function so as to output the distributed representation from the output layer by performing the calculation based on the first element and the weight of the first element with each element that belongs to the first element.

  In addition, the control unit 130 performs information processing in accordance with the second model M2 (model data MDT2) stored in the model information storage unit 125 on each layer other than the output layer with respect to the search query input to the input layer. The computer is made to function so that the probability that the search query belongs to each category is output from the output layer by performing an operation based on the first element and the weight of the first element with each element that belongs to the first element.

  As illustrated in FIG. 6, the control unit 130 includes an acquisition unit 131, an extraction unit 132, a generation unit 133, and a service providing unit 134, and implements or executes the information processing operation described below. Note that the internal configuration of the control unit 130 is not limited to the configuration shown in FIG. 6, and may be another configuration as long as it is a configuration for performing information processing described later.

(Acquisition unit 131)
The acquisition unit 131 acquires various information. Specifically, the acquisition unit 131 acquires the search query input by the user from the search server 50. When acquiring the search query input by the user, the acquisition unit 131 stores the acquired search query in the query information storage unit 121. The acquisition unit 131 also acquires vector information about a vector that is a distributed expression of the search query. When acquiring the vector information, the acquisition unit 131 stores the acquired vector information in the vector information storage unit 122. The acquisition unit 131 also acquires information defining the search query and the classification of the category to which the search query belongs. When acquiring the search query and the classification definition information that defines the classification of the category to which the search query belongs, the acquisition unit 131 stores the acquired classification definition information in the classification definition storage unit 123. Further, the acquisition unit 131 acquires category information regarding the category to which the search query belongs. When acquiring the category information, the acquisition unit 131 stores the acquired category information in the category information storage unit 124.

(Extractor 132)
The extraction unit 132 extracts various information. Specifically, the extraction unit 132 extracts, from the search queries acquired by the acquisition unit 131, a plurality of search queries input by the same user within a predetermined time period. For example, the extraction unit 132 extracts a plurality of search queries in which the time intervals at which the search queries are input by the same user are within a predetermined time. Next, the extraction unit 132 extracts a pair of search queries continuously input by the same user within a predetermined time from among a plurality of search queries input by the same user within a predetermined time. For example, the extraction unit 132 extracts a plurality of search queries in which the time interval at which each search query pair is input by the same user is within a predetermined time. For example, the extraction unit 132, among the search queries acquired by the acquisition unit 131, is a search query Q11 (“Roppongi pasta”) that is four search queries continuously input by the same user U1 within a predetermined time. ), Search query Q12 (“Roppongi Italian”), search query Q13 (“Akasaka pasta”), and search query Q14 (“Azabu pasta”). After arranging the search queries in the input order, the extraction unit 132 extracts the four search queries input in the order of the search query Q11, the search query Q12, the search query Q13, and the search query Q14. Subsequently, when the four search queries are extracted, the extraction unit 132 regards two search queries that are adjacent in time series as a pair of search queries and is a pair of three search queries (search query Q11, search query). Q12), (search query Q12, search query Q13), and (search query Q13, search query Q14) are extracted. The extraction unit 132 may extract a plurality of search queries in which all the search queries have been input by the same user within a predetermined time. Then, the extraction unit 132 selects two search queries from the plurality of extracted search queries, regardless of whether they are adjacent in time series, and sets the selected two search queries as a pair of search queries. You may extract.

  In addition, the extraction unit 132 extracts, from the search queries acquired by the acquisition unit 131, a predetermined search query and other search queries unrelated to the predetermined search query. For example, the extraction unit 132 extracts a predetermined search query from the search queries acquired by the acquisition unit 131. Subsequently, the extraction unit 132 randomly extracts another search query from the search queries acquired by the acquisition unit 131, regardless of the predetermined search query.

(Generator 133)
The generation unit 133 generates various information. Specifically, the generation unit 133 may learn that among the search queries acquired by the acquisition unit 131, a plurality of search queries input by the same user within a predetermined time have similar characteristics. , A learning model for predicting characteristic information of a predetermined search query from the predetermined search query is generated. Specifically, the generation unit 133 learns the learning model such that the distributed expressions of the plurality of search queries input by the same user within a predetermined time are similar to each other, thereby performing a predetermined search from the predetermined search query. A learning model for predicting query feature information is generated. For example, the generation unit 133 generates a learning model by learning so that the distributed expressions of a pair of search queries that are continuously input within a predetermined time period are similar to each other. For example, the generation unit 133 calculates the value of the similarity of the distributed expression (vector) before learning of the pair of search queries. In addition, the generation unit 133 calculates the value of the similarity of the distributed representation (vector) after learning of the pair of search queries. Subsequently, the generation unit 133 trains the learning model such that the similarity value of the distributed expression (vector) after learning is larger than the similarity value of the distributed expression (vector) before learning. As described above, the generation unit 133 trains the learning model so that the two vectors, which are a pair of distributed expressions corresponding to the pair of search queries, are similar in the distributed expression space, and thus the distributed expression (vector ) Is output to generate a learning model. More specifically, the generation unit 133 generates a learning model that outputs a distributed expression (vector) from a search query, using a DSSM technique that uses LSTM, which is a type of RNN, for distributed expression generation. For example, the generation unit 133 determines that, as correct answer data of the learning model, a pair of search queries input by the same user within a predetermined time have similar features and a distributed expression (vector) of the predetermined search query. , A predetermined search query and a distributed expression (vector) of another search query paired with each other are learned so as to exist close to each other in the distributed expression space. Further, when the generation unit 133 generates the first learning model, the generation unit 133 stores the generated first learning model (model data MDT1) in the model information storage unit 125 in association with the identification information for identifying the first learning model.

[Example of first learning model]
Here, an example of the first learning model generated by the information processing apparatus 100 will be described with reference to FIG. FIG. 12 is a diagram illustrating an example of the first learning model according to the embodiment. In the example illustrated in FIG. 12, the first learning model M1 generated by the information processing device 100 is composed of three layers of LSTM RNNs. In the example illustrated in FIG. 12, the extraction unit 132 includes a pair of search queries Q11 “Roppongi pasta” and a search query Q12 “Roppongi Italian” that are continuously input by the same user U1 within a predetermined time. Extract the search query. The generation unit 133 inputs the search query Q11 extracted by the extraction unit 132 into the input layer of the first learning model M1 (step S41).

  Then, the generation unit 133 outputs a 256-dimensional vector BQV11, which is a distributed expression of the search query Q11, from the output layer of the first learning model M1. Further, the generation unit 133 inputs the search query Q12 extracted by the extraction unit 132 into the input layer of the first learning model M1. Subsequently, the generation unit 133 outputs the 256-dimensional vector BQV12, which is the distributed expression of the search query Q12, from the output layer of the first learning model M1 (step S42).

  Subsequently, the generating unit 133 learns so that the distributed expressions (vectors) of the two consecutively input search queries are similar to each other, and thus the first learning model M1 that outputs the distributed expressions (vectors) from the search query. Is generated (step S43). For example, the size of the angle formed by the vector BQV11, which is the distributed expression of the search query Q11 before feedback (before learning) to the first learning model M1, and the vector BQV12, which is the distributed expression of the search query Q12, is Θ. Further, the size of the angle formed by the vector QV11, which is the distributed expression of the search query Q11 after feedback is applied to the first learning model M1 (after learning), and the vector QV12, which is the distributed expression of the search query Q12, is Φ. . At this time, the generation unit 133 trains the first learning model M1 so that Φ becomes smaller than Θ. For example, the generation unit 133 calculates the value of the cosine similarity between the vector BQV11 and the vector BQV12. The generation unit 133 also calculates the value of the cosine similarity between the vector QV11 and the vector QV12. Then, the generation unit 133 causes the learning model M1 so that the value of the cosine similarity between the vectors QV11 and QV12 becomes larger than the value of the cosine similarity between the vectors BQV11 and BQV12 (so that the value approaches 1). To learn. In this way, the generating unit 133 learns the first learning model M1 so that two vectors, which are a pair of distributed expressions corresponding to a pair of search queries, are similar in the distributed expression space, and thus the search query is distributed. A first learning model M1 that outputs an expression (vector) is generated. The generation unit 133 is not limited to the cosine similarity and may calculate the similarity between the distributed expressions (vectors) based on any index as long as it is an index applicable as a distance measure between vectors. Good. Further, the generation unit 133 may train the learning model M1 based on any index as long as it is an index applicable as a distance measure between vectors. For example, the generation unit 133 calculates a value of a predetermined distance function such as a Euclidean distance between distributed expressions (vectors), a distance in a non-Euclidean space such as a hyperbolic space, a Manhattan distance, a Mahalanobis distance, or the like. Subsequently, the generation unit 133 may train the learning model M1 so that the value of the predetermined distance function between the distributed expressions (vectors) (that is, the distance in the distributed expression space) becomes small.

  Further, the generation unit 133 learns that a plurality of search queries that are input by the same user within a predetermined time period have similar characteristics to each other including a character string delimited by a predetermined delimiter. By doing so, the first learning model is generated. For example, the generation unit 133 uses a search query “Roppongi pasta” in which “Roppongi” indicating a place name and the characters “pasta” indicating a food type are separated by a space that is a delimiter, and “Roppongi” indicating a place name. The first learning model is generated by learning that the search query “Roppongi Italian”, which is separated by a space that is a delimiter between the characters “Italian” indicating the type of food, has similar characteristics.

  Further, the generation unit 133 generates the first learning model by learning as the search queries acquired by the acquisition unit 131 that the plurality of randomly extracted search queries have different characteristics. Specifically, the generation unit 133 generates a first learning model by learning so that the distributed expressions of a pair of search queries that are randomly extracted among the search queries acquired by the acquisition unit 131 are different. To do. For example, the generation unit 133 causes the distributed expression of the predetermined search query extracted by the extraction unit 132 and the distributed expression of the search query randomly extracted irrespective of the predetermined search query to be far apart in the distributed expression space. The first learning model M1 is trained so as to be mapped.

  In addition, the generation unit 133 generates the second learning model. Specifically, the generation unit 133 refers to the model information storage unit 125 and acquires the first learning model (model data MDT1 of the first learning model M1) generated by the generation unit 133. Subsequently, the generation unit 133 uses the acquired first learning model to generate a second learning model that predicts a category to which the predetermined search query belongs from the predetermined search query. When the generation unit 133 acquires the first model M1, the generation unit 133 generates the second learning model M2 using the acquired first model M1. The generation unit 133 re-learns the first model M1 to generate the second model M2 having a connection coefficient which is a weight of the learning model from the first model M1. Specifically, when the search query is input to the learning model, the generation unit 133 learns such that the classification result of the distributed expressions output by the learning model corresponds to the category to which the search query belongs, and thus the predetermined result is obtained. A second model M2 that predicts a category to which a predetermined search query belongs is generated from the search query of.

  Specifically, when the search query is input to the learning model, the generation unit 133 learns such that the classification result of the distributed expressions output by the learning model corresponds to the category to which the search query belongs, and thus the predetermined result is obtained. A second learning model that predicts a category to which a predetermined search query belongs is generated from the search query. The generation unit 133 generates a second learning model that outputs the probability of each category to which the search query belongs as output information when the search query is input as input information to the learning model. For example, the generation unit 133 uses the first model M1 to calculate the probability that the distributed expression of the search query is classified as the output information into the category when the predetermined search query is input as the input information into the learning model. A second model M2 to be output for each is generated. When a predetermined search query is input as the input information, the generation unit 133 trains the second model so that the probability that the distributed expression of the predetermined search query is classified as the correct category as the output information exceeds the predetermined threshold. . Then, when a predetermined search query is input as the input information, the generation unit 133 sets a category whose probability that the distributed expression of the predetermined search query belongs to the category exceeds a predetermined threshold as the category of the predetermined search query. The second model M2 to be output is generated. Further, when the generation unit 133 generates the second learning model, the generation unit 133 stores the generated second learning model (model data MDT2) in the model information storage unit 125 in association with the identification information for identifying the second learning model.

  For example, the generation unit 133 refers to the model information storage unit 125 illustrated in FIG. 11 and acquires the first model M1 (model data MDT1 of the first model M1). Subsequently, the generation unit 133 refers to the classification definition storage unit 123 illustrated in FIG. 9 and selects a large classification of categories into which the search query is classified. Subsequently, when selecting the large classification, the generation unit 133 learns a set of the search query and the small classification to which the search query belongs as the learning data of the second model M2.

  For example, it is assumed that the correct category to which the search query Q11 (“Roppongi pasta”) belongs is CAT11 (“Find a restaurant”). When the search query Q11 (“Roppongi pasta”) is input to the second model M2 as input information, the generation unit 133 uses the distributed representation of the search query Q11 (“Roppongi pasta”) from the output layer of the second model M2. It outputs a certain vector BQV11. Here, the vector BQV11 is a distributed expression of the search query Q11 just output from the output layer of the second model M2, and represents a distributed expression before feedback is applied to the second model M2 (before learning). In this case, the generation unit 133 causes the probability that the vector BQV11, which is the distributed expression of the output search query Q11 (“Roppongi pasta”), is classified into the correct category CAT11 (“Find a restaurant”) exceeds a predetermined threshold. Thus, the second model M2 is trained.

  For example, when the search query Q11 (“Roppongi pasta”) is input to the second model M2 before learning, the generating unit 133 classifies the vector BQV11, which is a distributed expression, into CAT11 (“Find a restaurant”). 80% probability, 0% probability of being classified as CAT12 (“Search for a product”), 20% probability of being classified as CAT13 (“Reserve a restaurant”), CAT14 (“Purchase a product”) It is assumed that the probability of being classified as is output as 0%. In this case, the generation unit 133 trains the second model M2 so that the probability that the vector BQV11, which is a distributed expression, is classified into CAT11 (“search for a restaurant”) exceeds a predetermined threshold value (for example, 90%). . In addition, the generation unit 133 performs learning so that the probability that the vector BQV11, which is a distributed expression, is classified into CAT11 (“search for a restaurant”) exceeds a predetermined threshold value (for example, 90%), and the distribution is performed. The second model M2 is trained to reduce the probability that the expression vector BQV11 is classified into another category CAT13 (“reserve restaurant”) to 10%. Subsequently, when the search query Q11 (“Roppongi pasta”) is input as input information to the learned second model M2, the generation unit 133 generates a vector BQV11 that is a distributed expression of the search query Q11 (“Roppongi pasta”). Belongs to the category CAT11 (“search for restaurants”) exceeds 90%, the category to which the search query belongs is output as CAT11 (“searches for restaurants”) as output information.

  The generation unit 133 may select an arbitrary number of large categories as the large categories. Then, when the search query is input to the second model M2 as the input information, the generation unit 133 determines the probability of belonging to each of the small categories belonging to the arbitrary number of the large categories selected by the search query as the output information for each of the small categories. The second model M2 to be output to may be generated. Further, the generation unit 133 may select all the large classifications as the large classifications. Then, the generation unit 133 may generate the second model M2 that outputs the probability of belonging to each of the small classifications for every small classification when the search query is input to the second model M2.

[Example of Second Learning Model]
Here, an example of the second learning model generated by the information processing apparatus 100 will be described with reference to FIG. 13. FIG. 13 is a diagram illustrating an example of the second learning model according to the embodiment. In the example illustrated in FIG. 13, the second learning model M2 generated by the information processing device 100 is generated using the first learning model M1. That is, the information processing apparatus 100 re-learns the first learning model M1 to generate a second learning model M2 having a connection coefficient that is a weight of the learning model different from that of the first learning model M1.

  More specifically, the second learning model M2 generated by the information processing apparatus 100 is composed of three layers of LSTM RNNs, like the first learning model M1. In the example illustrated in FIG. 13, the extraction unit 132 inputs the search query Q11 “Roppongi pasta” input by the user U1 into the input layer of the second learning model M2 (step S51).

  Subsequently, the generation unit 133 outputs the 256-dimensional vector BQV11, which is the distributed expression of the search query Q11, from the output layer of the second learning model M2 (step S52).

  Subsequently, the generation unit 133 outputs the probability that the vector BQV11, which is the distributed expression of the search query Q11, is classified into each category (step S53).

  Then, the generation unit 133 learns the second learning model M2 so as to increase the probability that the vector BQV11, which is the distributed expression of the search query Q11, is classified into the correct answer category, thereby changing the category of the search query from the search query. A second model to be predicted is generated (step S54).

(Service providing unit 134)
The service providing unit 134 acquires the first learning model. Specifically, the service providing unit 134 refers to the model information storage unit 125 and acquires the first learning model generated by the generation unit 133. More specifically, the service providing unit 134 first learns the characteristics of the plurality of search queries, assuming that the plurality of search queries input by the same user within a predetermined time have similar characteristics. Get the model. For example, the service providing unit 134 learns, by the generating unit 133, that a plurality of search queries input by the same user within a predetermined time period have similar characteristics, so that a predetermined search query is performed. A first learning model generated as a learning model for predicting query feature information is acquired. Also, the service providing unit 134 acquires the first learning model that outputs the distributed expression of the predetermined search query as the output information when the predetermined search query is input as the input information. In addition, the service providing unit 134 learns the first learning model that has learned the characteristics of a plurality of search queries by learning so that the distributed expressions of a pair of search queries that are continuously input within a predetermined time are similar. get. In addition, the service providing unit 134 determines that a plurality of search queries including a character string delimited by a predetermined delimiter are similar to each other as a plurality of search queries input by the same user within a predetermined time. By learning, the 1st learning model which learned the characteristic which a some search query has is acquired. In addition, the service providing unit 134 acquires the first learning model in which the characteristics of the plurality of search queries are learned by learning that the plurality of randomly extracted search queries have different characteristics. In addition, the service providing unit 134 acquires the first learning model in which the characteristics of the plurality of search queries are learned by learning so that the distributed expressions of the pair of search queries that are randomly extracted are different.

  In addition, the service providing unit 134 acquires the second learning model. Specifically, the service providing unit 134 refers to the model information storage unit 125 and acquires the second learning model generated by the generation unit 133. More specifically, the service providing unit 134 uses the first learning model by the generating unit 133 to generate a second learning model as a learning model that predicts a category to which a predetermined search query belongs from a predetermined search query. To get. For example, when the search query is input to the learning model, the service providing unit 134 learns so that the classification result of the distributed expression output by the learning model corresponds to the category to which the search query belongs, thereby performing a predetermined search. A second learning model that predicts a category to which a predetermined search query belongs is acquired from the query. For example, when the search query is input to the learning model as the input information, the service providing unit 134 acquires, as the output information, the second learning model that outputs the probability that the search query belongs to the category for each category.

  The service providing unit 134 provides a search query analysis service to a client who is an advertiser. The service providing unit 134 determines that a plurality of search queries input by the same user within a predetermined time have similar features, and uses the first learning model that has learned the features of the plurality of search queries to determine a predetermined value. A similar query, which is a search query having characteristics similar to the search query of, is extracted. Specifically, the service providing unit 134 receives the analysis theme of the search query from the client who is the advertiser, and extracts the search query corresponding to the received analysis theme. For example, the service providing unit 134 receives, from the client terminal 20, an analysis query that is a keyword related to the analysis theme that analyzes the search trend of the user. The service providing unit 134 may receive a plurality of analysis queries regarding analysis themes. For example, assume that a client who is an advertiser desires "bone health-related" as an analysis theme of a search query. Then, the service providing unit 134 receives from the client terminal 20 a plurality of analysis queries such as “bone health”, “bone strength”, and “calcium small fish”. As a result, the service providing unit 134 can extract more similar queries than when receiving one analysis query at a time. That is, the service providing unit 134 can extract all the similar queries in line with the analysis theme of the client who is the advertiser. Then, the service providing unit 134 extracts the search query corresponding to the accepted analysis theme. For example, the service providing unit 134 uses the first learning model to extract a similar query that is a search query having characteristics similar to the analysis query.

  In addition, the service providing unit 134 uses the second learning model to classify similar queries that are search queries having characteristics similar to the predetermined search query. Specifically, the service providing unit 134 receives the classification axis of the search query from the client who is the advertiser, and classifies the search query into the category corresponding to the received classification axis. For example, the service providing unit 134 receives, from the client terminal 20, a classification axis (category) that classifies a similar query that is a search query similar to the analysis query. Note that the service providing unit 134 may receive a plurality of classification axes (categories) for one set of similar queries to be classified. Subsequently, the service providing unit 134 uses the second learning model to classify the similar queries into categories corresponding to the accepted classification axis.

  In the example illustrated in FIG. 1, the service providing unit 134 receives, from the client terminal 20, the analysis query “bone health” and a classification axis (category) including four small classifications. The service providing unit 134 selects a category axis (category) designated by the client who is the advertiser from the categories shown in the category definition storage unit 123 shown in FIG. 9 by the client who is the advertiser. The classified large classification may be accepted as the classification axis. In the example illustrated in FIG. 1, the client who is the advertiser selects the large category “health system” identified by the large category ID “CAT2” from the categories shown in the category definition storage unit 123 described later. Then, the service providing unit 134 receives the large category “health system” selected by the client who is the advertiser as the classification axis.

  Subsequently, when the service providing unit 134 receives the analysis query, the service providing unit 134 inputs the received analysis query “bone health” into the first model M1. When the service providing unit 134 inputs the analysis query to the first model M1, the service providing unit 134 outputs the distributed expression (vector) of the analysis query “bone health” from the first model M1.

  Next, when the service providing unit 134 outputs the distributed expression (vector) of the analysis query “bone health”, the search query whose similarity between the distributed expressions (vectors) exceeds a predetermined threshold is set as the analysis query “bone health”. Extract as a similar query. For example, the service providing unit 134 extracts a distributed expression (vector) located near the distributed expression (vector) of the output analysis query “bone health” in the distributed expression space. Subsequently, the service providing unit 134 calculates the similarity between the distributed expression (vector) of the analysis query “bone health” and the distributed expression (vector) located in the vicinity. For example, the service providing unit 134 calculates the cosine similarity between the distributed expressions (vectors). Subsequently, the service providing unit 134 extracts a distributed expression (vector) whose cosine similarity exceeds a predetermined threshold as a distributed expression (vector) similar to the distributed expression (vector) of the analysis query “bone health”. Note that the service providing unit 134 calculates the similarity between the distributed expressions (vectors) based on what kind of index, not limited to the cosine similarity, as long as it is an index applicable as a distance measure between vectors. Good. For example, the service providing unit 134 calculates a value of a predetermined distance function such as a Euclidean distance between distributed expressions (vectors), a distance in a non-Euclidean space such as a hyperbolic space, a Manhattan distance, a Mahalanobis distance, or the like. Next, the service providing unit 134 analyzes the distributed expression (vector) in which the value of the predetermined distance function (that is, the distance in the distributed expression space) between the distributed expressions (vectors) is less than the predetermined threshold value in the analysis query “bone health”. It may be extracted as a distributed expression (vector) similar to the distributed expression (vector). Subsequently, when the service providing unit 134 extracts a similar distributed expression (vector), the service providing unit 134 refers to the vector information storage unit 122 and specifies a search query corresponding to the extracted distributed expression (vector). Then, the service providing unit 134 extracts the identified search query as a similar query similar to the analysis query.

  Subsequently, when the service providing unit 134 extracts the similar query, the service providing unit 134 inputs the extracted similar query into the second model M2. When the similar query is input to the second model M2, the service providing unit 134 outputs the category to which the similar query belongs from the second model M2. The service providing unit 134 receives the classification axis (category) specified by the client who is the advertiser, and then learns the second model M2 so as to classify the search queries according to the received classification axis (category). You may let me. Alternatively, the service providing unit 134 preliminarily trains the second model M2 to classify the search queries according to the categories shown in the classification definition storage unit 123 described later. Then, the service providing unit 134 causes the client who is the advertiser to select the classification axis from the categories shown in the classification definition storage unit 123, and the classification axis (category) selected by the client who is the advertiser is selected from the client terminal 20. You may accept.

  Subsequently, the service providing unit 134 transmits the classification result of the query according to the classification axis (category) to the client terminal 20. For example, the service providing unit 134 provides, as the classification result of the query according to the classification axis (category), the configuration of the search query searched as the search query related to the analysis theme “bone health related”. For example, as a breakdown of the search queries extracted as the similar queries similar to the analysis query, the service providing unit 134 has a search query rate of “drink” of 40%, a search query rate of “food” of 30%, and “exercise”. The analysis result shows that the ratio of the search query regarding “” is 20% and the ratio of the search query regarding “supplement” is 10%.

[3. Flow of processing for generating first learning model]
Next, the procedure of the first learning model generation process according to the embodiment will be described with reference to FIG. 14. FIG. 14 is a flowchart showing a procedure for generating the first learning model according to the embodiment. In the example illustrated in FIG. 14, the information processing device 100 acquires the search query input by the user (step S101).

  Subsequently, the information processing apparatus 100 extracts a plurality of search queries input by the same user within a predetermined time (step S102).

  Subsequently, the information processing apparatus 100 learns that the extracted plurality of search queries have similar features, thereby generating a first learning model that predicts the feature information of the predetermined search query from the predetermined search query. (Step S103).

[4. Flow of second learning model generation processing]
Next, the procedure of the second learning model generation process according to the embodiment will be described with reference to FIG. 15. FIG. 15 is a flowchart showing the procedure of the second learning model generation process according to the embodiment. In the example illustrated in FIG. 15, the information processing device 100 acquires a first learning model (model data MDT1 of the first learning model M1) (step S201).

  Subsequently, the information processing apparatus 100 uses the first learning model to generate a second learning model that predicts a category of a predetermined search query from a predetermined search query (step S202).

[5. Information processing flow]
Next, a procedure of information processing according to the embodiment will be described with reference to FIG. FIG. 16 is a flowchart showing a procedure of information processing according to the embodiment. In the example illustrated in FIG. 16, the information processing device 100 acquires the analysis query and the classification axis (category) from the client (step S301). Subsequently, the information processing apparatus 100 uses the first model to extract a similar query similar to the analysis query (step S302). Subsequently, the information processing apparatus 100 classifies the similar queries into classification axes (categories) using the second model (step S303). Then, the information processing apparatus 100 provides the analysis result to the client (step S304).

[6. effect〕
As described above, the information processing device 100 according to the embodiment includes the service providing unit 134. The service providing unit 134 determines that a plurality of search queries input by the same user within a predetermined time have similar features, and uses the first learning model that has learned the features of the plurality of search queries to determine a predetermined value. A similar query, which is a search query having characteristics similar to the search query of, is extracted.

  In general, a user is considered to perform a search a plurality of times with a certain intention, and therefore it is assumed that search queries continuously input within a predetermined time have similar search intentions. Therefore, the information processing apparatus 100 according to the present invention is similar to each other in that a plurality of search queries continuously input within a predetermined time period are search queries searched under a predetermined search intention. The learning model M1 is trained on the assumption that the search query has a characteristic. Specifically, the information processing apparatus 100 according to the present invention learns the learning model M1 so that the distributed expressions of the search queries continuously input within a predetermined time period are similar to each other, so that the search intention is similar. The distributed expression of a query can be output to a close position on the distributed expression space. This allows the information processing apparatus 100 to output (interpret) the meaning (search intent) of the search query according to the context of the user who inputs the search query. Therefore, the information processing apparatus 100 can appropriately interpret the meaning of the search query. Furthermore, the information processing apparatus 100 extracts a search query corresponding to a distributed expression that is mapped in the vicinity of the distributed expression that includes the characteristic information of the predetermined search query, so that the predetermined search query is searched according to the search intention. Search queries can be extracted. That is, the information processing apparatus 100 enables extraction of a search query in consideration of the context of the user who inputs the search query. Therefore, the information processing apparatus 100 can analyze the search trend of the user by using the first learning model in consideration of the context of the user who inputs the search query. Therefore, the information processing apparatus 100 can improve the accuracy of analysis of the search trend of the user.

  Further, the learning model M1 generated by the information processing apparatus 100 can be made to function as a part of a search system (for example, a search system used for a search query analysis service provided by the information processing apparatus 100). Alternatively, the information processing apparatus 100 inputs to another system that uses the characteristic information of the search query predicted by the learning model M1 (for example, a search engine used for a search query analysis service provided by the information processing apparatus 100). The information can also be provided as a distributed representation of the search query output by the learning model M1. As a result, the search system can select the content output as the search result based on the characteristic information of the search query predicted by the learning model M1. That is, the search system can select the content output as the search result in consideration of the search intention and context of the user who inputs the search query. Furthermore, the search system can calculate the degree of similarity between the distributed expression of the character string included in the content output as the search result and the distributed expression of the search query, based on the characteristic information of the search query predicted by the learning model M1. become. Then, the search system can determine the display order of the content output as the search result based on the calculated similarity. That is, the search system can determine the display order of the content output as the search result in consideration of the search intention and the context of the user who inputs the search query. Therefore, the information processing apparatus 100 can improve usability in the search service.

  Also, the service providing unit 134 acquires the first learning model that outputs the distributed expression of the predetermined search query as the output information when the predetermined search query is input as the input information.

  With this, the information processing apparatus 100 can measure the abstract concept of the characteristics of a predetermined search query by a specific numerical value of the distributed expression.

  In addition, the service providing unit 134 learns the first learning model that has learned the characteristics of a plurality of search queries by learning so that the distributed expressions of a pair of search queries that are continuously input within a predetermined time are similar. get.

  Generally, it is considered that two search queries input by the same user in a short period of time have the same search intention, or have similar search intentions even if they are not the same. That is, it is considered that the pair of search queries that are continuously input within a predetermined time have the same search intention, or even if they are not the same, the search intention is close. That is, the information processing apparatus 100 can more accurately extract a search query according to the user's search intention by learning so that the distributed expressions of a pair of search queries that are continuously input within a predetermined time are similar. Can be

  In addition, the service providing unit 134 determines that a plurality of search queries including a character string delimited by a predetermined delimiter are similar to each other as a plurality of search queries input by the same user within a predetermined time. By learning, the 1st learning model which learned the characteristic which a some search query has is acquired.

  In general, a search query including a plurality of character strings is considered to have a clearer search intention than a search query including a single character string. That is, the information processing apparatus 100 can more accurately extract the search query according to the user's search intention by learning by using the search query including the character string delimited by the predetermined delimiter. .

  In addition, the service providing unit 134 acquires the first learning model in which the characteristics of the plurality of search queries are learned by learning that the plurality of randomly extracted search queries have different characteristics. In addition, the service providing unit 134 acquires the first learning model in which the characteristics of the plurality of search queries are learned by learning so that the distributed expressions of the pair of search queries that are randomly extracted are different.

  In general, a plurality of search queries that are randomly extracted are search queries that are input independently of each other, and thus it is considered that the search intentions are different or the search intentions are far. Therefore, the information processing apparatus 100 according to the present invention is a search query having mutually different characteristics in that the plurality of randomly extracted search queries are search queries searched under different search intentions. The learning model M1 is learned as if it exists. Specifically, the information processing apparatus 100 according to the present invention trains the learning model M1 so that the distributed expressions of a plurality of search queries that are randomly extracted are different, and thus the distributed expressions of the search queries having different search intentions. Can be output to a distant position in the distributed representation space. Thus, the learning model can learn incorrect answer data, which is a pair of search queries having a distant search intention, in addition to correct answer data, which is a pair of search queries having a close search intention. That is, the information processing apparatus 100 can improve the accuracy of the learning model. Therefore, the information processing apparatus 100 can appropriately interpret the meaning of the search query.

  The service providing unit 134 also acquires a second learning model that is a second learning model generated using the first learning model and that predicts a category to which a predetermined search query belongs from a predetermined search query.

  As a result, the information processing apparatus 100 efficiently utilizes the second learning model that predicts the category of the predetermined search query from the predetermined search query by utilizing the first learning model that has learned the features of the search query in consideration of the search intention. Can be generated dynamically. As a result, the information processing apparatus 100 enables the search query to be classified into categories in consideration of the search intention of the user who inputs the search query. Further, conventionally, it is necessary to prepare a sufficient amount of correct answer data in order to classify search queries into categories and obtain high classification accuracy. However, since the search queries themselves are diverse and have a long tail property, it is very troublesome and difficult to label the correct answer categories corresponding to a large number of search queries. Here, the information processing apparatus 100 learns the second model that predicts the category of the search query by using the user's search intention (the context of the user who inputs the search query) as a kind of correct answer, instead of labeling the correct answer category. Can be made. With this, the information processing apparatus 100 can train the second model without manually labeling the correct category of the search query. That is, the information processing apparatus 100 can obtain sufficient classification accuracy even when the correct answer data is small. In addition, the information processing apparatus 100 can obtain higher classification accuracy when there are many correct data. Therefore, the information processing device 100 can improve the classification accuracy of the search query.

  Further, when the search query is input to the learning model, the service providing unit 134 learns such that the classification result of the distributed expression output by the learning model corresponds to the category to which the search query belongs, thereby performing a predetermined search. A second learning model that predicts a category to which a predetermined search query belongs is acquired from the query. Further, when the search query is input to the learning model as the input information, the service providing unit 134 acquires, as the output information, the second learning model that outputs the probability that the search query belongs to the category for each category.

  Thereby, the information processing apparatus 100 utilizes the distributed expression including the characteristics of the search query in consideration of the search intention, and classifies the search query into a category in which the search intention of the user who inputs the search query is considered. Can be efficiently generated. That is, the information processing apparatus 100 enables the search queries to be classified into categories in consideration of the search intention of the user who inputs the search query. Therefore, the information processing device 100 can improve the classification accuracy of the search query.

  In addition, the service providing unit 134 uses the second learning model to classify similar queries that are search queries having characteristics similar to the predetermined search query.

  As a result, the information processing apparatus 100 enables the search query to be classified into categories in consideration of the search intention of the user who inputs the search query. Therefore, the information processing device 100 can improve the classification accuracy of the search query.

  Further, the service providing unit 134 receives the analysis theme of the search query from the client who is the advertiser, and extracts the search query corresponding to the received analysis theme.

  As a result, the information processing apparatus 100 can appropriately provide the analysis result of the search trend of the user that matches the analysis theme desired by the client who is the advertiser.

  Further, the service providing unit 134 receives the classification axis of the search query from the client who is the advertiser, and classifies the search query into the category corresponding to the received classification axis.

  As a result, the information processing apparatus 100 can appropriately provide the analysis result of the search trend of the user along the classification axis desired by the client who is the advertiser.

[7. Hardware configuration)
Further, the information processing apparatus 100 according to the above-described embodiment is realized by, for example, a computer 1000 configured as shown in FIG. FIG. 17 is a hardware configuration diagram illustrating an example of a computer that realizes the functions of the information processing device 100. The computer 1000 includes 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 unit. 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 a predetermined communication network, sends the data to the CPU 1100, and transmits the data generated by the CPU 1100 to another device via the predetermined communication network.

  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 obtains data from an 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) or 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 information processing apparatus 100 according to the embodiment, the CPU 1100 of the computer 1000 has a program or data loaded on the RAM 1200 (for example, model data MDT1 of the first model M1 and model data MDT1 of the second model M2). The function of the control unit 130 is realized by executing the model data MDT2). The CPU 1100 of the computer 1000 reads these programs from the recording medium 1800 and executes them, but as another example, these programs or data (for example, the model of the first model M1 from another device via a predetermined communication network). The data MDT1 and the model data MDT2 of the second model M2) may be acquired.

  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, including the modes described in the section 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-described embodiment and modified examples, all or part of the processes described as being automatically performed may be manually performed, or described as manually performed. It is also possible to automatically carry out all or part of the processing performed by a known method. In addition, the processing procedures, specific names, and information including 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 information shown.

  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 a 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.

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

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

1 Information processing system 10 User terminal 20 Client terminal 50 Search server 100 Information processing device 121 Query information storage unit 122 Vector information storage unit 123 Classification definition storage unit 124 Category information storage unit 125 Model information storage unit 131 Acquisition unit 132 Extraction unit 133 Generation Department 134 Service Department

Claims (14)

A plurality of analysis queries that are keywords related to the analysis theme of the search query and analysis categories that classify similar queries that are search queries similar to the analysis query are acquired from the client, and are input by the same user within a predetermined time period. As a search query having similar features, a similar query that is a search query having features similar to the analysis query is extracted using a first learning model that has learned the features of the plurality of search queries, The second learning model generated using the first learning model, the second learning model predicting a category to which the predetermined search query belongs from a predetermined search query, the extracted similar query is analyzed. For each analysis category, which is calculated based on the number of similar queries classified into the analysis category. The information processing apparatus comprising: a service providing unit that provides information about the component ratio of queries to the client similar. The service providing unit,
The information processing apparatus according to claim 1, wherein, when a predetermined search query is input as input information, a first learning model that outputs a distributed expression of the predetermined search query is output as output information. The service providing unit,
A first learning model in which the features of the plurality of search queries are learned by learning so that the distributed expressions of a pair of search queries that are continuously input within the predetermined time period are similar to each other. The information processing device according to claim 1 or 2 . The service providing unit,
By learning as a plurality of search queries input by the same user within a predetermined time, a plurality of search queries including a character string delimited by a predetermined delimiter have similar characteristics, Search query information processing apparatus according to any one of claims 1-3, characterized in that obtaining the first learning model trained features included in. The service providing unit,
By learning as a plurality of search queries that are extracted at random has a characteristic which differs, according to claim 1-4, characterized in that obtaining the first learning model learned features the plurality of search queries has The information processing apparatus described in any one of 1. The service providing unit,
By distributed representation of a pair of search queries extracted randomly learns to differences claims, characterized in that obtaining the first learning model learned features the plurality of search queries has 1-5 The information processing apparatus described in any one of 1. The service providing unit,
The second learning model generated by using the first learning model, wherein a second learning model that predicts a category to which the predetermined search query belongs is obtained from a predetermined search query. The information processing apparatus according to any one of items 1 to 6 . The service providing unit,
When the search query is input to the learning model, the classification result of the distributed expression output by the learning model is learned so as to correspond to the category to which the search query belongs, so that the predetermined search query is changed from the predetermined search query. the information processing apparatus according to any one of claims 1-7, characterized in that to obtain the second learning model that predicts the category to which belongs. The service providing unit,
When a search query as input information is input to the training model, the search query according to claim 1-8, characterized in that to obtain the second learning model for outputting the probability that belong to the category in each category as output information The information processing device described in any one of the above. The service providing unit,
The second with a learning model, the information processing apparatus according to any one of claims 7-9, characterized in that classifying the similar query is a search query having characteristics similar to the predetermined search query. The service providing unit,
Ad accepts Lord is analyzed themes search query from the client, the information processing apparatus according to any one of claims 1-10, characterized in that to extract the search query corresponding to the accepted analysis theme. The service providing unit,
Advertiser in which accepts classification axis search query from the client, the information processing apparatus according to any one of claims 1 to 11, characterized in that classifying the search query to the category corresponding to the received classification axis. An information processing method executed by a computer,
A plurality of analysis queries that are keywords related to the analysis theme of the search query and analysis categories that classify similar queries that are search queries similar to the analysis query are acquired from the client, and are input by the same user within a predetermined time period. As a search query having similar features, a similar query that is a search query having features similar to the analysis query is extracted using a first learning model that has learned the features of the plurality of search queries, The second learning model generated using the first learning model, the second learning model predicting a category to which the predetermined search query belongs from a predetermined search query, the extracted similar query is analyzed. For each analysis category, which is calculated based on the number of similar queries classified into the analysis category. The information processing method characterized by information about the component ratio of the similar queries including service providing step of providing to the client. A plurality of analysis queries that are keywords related to the analysis theme of the search query and analysis categories that classify similar queries that are search queries similar to the analysis query are acquired from the client, and are input by the same user within a predetermined time period. As a search query having similar features, a similar query that is a search query having features similar to the analysis query is extracted using a first learning model that has learned the features of the plurality of search queries, The second learning model generated using the first learning model, the second learning model predicting a category to which the predetermined search query belongs from a predetermined search query, the extracted similar query is analyzed. For each analysis category, which is calculated based on the number of similar queries classified into the analysis category. The information processing program characterized by executing the service providing means for providing information about the component ratio of queries to the client similar to a computer.

Images