JP2020047126A - Information processing apparatus, information processing method, and information processing program - Google Patents

Abstract

To improve the accuracy in analyzing the tendency of a user's search.SOLUTION: An information processing apparatus according to the present application comprises a service providing unit. Provided that a plurality of search queries input by a single user within a predetermined time have similar features, the service providing unit extracts a similar query that is a search query having a feature similar to that of a predetermined search query, by using a first learning model that learns the features of the plurality of search queries.SELECTED DRAWING: Figure 1

Description

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

  Conventionally, a technique for extracting information characteristic of a specific user has been known. Specifically, information on the search query used by the target user is classified into a first cluster based on the similarity between the vectors corresponding to the information on each search query used by the target user, and each search query of another user is classified. The information on the search query used by another user is classified into the second cluster based on the similarity between the vectors corresponding to the information on the search query. A technique has been proposed for extracting a feature cluster, which is a cluster showing a characteristic behavior of the target user, from the first cluster based on a difference between the first cluster and the second cluster.

JP 2018-60469 A

  However, in the above-described conventional technology, it is not always possible to improve the analysis accuracy of a user's search trend. Specifically, the meaning of a word may vary depending on the context in which the word is used. Therefore, even if the character string used in the context is extracted separately from the context, it is difficult to interpret the original meaning of the character string (that is, the meaning of the character string used in the context). There is. In view of the above, in the above-described related art, it is difficult to interpret the original meaning of the search query because only the similarity of the search query as a character string is considered. Therefore, it is difficult to analyze a user's search tendency in consideration of the context of the user who has input the search query in the above-described related art. Therefore, in the above-described conventional technology, it is not always possible to improve the analysis accuracy of the search trend of the user.

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

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

  According to an aspect of the embodiment, there is an effect that analysis accuracy of a search trend of a user can be improved.

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 process of generating a first learning model according to the embodiment. FIG. 3 is a diagram illustrating an example of a process of generating a first learning model according to the embodiment. FIG. 4 is a diagram illustrating an example of a process of generating a second learning model 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 apparatus according to the embodiment. FIG. 7 is a diagram illustrating an example of a 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 a 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 illustrating the generation processing procedure of the first learning model according to the embodiment. FIG. 15 is a flowchart illustrating the generation processing procedure of the second learning model according to the embodiment. FIG. 16 is a flowchart illustrating an 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, an information processing apparatus, an information processing method, and an embodiment (hereinafter, referred to as “embodiment”) for implementing an information processing program according to the present application will be described in detail with reference to the drawings. The information processing apparatus, the information processing method, and the information processing program according to the present application are not limited by the embodiment. In the following embodiments, the same portions are denoted by the same reference numerals, and overlapping 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 illustrated in FIG. 1 is performed by the user terminal 10, the client terminal 20, the search server 50, and the information processing device 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 device 100 are communicably connected 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. Is also good.

  The user terminal 10 is an information processing device used by a user who uses a 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 in some cases. That is, the user can be replaced with the user terminal 10 below.

  Hereinafter, the user specified by the user ID “U1” may be referred to as “user U1”. Thus, hereinafter, when "user U * (* is an arbitrary numerical value)" is described, it indicates that the user is a user specified by the user ID "U *". For example, when "user U2" is described, the user is a user specified by the user ID "U2".

  Hereinafter, the user terminal 10 will be described 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. In the following, the user terminals 10-1 and 10-2 are referred to as the user terminals 10 in the case where they are 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, when an operation of transmitting a search query is performed following an operation in which a character is input to the search box by the user, the user terminal 10 uses the character input in the search box via the search page as a search query. Send it to the search server 50. For example, when the user performs an operation of pressing a send button of a search query or an operation of pressing an enter key following an operation of inputting a character in a search box by a user, the user terminal 10 transmits the search query via a search page. The character entered in the search box is transmitted to the search server 50 as a search query.

  The client terminal 20 is an information processing device used by a client as 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) in some cases. That is, hereinafter, the client (advertiser) can be read as the client terminal 20.

  The client terminal 20 transmits, to the information processing apparatus 100, an analysis query that is a keyword related to an analysis theme for analyzing a user's search trend. Specifically, the client terminal 20 transmits a plurality of analysis queries to the information processing device 100. For example, suppose that a client who is an advertiser wants “bone health-related” as a search query analysis theme. More specifically, it is assumed that the client who is the advertiser intends “bone health”, “bone strong”, “calcium fish”, etc. as keywords related to the analysis theme. In this case, the client terminal 20 transmits a plurality of analysis queries such as “bone health”, “bone strength”, and “calcium fish” to the information processing apparatus 100 according to the operation of the client as the advertiser. Further, the client terminal 20 transmits to the information processing apparatus 100 information on a classification axis (category) for classifying a similar query that is a search query similar to the analysis query. For example, the client terminal 20 transmits the analysis query and the information on the classification axis (category) to the information processing apparatus 100 through a 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 a search query from the user terminal 10, the search server 50 selects content corresponding to the received search query and output as a search result. Subsequently, the search server 50 distributes a 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 a web page displayed by a web browser. For example, the content distributed by the search server 50 may be content displayed by a dedicated application installed on the user terminal 10. The contents distributed by the search server 50 include music contents, images (including moving images as well as still images), and text contents (including news articles and articles posted on SNS (Social Networking Service)). Any content such as content combining an image and text, game content, and the like may be used.

  Further, the search server 50 stores information related to the search query input by the user. Specifically, the search server 50 stores information on 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, a user ID for identifying a user who is the source of the search query, and a transmission date and time of the search query in association with each other in the database. . The search server 50 transmits information on the search query input by the user to the information processing device 100 in response to a request from the information processing device 100.

  The information processing device 100 is a server device that provides a search query analysis service to a client as 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 for analyzing a search trend of the user. Note that the information processing device 100 may receive a plurality of analysis queries on the analysis theme. For example, suppose that a client who is an advertiser wants “bone health-related” as a search query analysis theme. Then, the information processing apparatus 100 receives a plurality of analysis queries such as “bone health”, “bone strength”, and “calcium small fish” from the client terminal 20. Thus, 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 completely extract similar queries according to the analysis theme of the client as the advertiser. Subsequently, the information processing device 100 extracts a search query corresponding to the received 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 later. Here, the first learning model means that a plurality of search queries input by the same user within a predetermined time period are learned as having similar characteristics, so that a predetermined search query This is a learning model for predicting feature information. Hereinafter, the first learning model is referred to as a first model (or a first model M1) as appropriate.

  Further, the information processing apparatus 100 receives, from the client terminal 20, a classification axis (category) for classifying a similar query that is a search query similar to the analysis query. Note that the information processing apparatus 100 may accept a plurality of classification axes (categories) for a set of one similar query to be classified. Subsequently, the information processing device 100 classifies the similar query into a category corresponding to the received classification axis. For example, the information processing apparatus 100 classifies the similar query into a category corresponding to the received classification axis using a 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 is appropriately described as a second model (or a second model M2).

  Further, 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 a plurality of search queries input by the same user within a predetermined time are similar, so that the predetermined search query A first learning model for predicting feature information of a predetermined search query is generated. Here, the information processing apparatus 100 learns the entire character string input to the search box as a search query input by the user for each search performed by the user. For example, when a search query including a plurality of character strings such as “Roppongi pasta” is input to the search box in a single 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 such that an interval of time when each search query is input by the same user is within a predetermined time (for example, within 2 minutes) by the same user for a predetermined time. 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 quick succession have similar characteristics. Note that, when a predetermined search query is input and another search query is input after a predetermined time (for example, after two minutes or more), the predetermined search query and the predetermined search query are input. Learning is performed as having different characteristics from other search queries.

  In general, when a searcher performs a search, the searcher performs a search multiple times using different search queries, rather than the case where the searcher reaches the information intended by the searcher, and the searcher's intention It is thought that there are more cases where information is reached. That is, since it is considered that the user performs a search a plurality of times with a certain intention, it is presumed that a search query continuously input within a predetermined time has a similar search intention. Therefore, the information processing apparatus 100 according to the present invention learns 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 on the search query input by the user from the search server 50. Subsequently, the information processing apparatus 100 extracts a plurality of search queries input by the same user within a predetermined time from the search queries acquired from the search server 50. Subsequently, the information processing apparatus 100 generates a first learning model that predicts feature information of a predetermined search query from a predetermined search query by learning as a plurality of extracted search queries having similar characteristics. . For example, the information processing apparatus 100 trains the first learning model so that the distributed expressions of the plurality of extracted search queries are similar to each other, so that the distributed expression (vector ) Is generated.

  More specifically, the information processing apparatus 100 uses a DSSM (Deep Structured Sematic Model) technique that uses LSTM (Long Short-Term Memory), which is a type of RNN (Recurrent Neural Network), for generating a distributed expression. A first learning model that outputs a distributed expression (vector) from the search query is generated. For example, the information processing apparatus 100 determines that a pair of search queries input by the same user within a predetermined time period have similar characteristics as the correct answer data of the first learning model, and the distributed expression of the predetermined search query ( (A vector) and a distributed expression (vector) of another search query paired with a predetermined search query are learned so as to be close to each other in the distributed expression space. Learning to make two vectors close to each other in a distributed expression space can be rephrased as learning to make two vectors similar in a distributed expression space. As described above, since the information processing apparatus 100 generates the first learning model using the DSSM technique using the LSTM for generating the distributed expression, the first learning model generated by the information processing apparatus 100 includes an arbitrary character. A distributed representation (vector) can be generated from the columns. In other words, 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).

  Further, the information processing device 100 generates a second learning model. Specifically, the information processing device 100 generates a second learning model that predicts a category to which a predetermined search query belongs from a predetermined search query using the first learning model. More specifically, after generating 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 acquiring the first model M1, the information processing apparatus 100 generates a second learning model M2 using the acquired first model M1. The information processing apparatus 100 generates a second model M2 having a different connection coefficient, which is a weight of the learning model, from the first model M1 by re-learning the first model M1. For example, when the search query is input to the learning model, the information processing apparatus 100 performs learning 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 for predicting a category to which a predetermined search query belongs from the query is generated.

  Now, the flow of information processing will be described with reference to 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 as the advertiser is “related to bone health”. For example, an advertiser client is a business that runs a health-related business. In addition, a client who is an advertiser wants to analyze search trends of users who are interested in “bone health” when developing advertisements for their products. The information processing apparatus 100 receives the analysis query “bone health” and a classification axis (category) including four small classifications from the client terminal 20 (step S1). Note that the information processing apparatus 100, instead of receiving the classification axis (category) specified by the client as the advertiser, selects the client as the advertiser from the categories shown in the classification definition storage unit 123 shown in FIG. May be received as a classification axis. In the example illustrated in FIG. 1, the client as the advertiser selects the large category “healthy” identified by the large category ID “CAT2” from the categories shown in the category definition storage unit 123 described below. Then, the information processing apparatus 100 receives the major category “healthy” selected by the client as the advertiser as a category axis.

  Subsequently, when receiving the analysis query, the information processing apparatus 100 inputs the received analysis query “bone health” to the first model M1 (Step S2). The information processing apparatus 100 generates a first model M1 in which a plurality of search queries input by the same user within a predetermined period of time previously have similar characteristics, and learns the features of the plurality of search queries. It is assumed that When the analysis query is input to the first model M1, the information processing apparatus 100 outputs a 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 converted to the analysis query “bone health”. A similar query is extracted as a similar query (step S4). For example, the information processing apparatus 100 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 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 the distributed expressions (vectors). Subsequently, the information processing apparatus 100 extracts a distributed expression (vector) having a cosine similarity exceeding 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 the distributed expressions (vectors) based on not only the cosine similarity but any index that can be applied as a distance measure between the vectors. Is also 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, and a Mahalanobis distance. Subsequently, the information processing apparatus 100 determines 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 smaller than the predetermined threshold value by the analysis query “bone health”. It may be extracted as a distributed expression (vector) similar to the distributed expression (vector). Subsequently, upon extracting a similar distributed expression (vector), the information processing apparatus 100 specifies a search query corresponding to the extracted distributed expression (vector) with reference to the vector information storage unit 122 described later. Then, the information processing device 100 extracts the specified search query as a similar query similar to the analysis query.

  Subsequently, when the similar query is extracted, the information processing apparatus 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 as the advertiser, and then learns the second model M2 so as to classify the search query according to the received classification axis (category). May be. Alternatively, the information processing device 100 learns the second model M2 in advance so as to classify the search query according to the category shown in the classification definition storage unit 123 described later. Then, the information processing apparatus 100 causes the client as the advertiser to select a classification axis from the categories shown in the classification definition storage unit 123, and outputs the classification axis (category) selected by the client as the advertiser from the client terminal 20. May be accepted.

  Subsequently, the information processing apparatus 100 transmits the result of the classification of the query according to the classification axis (category) to the client terminal 20 (step S7). For example, the information processing apparatus 100 provides a configuration of a search query searched as a search query related to “analysis related to bone health”, which is an analysis theme, as a query classification result according to a 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 ratio of “drink” of 40%, a search query of “food” of 30%, and “exercise”. The analysis result that the ratio of the search query about "supplement" is 20% and the ratio of the search query about "supplement" is 10% is provided.

  As described above, the information processing apparatus 100 determines that the plurality of search queries input by the same user within a predetermined time period have similar characteristics, and the first learning model that has learned 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 appropriately interpret the meaning of the search query based on the context of the user who has input the search query. Therefore, by using the first learning model, the information processing apparatus 100 can analyze the search trend of the user in consideration of the context of the user who has input the search query. Therefore, the information processing apparatus 100 can improve the analysis accuracy of the search trend of the user.

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

  Subsequently, the information processing apparatus 100 inputs the extracted search query Q11 to 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 that has just been output from the output layer of the first model M1, and indicates a distributed expression before applying feedback to the first model M1 (before learning). Further, the information processing device 100 inputs the extracted search query Q12 to the first model M1, and outputs a 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 that has just been output from the output layer of the first model M1, and indicates a distributed expression before applying feedback to the first model M1 (before learning). Thus, the information processing apparatus 100 outputs the vector BQV11, which is a distributed expression of the search query Q11, and the vector BQV12, which is a distributed expression of the search query Q12 (step S12).

  Subsequently, the information processing apparatus 100 generates a pair of search queries including a search query Q11 (“Roppongi pasta”) and a search query Q12 (“Roppongi Italian”) continuously input within the predetermined time by the same user U1. Since the query is assumed to be a search query input with a predetermined search intention (for example, a search intention of “searching for a restaurant in a certain place”), the search query Q11 is assumed to have mutually similar characteristics. The first model M1 is trained so that the distributed expression (vector QV11) 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 between the vector BQV11 that is the distributed expression of the search query Q11 and the vector BQV12 that is the distributed expression of the search query Q12 before applying feedback to the first model M1 (before learning) is denoted by Θ. Further, the magnitude of the angle formed between the vector QV11, which is the distributed expression of the search query Q11, and the vector QV12, which is the distributed expression of the search query Q12, after applying feedback to the first model M1 (after learning) is Φ. At this time, the information processing device 100 trains the first model M1 so that Φ is 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. Further, the information processing apparatus 100 calculates a cosine similarity value between the vector QV11 and the vector QV12. Subsequently, the information processing apparatus 100 sets the first value of the cosine similarity between the vector QV11 and the vector QV12 to be larger (the value approaches 1) than the value of the cosine similarity between the vector BQV11 and the vector BQV12. The model M1 is learned. 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. A first model M1 that outputs an expression (vector) is generated (step S13). Note that the information processing apparatus 100 calculates the similarity between the distributed expressions (vectors) based on not only the cosine similarity but any index that can be applied as a distance measure between the vectors. Is also good. The information processing apparatus 100 may cause the first model M1 to learn 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, and a Mahalanobis distance. Subsequently, the information processing apparatus 100 may cause the first model M1 to learn such that the value of the predetermined distance function between the distributed expressions (vectors) (that is, the distance in the distributed expression space) is reduced.

  Next, the flow of the process of generating the first learning model will be described in more detail with reference to FIG. In the description of FIG. 3, portions that are the same as those in 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 illustrated in FIG. 3, a state is illustrated in which the distributed expression (vector) output by the first model M1 generated by the information processing device 100 is mapped to the distributed expression space. The information processing apparatus 100 performs the training of the first model M1 such that the distributed expression of the predetermined search query and the distributed expression of another search query paired with the predetermined search query are mapped close to each other in the distributed expression space. Do.

  In the example illustrated in the upper part of FIG. 3, the information processing apparatus 100 includes a search query Q11 (“Roppongi pasta”), which is four search queries continuously input by the same user U1 within a predetermined time period, 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 in which the time interval of inputting each search query by the same user U1 is within a predetermined time. The information processing apparatus 100 extracts a plurality of search queries in which a time interval when a pair of search queries described later is input by the same user U1 is within a predetermined time. When the search queries are arranged in the input order, the information processing apparatus 100 extracts 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 the four search queries are extracted, the information processing apparatus 100 sets two search queries that are adjacent in time to 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 (step S21-1). Note that the information processing apparatus 100 may extract a plurality of search queries in which all search queries are 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 they are adjacent in time series, and converts the selected two search queries into a pair of search queries. May be extracted as

  Subsequently, the information processing apparatus 100 inputs the extracted search query Q1k (k = 1, 2, 3, 4) to the first model M1, and outputs the search query Q1k (k = 1, 2, 3, 4). A vector BQV1k (k = 1, 2, 3, 4) as a distributed representation is output. Here, the vector BQV1k (k = 1, 2, 3, 4) is a distributed expression 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 applying feedback to the first model M1 (before learning) 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, “eating and drinking at a certain location (near Minato-ku, Tokyo)”). It is assumed that the search query is a search query input by “search for a store”), so that the search query Q11 has distributed characteristics (vector QV11), The first model M1 is learned 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 expression of the search query Q12 (vector QV12) and the distributed expression of the search query Q13 paired with the search query Q12 (vector QV13) are similar in the distributed expression space. One model M1 is trained. Further, the information processing apparatus 100 determines that the distributed expression of the search query Q13 (vector QV13) and the distributed expression of the search query Q14 paired with the search query Q13 (vector QV14) are similar in the distributed expression space. One 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. 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, a vector QV1k (k = 1, 2, 3, 4) which is a distributed expression of the search query Q1k (k = 1, 2, 3, 4) is close to the distributed expression space. The state where the position is mapped as the cluster CL11 is shown. 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 intention of “search for a restaurant in a certain place (near Minato-ku, Tokyo)”. Is estimated. That is, the search query Q1k (k = 1, 2, 3, 4) is a search query searched under the search intention of “search for a restaurant in a certain place (near Minato-ku, Tokyo)”. , Are presumed to be search queries having mutually similar characteristics. Here, when a predetermined search query input with a search intention of “search for a restaurant in a certain place (near Minato-ku, Tokyo)” is input to the first model, the information processing apparatus 100 determines the position of the cluster CL11. Can be output. Thus, for example, the information processing apparatus 100 extracts a search query corresponding to the distributed expression mapped to the position of the cluster CL11, thereby performing a search of “search for a restaurant in a certain place (near Minato-ku, Tokyo)”. A search query according to the intention can be extracted. 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 includes a search query Q21 (“refrigerator 400L”), which is three search queries continuously input within a predetermined time by the same user U2, and a search query. Q22 (“refrigerator medium-sized”) and search query Q23 (“refrigerator medium-sized recommended”) are extracted. When the search queries are arranged in the input order, the information processing apparatus 100 extracts three search queries input in the order of the search query Q21, the search query Q22, and the search query Q23. When the three search queries are extracted, the information processing device 100 sets two search queries adjacent in time series as a pair of search queries, and is a pair of two search queries (search query Q21, search query Q22). , (Search query Q22, search query Q23) (step S21-2).

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

  Subsequently, the information processing apparatus 100 determines that a pair of search queries continuously input by the same user U2 within a predetermined time period has a predetermined search intention (for example, a search intention of “checking a medium-sized refrigerator”). Since the input search query is presumed to be an input search query, it is assumed that the search query Q21 has characteristics similar to each other, and a distributed expression (vector QV21) of the search query Q21 and a distributed expression (vector QV22) is trained in the first model M1 so as to be similar in the distributed expression space. Further, the information processing apparatus 100 determines that the distributed expression of the search query Q22 (vector QV22) and the distributed expression of the search query Q23 paired with the search query Q22 (vector QV23) are similar in the distributed expression space. One 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. 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, a vector QV2m (m = 1, 2, 3) which is a distributed expression of the search query Q2m (m = 1, 2, 3) is located at a position close to the distributed expression space in the cluster CL21. Is shown. For example, the search query Q2m (m = 1, 2, 3) is estimated to be a set of search queries searched for by the user U2 under the search intention of “check medium-sized refrigerator”. In other words, Q2m (m = 1, 2, 3) is a search query having characteristics similar to each other in that it is a search query searched under the search intention of “check medium-sized refrigerator”. Presumed. Here, when a predetermined search query input with a search intention of “check medium-sized refrigerator” is input to the first model, the information processing apparatus 100 generates a distributed expression that is mapped to the position of the cluster CL21. Can be output. Thus, for example, the information processing apparatus 100 extracts a search query corresponding to the search intention of “check medium-sized refrigerator” by extracting a search query corresponding to the distributed expression mapped to the position of the cluster CL21. 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 different characteristics in that a plurality of search queries randomly extracted are search queries searched under different search intentions. Assuming that there is, the first model M1 is learned. 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 independently of the predetermined search query are distantly mapped in the distributed expression space. Training of the first model M1 is performed as described above. In the example illustrated in FIG. 3, it is assumed that the information processing apparatus 100 randomly extracts a search query irrespective of the search query Q11 and then extracts the search query Q21. In this case, the information processing apparatus 100 stores the distributed expression of the search query Q11 (vector QV11) and the distributed expression of the search query Q21 randomly extracted independently of the search query Q11 (vector QV21) on the distributed expression space. The first model M1 is trained so as to be mapped far away. As a result, a vector QV1k that is a distributed expression of the search query Q1k (k = 1, 2, 3, 4) searched under the search intention of “search for a restaurant in a certain place (near Minato-ku, Tokyo)” This is a distributed expression of a cluster CL11 including (k = 1, 2, 3, 4) and a search query Q2m (m = 1, 2, 3) searched under a search intention of “check medium-sized refrigerator”. The cluster CL21 including the vector QV2m (m = 1, 2, 3) is mapped far away 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 having different search intentions by learning the first model M1 so that the distributed expressions of the plurality of search queries randomly extracted are different. Enables output to a distant position in the expression space.

  Note that, as a result of the distributed expression (vector) output by the first model M1 generated by the information processing apparatus 100 being mapped to the distributed expression space, the same user as the predetermined user in addition to the cluster CL11 and the cluster CL21 described above. Are generated, 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.

  As described above, the information processing device 100 acquires the search query input by the user. In addition, the information processing apparatus 100 learns a plurality of search queries input by the same user within a predetermined period of time from among the acquired search queries as having similar characteristics. And generating a first model for predicting 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 a plurality of search queries continuously input within a predetermined time are search queries searched under a predetermined search intention. The first model is learned as a search query having the following characteristics. More specifically, the information processing apparatus 100 learns the first model so that the distributed expressions of a plurality of search queries input by the same user within a predetermined time are similar to each other. And generating a first model that outputs a distributed expression including feature information of the search query. 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. 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 has input the search query. Therefore, the information processing apparatus 100 can appropriately interpret the meaning of the search query. Further, the information processing apparatus 100 extracts a search query corresponding to the distributed expression mapped to the vicinity of the distributed expression including the characteristic information of the predetermined search query, and thereby, according to the search intention in which the predetermined search query is searched. Search queries can be extracted. That is, the information processing apparatus 100 enables analysis of a user's search trend in consideration of the search intention and context of the user who has input the search query. Therefore, the information processing apparatus 100 can improve the analysis accuracy of the search trend of the user. Further, the first model M1 generated by the information processing apparatus 100 can also 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) using the feature information of the search query predicted by the first model M1. An expression can also be provided. Thereby, the search system can select the content output as the search result based on the feature information of the search query predicted by the first model M1. That is, the search system can select the content output as a search result in consideration of the search intention and context of the user who has input 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. 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 context of the user who has input the search query. Therefore, the information processing device 100 can improve usability in the search service.

[Second learning model generation process]
Next, the flow of the process of generating the second learning model will be described with reference to FIG. FIG. 4 is a diagram illustrating an example of a process of generating a second learning model according to the embodiment. In the following, the second learning model is appropriately described as a second model (or a second model M2). In the example shown in the upper part of FIG. 4, the information processing apparatus 100 includes a search query Q11 (“Roppongi pasta”), which is four search queries continuously input within a predetermined time by the same user U1, 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 interval of inputting each search query by the same user U1 is within a predetermined time. In addition, the information processing apparatus 100 extracts a plurality of search queries in which the time interval when each search query pair is input by the same user U1 is within a predetermined time. Here, it is assumed that the four search queries are search queries input by the user U1 within a predetermined time in the order of the search query Q11, the search query Q12, the search query Q13, and the search query Q14. When the four search queries are extracted, the information processing apparatus 100 sets two search queries that are adjacent in time to 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). After extracting three search query pairs, the information processing device 100 inputs the extracted search query Q1k (k = 1, 2, 3, 4) to the first model M1 (step S31). Note that the information processing apparatus 100 may extract a plurality of search queries in which all search queries have been input within a predetermined time by the same user U1. Then, the information processing apparatus 100 selects two search queries from among the plurality of extracted search queries regardless of whether they are adjacent in time series, and converts the selected two search queries into a pair of search queries. May be extracted as

  Subsequently, the information processing apparatus 100 uses a vector BQV1k (k = 1, 2, 3, 4), which is a distributed expression of the search query Q1k (k = 1, 2, 3, 4), as output data of the first model M1. Output (Step S32). Here, the vector BQV1k (k = 1, 2, 3, 4) is a distributed expression 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 applying feedback 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, “a certain place (near Minato-ku, Tokyo)” by the user U1. It is presumed that it is a set of search queries searched under the search intention of "Find a restaurant in." That is, the search query Q1k (k = 1, 2, 3, 4) is a search query searched under the search intention of “search for a restaurant in a certain place (near Minato-ku, Tokyo)”. , Are presumed to be search queries having mutually similar characteristics. Therefore, the information processing apparatus 100 generates a first model that predicts feature information of a predetermined search query from a predetermined search query by learning that search queries that are continuously input have similar characteristics. (Step S33). Specifically, the information processing apparatus 100 learns that the distributed expressions of the continuously input search queries are similar, 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 expression (vector QV11) of the search query Q11 and the distributed expression (vector QV12) of the search query Q12 paired with the search query Q11 are similar in the distributed expression space. One model M1 is trained. Further, the information processing apparatus 100 determines that the distributed expression of the search query Q12 (vector QV12) and the distributed expression of the search query Q13 paired with the search query Q12 (vector QV13) are similar in the distributed expression space. One model M1 is trained. Further, the information processing apparatus 100 determines that the distributed expression of the search query Q13 (vector QV13) and the distributed expression of the search query Q14 paired with the search query Q13 (vector QV14) are similar in the distributed expression space. One model M1 is trained.

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

  When 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 acquiring the first model M1, the information processing apparatus 100 generates a second learning model M2 using the acquired first model M1. Specifically, the information processing apparatus 100 re-learns the first model M1, thereby generating a second model M2 having a different connection coefficient as a weight of the learning model from 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 the predetermined search query belongs from the 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 CAT11 (“Search for a restaurant”), CAT12 (“Search for a product”), and CAT13 ( A second model M2 that predicts which of the four categories of “reserve a restaurant” and CAT14 (“purchase a product”) is generated. Specifically, when the search query is input to the second model M2 as input information, the information processing apparatus 100 outputs, as output information, the second model M2 that outputs the probability that the search query belongs to the category for each category. Generate. For example, the information processing apparatus 100 learns a set of a search query and a category (any one of CAT11 to CAT14) to which the search query belongs as the correct answer data of the second model M2.

  It should be noted that the fact that the search query belongs to CAT11 (“Find a restaurant”) indicates that the search query is a search query input with the intention of searching for a restaurant. Also, belonging to the CAT 12 (“Search for a product”) indicates that the search query is a search query input with the intention of searching for a product. Further, the fact that the search query belongs to CAT13 (“Reserve a restaurant”) indicates that the search query is a search query input with the intention of reserving a restaurant. The fact that the search query belongs to the CAT 14 (“purchase a product”) indicates that the search query is a search query input with the intention of purchasing a product.

  Specifically, when the search query is input to the learning model, the information processing apparatus 100 performs learning 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 from the predetermined search query is generated. Then, for example, when the search query is input to the second model M2 as input information, the information processing apparatus 100 outputs the probability that the search query belongs to the category as output information for each of the categories CAT11 to CAT14. Generate M2.

  For example, when the search query Q11 (“Roppongi pasta”) is input as input information to the second model M2 (step S35), the information processing apparatus 100 distributes the search query Q11 (“Roppongi pasta”) as output information. A vector BQV11, which is an expression, is output. Here, the vector BQV11 is a distributed expression of the search query Q11 that has just been output from the output layer of the second model M2, and indicates a distributed expression before applying feedback to the second model M2 (before learning). Here, it is assumed that the correct answer category to which the search query Q11 (“Roppongi pasta”) belongs is CAT11 (“Find a restaurant”). In this case, the information processing apparatus 100 determines that the probability that the vector BQV11, which is a distributed expression of the output search query Q11 (“Roppongi pasta”), is classified as CAT11 (“Find a restaurant”) exceeds a predetermined threshold. To learn the second model M2. In addition, the information processing apparatus 100 learns the second model using the correct data prepared in advance. The information processing device 100 may generate the correct answer data of the second model M2. Then, the information processing apparatus 100 may cause the second model M2 to learn using the generated correct answer data. Specifically, the information processing apparatus 100 determines the correct answer category to which the search query belongs based on the behavior of the user who searched the search query after the search. More specifically, the information processing apparatus 100 associates a predetermined behavior in which a ratio of a user who has performed a predetermined action after the search exceeds a predetermined threshold with respect to a user who has searched for a predetermined search query to a correct answer category. Is determined as the action to be performed. For example, as a percentage of the users who searched for the search query Q11 (“Roppongi pasta”), the percentage of users who took a predetermined action after the search was 90%, and the percentage of users who took an action to search for restaurants was 90%. It is assumed that the ratio of users who have caused the action of 0%, the rate of the user who has reserved the restaurant after the search is 10%, and the rate of the user who has purchased the product after the search is 0%. In this case, the information processing apparatus 100 determines the activity of searching for a restaurant by the search query Q11 (“Roppongi pasta”) because the ratio of users who have performed the activity of searching for a restaurant exceeds a predetermined threshold (for example, 90%). The action is determined as the action corresponding to the correct answer category. Then, since the information processing apparatus 100 determines that the action corresponding to the correct answer category is an 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”). 80%, 0% probability of being classified as CAT12 ("Search for products"), 20% probability of being classified as CAT13 ("Reserving restaurants"), and CAT14 ("Purchase products"). Suppose that the probability of being classified as) is output as 0%. In this case, the information processing apparatus 100 learns the second model M2 such that the probability that the vector BQV11, which is a distributed expression, is classified as CAT11 (“find a restaurant”) exceeds a predetermined threshold (for example, 90%). Let it. In addition, the information processing apparatus 100 learns such that the probability that the vector BQV11, which is a distributed expression, is classified as CAT11 (“find a restaurant”) exceeds a predetermined threshold (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 performs the second processing so that the probability that the distributed expression of the predetermined search query is classified into the correct category as the output information exceeds the predetermined threshold. Train two models. Then, when a predetermined search query is input as input information, the information processing apparatus 100 assigns a category in which the probability that the distributed expression of the predetermined search query belongs to the category exceeds a predetermined threshold to the category of the predetermined search query. Output as For example, when the search query Q11 (“Roppongi pasta”) is input as input information to the learned second model M2, the information processing apparatus 100, the vector BQV11 that is a distributed expression of the search query Q11 (“Roppongi pasta”). Since the probability of belonging to the category CAT11 ("Find a restaurant") exceeds 90%, the category to which the search query belongs is output as CAT11 ("Find a restaurant") as output information (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, thereby generating the second model that predicts the category of the predetermined search query from the predetermined search query (step S37). ).

  In general, it is considered that a user performs a search a plurality of times with a certain intention. Therefore, it is assumed that a search query input continuously within a predetermined time has a similar search intention. 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 are search queries searched under a predetermined search intention. The first model M1 is learned as a search query having the following characteristics. Thereby, the information processing apparatus 100 can make the first model M1 learn the characteristics of the search query in consideration of the search intention. Then, the information processing device 100 efficiently utilizes the first model M1 that has learned the characteristics of the search query in consideration of the search intention to efficiently predict the second model that predicts the category of the predetermined search query from the predetermined search query. Can be generated. Thereby, the information processing apparatus 100 can classify the search query into a category considering the search intention of the user who has input the search query. Also, conventionally, in order to classify search queries into categories and obtain high classification accuracy, it was necessary to prepare a sufficient amount of correct answer data. However, since search queries themselves are diverse and have long-tail properties, labeling correct categories corresponding to a large number of search queries is extremely troublesome and difficult. Here, instead of labeling the correct answer category, the information processing apparatus 100 learns the second model that predicts the category of the search query using the search intention of the user (the context of the user who has input the search query) as a kind of correct answer. Can be done. Thus, the information processing apparatus 100 can learn the second model without manually labeling the correct answer category of the search query. That is, the information processing apparatus 100 can obtain sufficient classification accuracy even when the number of correct answer data is small. In addition, the information processing apparatus 100 can obtain a higher classification accuracy when there are many correct answer 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 apparatus 100 according to the embodiment. As illustrated 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) for receiving various operations from an administrator or the like of the information processing apparatus 100, and a display unit (for example, a liquid crystal display) for displaying various information. May have.

(Communication unit 110)
The communication unit 110 is realized by, for example, a NIC (Network Interface Card) or the like. The communication unit 110 is connected to 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 element such as a random access memory (RAM) or a flash memory, or a storage device such as a hard disk or an optical disk. As illustrated 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 types of information related to a search query input by a user. FIG. 7 illustrates an example of the query information storage unit according to the embodiment. In the example illustrated in FIG. 7, the query information storage unit 121 has items such as “user ID”, “date and time”, “search query”, and “search query ID”.

  “User ID” indicates identification information for identifying a user who has input a search query. “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. “Search query ID” indicates identification information for identifying a search query input by a user.

  In the example shown in the first record in 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 has input the search query Q11 is the user identified by the user ID “U1” (user U1). Further, the date and time “2018/9/1 PM 17:00” indicates that the date and time when the search server received 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 the characters “Roppongi” indicating the place name and the character “Pasta” indicating the type of food are character strings separated by spaces that are delimiters.

(Vector information storage unit 122)
The vector information storage unit 122 stores various types of information regarding vectors, which are distributed expressions of search queries. FIG. 8 illustrates an example of the vector information storage unit according to the embodiment. In the example illustrated 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 the search query. “Search query ID” indicates identification information for identifying a search query corresponding to a vector. “Vector information” indicates an N-dimensional vector that is a distributed expression of the search query. The vector that is a 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 a distributed expression of the search query Q11 shown in FIG. The search query identified by the search query ID “Q11” (search query 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 information related to the definition of the category into which the search query is classified. FIG. 9 illustrates an example of the classification definition storage unit according to the embodiment. In the example illustrated in FIG. 9, the classification definition storage unit 123 has items such as “major classification ID”, “major classification”, “small classification ID”, and “small classification”.

  The “major category” indicates a major category of a category into which the search query is classified. The “major category ID” indicates identification information for identifying the major category. In the example illustrated in FIG. 9, the large classification “purchasing behavior system” corresponds to the large classification described in the example illustrated in the lower part of FIG. 1. The major category "purchasing behavior type" indicates a major category of a category for classifying the search query based on the purchasing behavior of the user. In the example shown in FIG. 9, the large category “purchasing behavior” has four more small categories. The major category ID “CAT1” indicates identification information for identifying the major category “purchasing behavior system”.

  “Small classification” indicates a small classification of a category into which the search query is classified. “Small classification ID” indicates identification information for identifying a small classification. In the example shown in FIG. 9, the small category “Find a restaurant” is a category belonging to the large category “Purchase behavior”, and a search query classified into the small category is input by the user with the intention of searching for a restaurant. Indicates that the search query was performed. The small category ID “CAT11” indicates identification information for identifying the small category “Find a restaurant”.

  The sub-category “Search for products” is a classification belonging to the main classification “Purchase behavior”, and indicates that a search query classified into the sub-category is a search query input by a user with the intention of searching for products. . The small category ID “CAT12” indicates identification information for identifying the small category “search for product”.

  The sub-class “Reservation of restaurants” is a classification belonging to the main category “Purchase behavior”, and the search query classified into the sub-class is a search query input by the user with the intention of reserving a restaurant. Indicates that The small category ID “CAT13” indicates identification information for identifying the small category “Reserve restaurant”.

  The small category "Purchase goods" is a classification belonging to the large category "Purchase behavior system". Show. The small category ID “CAT14” indicates identification information for identifying the small category “purchase goods”.

(Category information storage unit 124)
The category information storage unit 124 stores various types of information related to the category to which the search query belongs. Specifically, the category information storage unit 124 stores various types of information on the category output by the second learning model when a search query is input to the learned second learning model. FIG. 10 illustrates 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”, “major category ID”, “small category ID”, and “probability (%)”.

  “Search query ID” indicates identification information for identifying a search query input by a 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 category ID” indicates identification information for identifying the major category. “Small classification ID” indicates identification information for identifying a small classification. “Probability (%)” indicates the probability of each small classification output by the second learning model when a search query is input to the learned second learning model. In the example illustrated 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 types of information regarding the learning model generated by the information processing device 100. FIG. 11 illustrates an example of the model information storage unit according to the embodiment. In the example illustrated in FIG. 11, the model information storage unit 125 has items such as “model ID” and “model data”.

  “Model ID” indicates identification information for identifying a 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, the "model data" stores data for converting a search query into a distributed expression.

  In the example shown in the first record in 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 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 one of the layers from the input layer to the output layer other than the output layer, a first element and a first element. And a second element whose value is calculated based on the weight of one element. The 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 function.

  Here, it is assumed that the model data MDT1 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 MDT1 corresponds to input data (xi) such as x1 or x2. Further, 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 corresponds to any node of the input layer, and the second element can be regarded as a node of 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 a DNN (Deep Neural Network). In this case, the first element included in the model data MDT1 corresponds to one of the nodes of the input layer or the intermediate layer. The second element corresponds to a next-stage node that is a node to which a value is transmitted from a node corresponding to the first element. Further, the weight of the first element corresponds to a connection coefficient that is a weight considered for a value transmitted from a node corresponding to the first element to a node corresponding to the second element.

  The information processing device 100 calculates a distributed expression using a model having an arbitrary structure, such as the above-described regression model or neural network. Specifically, coefficients are set in the model data MDT1 so as to output a distributed expression when a search query is input. The information processing apparatus 100 calculates a distributed expression using such model data MDT1.

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

  Further, when the information processing apparatus 100 performs the estimation processing using GAN (Generative Adversarial Networks), the model data MDT1 may be a model constituting 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. 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 one of the layers from the input layer to the output layer other than the output layer, And a second element whose value is calculated based on the weight of one element. The information processing apparatus 100 may be caused to function so as to output the information.

  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 input data (xi) such as x1 or x2. Further, 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 corresponds to any node of the input layer, and the second element can be regarded as a node of the output layer.

  It is also assumed that the model data MDT2 is realized by a neural network having one or a plurality of intermediate layers, such as a DNN (Deep Neural Network). In this case, the first element included in the model data MDT2 corresponds to one of the nodes of the input layer or the intermediate layer. The second element corresponds to a next-stage node that is a node to which a value is transmitted from a node corresponding to the first element. Further, the weight of the first element corresponds to a connection coefficient that is a weight considered for a value transmitted from a node corresponding to the first element to a 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 above-described regression model or neural network. Specifically, a coefficient is set in the model data MDT2 so that, when a search query is input, the probability that the search query belongs to each category is output. The information processing apparatus 100 calculates the probability that the search query belongs to each category using such model data MDT2.

  Note that, in the above example, an example in which the model data MDT2 is a model (hereinafter, referred to as a model X2) that outputs a 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 repeatedly inputting and outputting data to and from the model X2. For example, the model data MDT2 may be a model (hereinafter, referred to as a model Y2) learned by inputting the distributed expression output by the model X2 when a search query is input. Alternatively, the model data MDT2 may be a model that has been learned so that a search query is input and an output value of the model Y2 is output.

  Further, when the information processing apparatus 100 performs the estimation process using GAN (Generative Adversarial Networks), the model data MDT2 may be a model constituting 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 apparatus 100 by, for example, a CPU (Central Processing Unit) or an MPU (Micro Processing Unit). Are implemented by executing various programs (corresponding to an example of a generation program) using a 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 according to the first model M1 (model data MDT1) stored in the model information storage unit 125, and in response to a search query input to the input layer, The computer is made to function to output the distributed representation from the output layer by performing an operation based on the first element and the weight of the first element with each element belonging to the element as the first element.

  Further, the control unit 130 performs information processing according to the second model M2 (model data MDT2) stored in the model information storage unit 125, and in response to a search query input to the input layer, The computer is operated 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 as 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 an information processing operation described below. Note that the internal configuration of the control unit 130 is not limited to the configuration illustrated in FIG. 6, and may be another configuration as long as it performs information processing described below.

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

(Extraction unit 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. For example, the extraction unit 132 extracts a plurality of search queries in which the time interval of inputting each search query by the same user is within a predetermined time. Subsequently, the extraction unit 132 extracts a pair of search queries continuously input within a predetermined time by the same user 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 between input of each search query pair by the same user is within a predetermined time. For example, the extraction unit 132 outputs a search query Q11 (“Roppongi pasta”), which is four search queries consecutively input by the same user U1 within a predetermined period of time among the search queries acquired by the acquisition unit 131. ), A search query Q12 (“Roppongi Italian”), a search query Q13 (“Akasaka pasta”), and a search query Q14 (“Azabu pasta”). After arranging the search queries in the input order, the extraction unit 132 extracts 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 extraction unit 132 extracts four search queries, the two search queries that are adjacent in time are taken as a pair of search queries, and are a pair of three search queries (search query Q11, search query Q11). Q12), (search query Q12, search query Q13), and (search query Q13, search query Q14) are extracted. Note that the extraction unit 132 may extract a plurality of search queries in which all 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 treats the selected two search queries as a pair of search queries. May be extracted.

  The extraction unit 132 extracts a predetermined search query and other search queries irrelevant to the predetermined search query from among the search queries acquired by the acquisition unit 131. 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 irrespective of the predetermined search query.

(Generation unit 133)
The generation unit 133 generates various information. Specifically, the generation unit 133 learns, among the search queries acquired by the acquisition unit 131, that a plurality of search queries input by the same user within a predetermined time have similar characteristics. And generating a learning model for predicting feature information of the predetermined search query from the predetermined search query. Specifically, the generation unit 133 trains the learning model so that the distributed expressions of a plurality of search queries input by the same user within a predetermined time are similar to each other. Generate a learning model that predicts query feature information. For example, the generation unit 133 generates a learning model by learning so that the distributed expressions of a pair of search queries input continuously within a predetermined time are similar. For example, the generation unit 133 calculates the value of the similarity of the distributed expression (vector) of the pair of search queries before learning. In addition, the generation unit 133 calculates a similarity value of the distributed expression (vector) after learning of the pair of search queries. Subsequently, the generation unit 133 trains the learning model such that the value of the similarity of the distributed expression (vector) after learning becomes larger than the value of the similarity of the distributed expression (vector) before learning. As described above, the generation unit 133 trains the learning model 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 thereby the distributed expression (vector ) Is generated. More specifically, the generation unit 133 generates a learning model that outputs a distributed expression (vector) from a search query by using a DSSM technique using LSTM, which is a kind of RNN, for generating a distributed expression. For example, the generating unit 133 determines that a pair of search queries input by the same user within a predetermined time period have similar characteristics as the correct answer data of the learning model, , Is learned so that a predetermined search query and a distributed expression (vector) of another search query to be paired exist close in the distributed expression space. When generating the first learning model, the generating 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 configured by a three-layer LSTM RNN. In the example illustrated in FIG. 12, the extraction unit 132 includes a pair of a search query Q11 “Roppongi pasta” and a search query Q12 “Roppongi Italian” that are continuously input within a predetermined time by the same user U1. Extract search queries. The generation unit 133 inputs the search query Q11 extracted by the extraction unit 132 to the input layer of the first learning model M1 (Step S41).

  Subsequently, the generation unit 133 outputs a 256-dimensional vector BQV11 that 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 to the input layer of the first learning model M1. Subsequently, the generation unit 133 outputs a 256-dimensional vector BQV12 that is a distributed expression of the search query Q12 from the output layer of the first learning model M1 (Step S42).

  Subsequently, the generation unit 133 learns such that the distributed expressions (vectors) of two consecutively input search queries are similar to each other, thereby outputting a distributed expression (vector) from the search queries. Is generated (step S43). For example, the magnitude of the angle between 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, before applying feedback to the first learning model M1 (before learning) is denoted by Θ. Further, the magnitude of the angle between the vector QV11, which is the distributed expression of the search query Q11, and the vector QV12, which is the distributed expression of the search query Q12, after applying feedback to the first learning model M1 (after learning) is denoted by Φ. . At this time, the generation unit 133 trains the first learning model M1 such that Φ is smaller than Θ. For example, the generation unit 133 calculates the value of the cosine similarity between the vector BQV11 and the vector BQV12. Further, the generation unit 133 calculates the value of the cosine similarity between the vector QV11 and the vector QV12. Subsequently, the generation unit 133 sets the learning model M1 such 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. As described above, the generation 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 generation unit 133 performs distribution from the search query. A first learning model M1 that outputs an expression (vector) is generated. Note that the generation unit 133 calculates the similarity between the distributed expressions (vectors) based on not only the cosine similarity but also any index that is applicable as a distance measure between the vectors. Good. Further, the generation unit 133 may cause the learning model M1 to learn 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, and a Mahalanobis distance. Subsequently, the generation unit 133 may cause the learning model M1 to learn such that the value of the predetermined distance function between the distributed expressions (vectors) (that is, the distance in the distributed expression space) is reduced.

  In addition, the generation unit 133 learns, as a plurality of search queries input by the same user within a predetermined time period, a plurality of search queries including a character string separated by a predetermined delimiter having similar characteristics. By doing so, a first learning model is generated. For example, the generation unit 133 generates a search query “Roppongi pasta” in which characters “Roppongi” indicating a place name and “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” separated from the character “Italian” indicating the type of dish by a space that is a delimiter has similar characteristics.

  Further, the generation unit 133 generates a first learning model by learning as a plurality of search queries randomly extracted from the search queries acquired by the acquisition unit 131 as having different characteristics. Specifically, the generation unit 133 generates the first learning model by learning the search queries acquired by the acquisition unit 131 so that the distributed expressions of the pair of search queries randomly extracted are different. I do. For example, the generation unit 133 determines that the distributed expression of the predetermined search query extracted by the extraction unit 132 and the distributed expression of the search query randomly extracted regardless of the predetermined search query are far apart in the distributed expression space. The training of the first learning model M1 is performed so as to be mapped.

  Further, the generation unit 133 generates a second learning model. Specifically, the generation unit 133 acquires the first learning model (model data MDT1 of the first learning model M1) generated by the generation unit 133 with reference to the model information storage unit 125. Subsequently, using the acquired first learning model, the generation unit 133 generates a second learning model that predicts a category to which the predetermined search query belongs from the predetermined search query. Upon acquiring the first model M1, the generating unit 133 generates a second learning model M2 using the acquired first model M1. The generation unit 133 generates a second model M2 having a different connection coefficient, which is a weight of the learning model, from the first model M1 by re-learning the first model M1. Specifically, when the search query is input to the learning model, the generation unit 133 performs learning so that the classification result of the distributed expression output by the learning model corresponds to the category to which the search query belongs, and A second model M2 that predicts a category to which a predetermined search query belongs from the search query is generated.

  Specifically, when the search query is input to the learning model, the generation unit 133 performs learning so that the classification result of the distributed expression output by the learning model corresponds to the category to which the search query belongs. And generating a second learning model for predicting a category to which a predetermined search query belongs from the search query. The generation unit 133 generates a second learning model that outputs a probability for each category to which the search query belongs as output information when the search query is input to the learning model as input information. For example, when a predetermined search query is input as input information to the learning model using the first model M1, the generation unit 133 determines the probability that the distributed expression of the search query is classified into that category as output information. A second model M2 to be output every time is generated. When a predetermined search query is input as input information, the generation unit 133 learns the second model so that the probability that the distributed expression of the predetermined search query is classified into the correct category as output information exceeds a predetermined threshold. . Then, when a predetermined search query is input as input information, the generation unit 133 sets a category in which the probability that the distributed expression of the predetermined search query belongs to the category exceeds a predetermined threshold as a category of the predetermined search query. A second model M2 to be output is generated. After generating the second learning model, the generating 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 acquires the first model M1 (model data MDT1 of the first model M1) with reference to the model information storage unit 125 illustrated in FIG. Subsequently, the generation unit 133 refers to the classification definition storage unit 123 illustrated in FIG. 9 and selects a large classification of a category into which the search query is classified. Subsequently, when the large classification is selected, the generation unit 133 learns a set of a search query and a small classification to which the search query belongs as 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 a distributed expression of the search query Q11 (“Roppongi pasta”) from the output layer of the second model M2. A certain vector BQV11 is output. Here, the vector BQV11 is a distributed expression of the search query Q11 that has just been output from the output layer of the second model M2, and indicates a distributed expression before applying feedback to the second model M2 (before learning). In this case, the generation unit 133 determines that the probability that the vector BQV11, which is a distributed expression of the output search query Q11 (“Roppongi pasta”), is classified into the correct category CAT11 (“Search for restaurants”) exceeds a predetermined threshold. The second model M2 is learned as described above.

  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%, the probability of being classified as CAT12 ("Find a product") is 0%, the probability of being classified as CAT13 ("Reserving a restaurant") is 20%, and CAT14 ("Purchase a product"). Is output as 0%. In this case, the generation unit 133 learns the second model M2 such that the probability that the vector BQV11, which is a distributed expression, is classified as CAT11 (“find a restaurant”) exceeds a predetermined threshold (for example, 90%). . In addition, the generation unit 133 adjusts the distribution so that the probability that the vector BQV11, which is a distributed expression, is classified as CAT11 (“Find a restaurant”) exceeds a predetermined threshold (for example, 90%). The second model M2 is trained so that the probability that the expression vector BQV11 is classified into another category CAT13 ("reserve a restaurant") is reduced 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”). Since the probability of belonging to the category CAT11 (“Find a restaurant”) exceeds 90%, the category to which the search query belongs is output as CAT11 (“Find a restaurant”) as output information.

  Note that 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 calculates the probability that the search query belongs to each of the small classes belonging to the arbitrary number of the large classes selected by the search query as the output information for each of the small classes. May be generated. Further, the generation unit 133 may select all the large categories as the large categories. Then, when the search query is input to the second model M2, the generation unit 133 may generate the second model M2 that outputs the probability belonging to each sub-class for every sub-class.

[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. 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, thereby generating a second learning model M2 having a different connection coefficient as a weight of the learning model from the first learning model M1.

  More specifically, similarly to the first learning model M1, the second learning model M2 generated by the information processing device 100 is configured by a three-layer LSTM RNN. In the example illustrated in FIG. 13, the extraction unit 132 inputs the search query Q11 “Roppongi pasta” input by the user U1 to the input layer of the second learning model M2 (Step S51).

  Subsequently, 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 second learning model M2 (Step S52).

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

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

(Service providing unit 134)
The service providing unit 134 acquires a first learning model. Specifically, the service providing unit 134 acquires the first learning model generated by the generating unit 133 with reference to the model information storage unit 125. More specifically, the service providing unit 134 determines that the plurality of search queries input by the same user within a predetermined period of time have similar characteristics, and the first learning that has learned the characteristics of the plurality of search queries Get the model. For example, the service providing unit 134 learns, by the generation unit 133, that a plurality of search queries input by the same user within a predetermined time period have similar characteristics, thereby performing a predetermined search from a predetermined search query. A first learning model generated as a learning model for predicting feature information of a query is obtained. In addition, when a predetermined search query is input as input information, the service providing unit 134 acquires a first learning model that outputs a distributed expression of the predetermined search query as output information. In addition, the service providing unit 134 learns the first learning model that has learned the features of the plurality of search queries by learning so that the distributed expressions of the pair of search queries 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 separated by a predetermined delimiter have similar characteristics as a plurality of search queries input within a predetermined time by the same user. By learning, a first learning model that has learned features of a plurality of search queries is obtained. In addition, the service providing unit 134 acquires a first learning model in which the features of the plurality of search queries are learned by learning that the plurality of search queries randomly extracted have different features. In addition, the service providing unit 134 obtains a first learning model in which the features of a plurality of search queries are learned by learning so that the distributed expressions of a pair of search queries randomly extracted are different.

  Further, the service providing unit 134 acquires the second learning model. Specifically, the service providing unit 134 acquires the second learning model generated by the generating unit 133 with reference to the model information storage unit 125. More specifically, the service providing unit 134 generates a second learning model generated by the generation unit 133 as a learning model for predicting a category to which a predetermined search query belongs from a predetermined search query using the first learning model. To get. For example, when the search query is input to the learning model, the service providing unit 134 learns that the classification result of the distributed expression output by the learning model corresponds to the category to which the search query belongs, so that the predetermined search is performed. A second learning model that predicts a category to which a predetermined search query belongs from the query is acquired. For example, when a search query is input to a learning model as input information, the service providing unit 134 acquires, as output information, a second learning model that outputs, for each category, the probability that the search query belongs to a category.

  The service providing unit 134 provides a search query analysis service to a client as an advertiser. The service providing unit 134 determines that a plurality of search queries input by the same user within a predetermined period of time have similar characteristics, and uses a first learning model that has learned features of the plurality of search queries, A similar query, which is a search query having characteristics similar to those of the search query, is extracted. Specifically, the service providing unit 134 receives an analysis theme of a search query from a client that is an advertiser, and extracts a 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 an analysis theme for analyzing a search trend of the user. The service providing unit 134 may receive a plurality of analysis queries on the analysis theme. For example, suppose that a client who is an advertiser wants “bone health-related” as a search query analysis theme. 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”. Thereby, 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 completely extract similar queries according to the analysis theme of the client as the advertiser. Subsequently, the service providing unit 134 extracts a search query corresponding to the received 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.

  Further, the service providing unit 134 classifies similar queries that are search queries having characteristics similar to a predetermined search query using the second learning model. Specifically, the service providing unit 134 receives a classification axis of the search query from the client as the advertiser, and classifies the search query into a category corresponding to the received classification axis. For example, the service providing unit 134 receives, from the client terminal 20, a classification axis (category) for classifying 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 classifies the similar query into a category corresponding to the received classification axis using the second learning model.

  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. Note that, instead of accepting the classification axis (category) specified by the client as the advertiser, the service providing unit 134 selects a category from the classification definition storage unit 123 shown in FIG. 9 by the client as the advertiser. The obtained large classification may be received as a classification axis. In the example illustrated in FIG. 1, the client as the advertiser selects the large category “healthy” identified by the large category ID “CAT2” from the categories shown in the category definition storage unit 123 described below. Then, the service providing unit 134 receives the major category “healthy” selected by the client as the advertiser as a category axis.

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

  Subsequently, when the service providing unit 134 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 value is converted to the analysis query “bone health”. Extract as similar similar queries. 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 variance expression (vector) having a cosine similarity exceeding a predetermined threshold as a variance expression (vector) similar to the variance 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 not only the cosine similarity but any index that can be applied as a distance measure between the vectors. Is also 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, and a Mahalanobis distance. Subsequently, the service providing unit 134 determines the distributed expression (vector) in which the value of the predetermined distance function (ie, the distance in the distributed expression space) between the distributed expressions (vectors) is smaller than a predetermined threshold value by the analysis query “bone health”. It may be extracted as a distributed expression (vector) similar to the distributed expression (vector). Next, upon extracting 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 specified search query as a similar query similar to the analysis query.

  Subsequently, when extracting the similar query, the service providing unit 134 inputs the extracted similar query to 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 as the advertiser, and then learns the second model M2 so as to classify the search query according to the received classification axis (category). May be. Alternatively, the service providing unit 134 previously trains the second model M2 so as to classify the search query according to the category shown in the classification definition storage unit 123 described later. Then, the service providing unit 134 causes the client, which is the advertiser, to select a classification axis from the categories shown in the classification definition storage unit 123, and displays the classification axis (category) selected by the client, which is the advertiser, from the client terminal 20. May be accepted.

  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 a classification result of the query according to the classification axis (category), a configuration of a search query searched as a search query related to “analysis related to bone health”. For example, the service providing unit 134 breaks down the search query extracted as a similar query similar to the analysis query, where the ratio of the search query regarding “drink” is 40%, the ratio of the search query regarding “food” is 30%, and the ratio of “exercise” The analysis result that the ratio of the search query about "supplement" is 20% and the ratio of the search query about "supplement" is 10% is provided.

[3. Flow of generation processing of first learning model]
Next, a procedure of a process of generating a first learning model according to the embodiment will be described with reference to FIG. FIG. 14 is a flowchart illustrating the generation processing procedure of the first learning model according to the embodiment. In the example illustrated in FIG. 14, the information processing apparatus 100 acquires a search query input by a 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 generates a first learning model that predicts feature information of a predetermined search query from a predetermined search query by learning as a plurality of extracted search queries having similar characteristics. (Step S103).

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

  Subsequently, the information processing apparatus 100 generates a second learning model that predicts a category of a predetermined search query from a predetermined search query using the first learning model (Step S202).

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

[6. effect〕
As described above, the information processing apparatus 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 period of time have similar characteristics, and uses the first learning model that has learned the characteristics of the plurality of search queries, A similar query, which is a search query having characteristics similar to those of the search query, is extracted.

  In general, it is considered that a user performs a search a plurality of times with a certain intention. Therefore, it is assumed that a search query input continuously within a predetermined time has a similar search intention. 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 are search queries searched under a predetermined search intention. The learning model M1 is trained by regarding the search query as having the following characteristics. 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 are similar to each other. It is possible to output a distributed expression of a query to a close position in a 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 has input the search query. Therefore, the information processing apparatus 100 can appropriately interpret the meaning of the search query. Further, the information processing apparatus 100 extracts a search query corresponding to the distributed expression mapped to the vicinity of the distributed expression including the characteristic information of the predetermined search query, and thereby, according to the search intention in which the predetermined search query is searched. Search queries can be extracted. That is, the information processing apparatus 100 can extract a search query in consideration of the context of the user who has input the search query. Therefore, by using the first learning model, the information processing apparatus 100 can analyze the search trend of the user in consideration of the context of the user who has input the search query. Therefore, the information processing apparatus 100 can improve the analysis accuracy of the search trend of the user.

  Further, the learning model M1 generated by the information processing apparatus 100 can also 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 (for example, a search engine used for a search query analysis service provided by the information processing apparatus 100) using the feature information of the search query predicted by the learning model M1. As information, a distributed expression of the search query output by the learning model M1 can be provided. Thereby, the search system can select the content output as the search result based on the feature information of the search query predicted by the learning model M1. That is, the search system can select the content output as a search result in consideration of the search intention and context of the user who has input the search query. Further, the search system can calculate 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 feature 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 context of the user who has input the search query. Therefore, the information processing device 100 can improve usability in the search service.

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

  Thereby, the information processing apparatus 100 can measure the abstract concept of the characteristic of the predetermined search query by using a specific numerical value called a distributed expression.

  In addition, the service providing unit 134 learns the first learning model that has learned the features of the plurality of search queries by learning so that the distributed expressions of the pair of search queries 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 time have the same search intention or similar search intentions even if they are not the same. That is, it is considered that a pair of search queries that are continuously input within a predetermined time have the same search intention, or have a similar search intention even if they are not the same. That is, the information processing apparatus 100 can more accurately extract a search query according to a user's search intention by learning so that the distributed expressions of a pair of search queries continuously input within a predetermined time are similar. It can be.

  In addition, the service providing unit 134 determines that a plurality of search queries including a character string separated by a predetermined delimiter have similar characteristics as a plurality of search queries input within a predetermined time by the same user. By learning, a first learning model that has learned features of a plurality of search queries is obtained.

  Generally, it is considered that a search query including a plurality of character strings has a clearer search intention than a search query including a single character string. That is, the information processing apparatus 100 can use a search query including a character string delimited by a predetermined delimiter to perform learning, so that a search query corresponding to a user's search intention can be extracted with higher accuracy. .

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

  In general, a plurality of search queries 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 distant. Therefore, the information processing apparatus 100 according to the present invention is a search query having characteristics different from each other in that a plurality of search queries randomly extracted are search queries searched under different search intentions. The learning model M1 is learned assuming that there is. More 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 randomly extracted are different, so that the distributed expressions of search queries having different search intents are obtained. Can be output to a distant position in the distributed representation space. Accordingly, the learning model can learn incorrect answer data, which is a pair of search queries with distant search intentions, in addition to correct answer data, which is a pair of search queries with close search intentions. 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.

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

  As a result, the information processing apparatus 100 efficiently utilizes the first learning model, which has learned the characteristics of the search query in consideration of the search intention, to predict the category of the predetermined search query from the predetermined search query. Can be generated dynamically. Thereby, the information processing apparatus 100 can classify the search query into a category considering the search intention of the user who has input the search query. Also, conventionally, in order to classify search queries into categories and obtain high classification accuracy, it was necessary to prepare a sufficient amount of correct answer data. However, since search queries themselves are diverse and have long-tail properties, labeling correct categories corresponding to a large number of search queries is extremely troublesome and difficult. Here, instead of labeling the correct answer category, the information processing apparatus 100 learns the second model that predicts the category of the search query using the search intention of the user (the context of the user who has input the search query) as a kind of correct answer. Can be done. Thus, the information processing apparatus 100 can learn the second model without manually labeling the correct answer category of the search query. That is, the information processing apparatus 100 can obtain sufficient classification accuracy even when the number of correct answer data is small. In addition, the information processing apparatus 100 can obtain a higher classification accuracy when there are many correct answer data. Therefore, the information processing device 100 can improve the classification accuracy of the search query.

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

  As a result, the information processing apparatus 100 uses the distributed expression including the characteristics of the search query in consideration of the search intention to classify the search query into a category in which the search intention of the user who has input the search query is considered. Can be generated efficiently. That is, the information processing apparatus 100 enables the search query to be classified into a category that considers the search intention of the user who has input the search query. Therefore, the information processing device 100 can improve the classification accuracy of the search query.

  Further, the service providing unit 134 classifies similar queries that are search queries having characteristics similar to a predetermined search query using the second learning model.

  Thereby, the information processing apparatus 100 can classify the search query into a category considering the search intention of the user who has input 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 receives a search query analysis theme from a client that is an advertiser, and extracts a search query corresponding to the received analysis theme.

  Thereby, 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 as the advertiser.

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

  Thereby, the information processing apparatus 100 can appropriately provide the analysis result of the user's search trend along the classification axis desired by the client as the advertiser.

[7. Hardware configuration)
The information processing apparatus 100 according to the embodiment described above is realized by, for example, a computer 1000 having a configuration as shown in FIG. FIG. 17 is a hardware configuration diagram illustrating an example of a computer that implements the functions of the information processing apparatus 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 that depends on hardware of the computer 1000, and the like.

  The HDD 1400 stores a program executed by the CPU 1100, data used by the program, and the like. Communication interface 1500 receives data from another device via a predetermined communication network, sends the data to CPU 1100, and transmits the data generated by CPU 1100 to the other 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, CPU 1100 outputs the generated data to an output device via 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), a PD (Phase Change Rewritable Disk), a magneto-optical recording medium such as an MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory. And so on.

  For example, when the computer 1000 functions as the information processing apparatus 100 according to the embodiment, the CPU 1100 of the computer 1000 loads the program or data (for example, the model data MDT1 of the first model M1, the 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 and executes these programs from the recording medium 1800. As another example, these programs or data (for example, the model of the first model M1) may be transmitted from another device via a predetermined communication network. Data MDT1 and model data MDT2) of the second model M2 may be obtained.

  As described above, some of the embodiments of the present application have been described in detail with reference to the drawings. However, these are exemplifications, and various modifications based on the knowledge of those skilled in the art, including the aspects described in the section of the disclosure of the invention, The invention can be implemented in other modified forms.

[8. Others)
Further, among the processes described in the above embodiments and modifications, all or a part of the processes described as being performed automatically can be manually performed, or described as being performed manually. All or a part of the processing can be automatically 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.

  Each component of each device illustrated is a functional concept, and does not necessarily need to be physically configured as illustrated. That is, the specific form of distribution / integration of each device is not limited to the one shown in the figure, and all or a part thereof may be functionally or physically distributed / arbitrarily divided into arbitrary units according to various loads and usage conditions. 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.

  Further, the “section (section, module, unit)” described above can be read as “means”, “circuit”, or the like. For example, the service providing unit can be replaced with a service providing unit or a service providing circuit.

Reference Signs List 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 Provision Department

Claims (14)

Assuming that a plurality of search queries input by the same user within a predetermined period of time have similar characteristics, a similarity to the predetermined search query is determined using a first learning model that has learned the characteristics of the plurality of search queries. An information processing apparatus comprising: a service providing unit that extracts a similar query that is a search query having the following characteristics. The service providing unit includes:
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 obtained as output information. The service providing unit includes:
Acquiring a first learning model that has learned features of the plurality of search queries by learning so that the distributed expressions of the pair of search queries input continuously within the predetermined time are similar. The information processing apparatus according to claim 1, wherein The service providing unit includes:
By learning as a plurality of search queries input within a predetermined time by the same user as a plurality of search queries including a character string delimited by a predetermined delimiter, the plurality of search queries have similar characteristics. The information processing device according to any one of claims 1 to 3, wherein a first learning model that has learned a feature of the search query is acquired. The service providing unit includes:
A first learning model that learns features of the plurality of search queries is obtained by learning the plurality of search queries randomly extracted as having different features. An information processing device according to any one of the above. The service providing unit includes:
A first learning model that learns the features of the plurality of search queries is obtained by learning so that the distributed expressions of a pair of search queries randomly extracted are different from each other. An information processing device according to any one of the above. The service providing unit includes:
2. A second learning model generated by using the first learning model, the second learning model predicting a category to which the predetermined search query belongs from a predetermined search query is acquired. 3. 7. The information processing apparatus according to any one of claims 6 to 6. The service providing unit includes:
When a 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 converted to the predetermined search query. The information processing apparatus according to any one of claims 1 to 7, wherein a second learning model that predicts a category to which the information belongs belongs. The service providing unit includes:
When a search query is input to a learning model as input information, a second learning model that outputs, for each category, the probability that the search query belongs to a category is acquired as output information. An information processing device according to any one of the above. The service providing unit includes:
The information processing apparatus according to claim 9, wherein a similar query that is a search query having characteristics similar to a predetermined search query is classified using the second learning model. The service providing unit includes:
The information processing apparatus according to claim 1, wherein the information processing apparatus receives an analysis theme of a search query from a client serving as an advertiser and extracts a search query corresponding to the received analysis theme. The service providing unit includes:
The information processing apparatus according to any one of claims 1 to 11, wherein a classification axis of the search query is received from a client as an advertiser, and the search query is classified into a category corresponding to the received classification axis. An information processing method executed by a computer,
Assuming that a plurality of search queries input by the same user within a predetermined period of time have similar characteristics, a similarity to the predetermined search query is determined using a first learning model that has learned the characteristics of the plurality of search queries. An information processing method comprising: a service providing step of extracting a similar query that is a search query having the following characteristics. Assuming that a plurality of search queries input by the same user within a predetermined period of time have similar characteristics, a similarity to the predetermined search query is determined using a first learning model that has learned the characteristics of the plurality of search queries. An information processing program for causing a computer to execute service providing means for extracting a similar query that is a search query having the following characteristics.

Images