JP7224392B2 - Information processing device, information processing method and program - Google Patents

Description

The present invention relates to an information processing device, an information processing method, and a program, and more particularly to technology for providing search results in response to an input query.

A query is a search request that is submitted to a search engine to cause the search engine to output desired search results.
For example, an EC (Electronic Commerce) site built on the web implements a search engine, receives input of a query for searching for an item that a user wishes to purchase, and obtains a search key and The search engine searches for items tagged with tags that match a keyword, and presents the user with a list of the resulting items.

Many search engines allow multi-word queries that include multiple keywords in one query. A multi-word query makes it easier for the user to reflect the user's intentions in the query from multiple viewpoints. The desired search results will not be obtained because the search results that should be expected will be omitted from the output.

For this reason, search engines that allow multi-word queries usually use partial match to output search results that match either a key phrase (phrase) or a keyword (word) that constitutes a part of one query. By adopting search, the search range is expanded and more search results are output.
However, each of the multiple key phrases or keywords included in a single query does not necessarily equally represent the user's intention of what he or she actually wants to purchase. For this reason, partial match searches may output search results that are significantly different from the user's intentions, which may reduce search accuracy.

Patent Literature 1 discloses an information processing system that classifies queries into phrases or keywords, and selects phrases or keywords determined to express a user's desire from the classified phrases or keywords.
Specifically, in the information processing system of Patent Document 1, a query type classification unit classifies a query into either phrases, keywords, or any combination of these data, and a query type determination unit uses a request dictionary Searching a database to obtain query types and query type scores corresponding to the classified phrases, keywords, or combination data, and selecting phrases, keywords, or combination data with high obtained query type scores.

Query type is a classification pattern according to the nature and request of phrases and keywords, such as "guidance type" and "request type". It is a score that quantifies the accuracy. The wish dictionary database stores the corresponding query type and query type score for each phrase, keyword, or combination data.

Japanese Unexamined Patent Application Publication No. 2017-41030

According to the technique disclosed in Patent Literature 1, it is possible to select a key phrase or keyword that is highly likely to be the user's desire, among the key phrases or keywords that make up one query.
However, with the technique of Patent Literature 1, a query type and a query type score must be defined in a dictionary in advance for each of the key phrases and keyword candidates that are components that can be included in a query.
As a result, the load of registration and maintenance related to predefined definitions increases, and query type scores cannot be obtained for query components that are not predefined, resulting in lack of flexibility and reduced search accuracy. There is a risk that it will be lost.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problems, and its object is to provide a configuration that more closely matches the user's intention from among the constituent elements of a query without predefining each of the constituent elements included in the query. An object of the present invention is to provide an information processing device, an information processing method, and a program capable of specifying an element with high accuracy.

In order to solve the above problems, one aspect of an information processing apparatus according to the present invention includes: a first query acquisition unit that acquires a first query; Obtaining a query subset generation unit that divides into a plurality of tokens and generates a plurality of query subsets composed of one or more divided tokens, and a second query that includes the query subset generated by the query subset generation unit a second query acquisition unit, a similarity calculation unit that calculates the similarity of the second query to the first query based on the user's action history, and the similarity calculation unit, a score calculation unit that calculates a score of the query subset based on the similarity of the second query; and a query subset having a higher score calculated by the score calculation unit among the plurality of query subsets. and a query subset identification unit to identify.

The similarity calculation unit may calculate the similarity as a degree of similarity between the search result set of the second query and the search result set of the first query.
The information processing device connects the first query or the second query to the attribute of the query with a link based on the action history of the user, and assigns a weight to the link, thereby forming a query model and storing the generated query model in a storage device, wherein the similarity calculation unit refers to the query model stored in the storage device to obtain the second query may be calculated.

The similarity calculation unit calculates the weight given to the link connected to the attribute shared between the query model of the first query and the query model of the second query , the similarity of the second query may be calculated.

The similarity calculation unit calculates a plurality of the weights respectively given to the pairs of the links connected to the attributes shared between the query model of the first query and the query model of the second query. and adding the weights with smaller values for a plurality of the attributes to calculate the similarity of the second query.

The similarity calculation unit calculates the similarity of the second query by comparing information associated with an item on which the user has performed an operation in the past, between the first query and the second query. can be calculated.

The query model generation unit may generate the query model by setting information associated with the item as the attribute and assigning the weight based on the frequency of the operation.

The score calculator may calculate the score of the query subset by calculating a geometric mean of the similarities of all the second queries.

The second query acquisition unit acquires the second query for each of the plurality of query subsets generated by the query subset generation unit, and the score calculation unit acquires the second query for each of the plurality of query subsets, The score may be calculated.

The query subset generation unit may generate the query subset from a plurality of discontinuous tokens in the character string of the first query.

The information processing apparatus may further include an output control unit that controls output such that search results corresponding to the query subset identified by the query subset identification unit are preferentially presented.

One aspect of an information processing method according to the present invention is an information processing method executed by an information processing apparatus, comprising: acquiring a first query; dividing the acquired first query into a plurality of tokens; , generating a plurality of query subsets consisting of one or more divided tokens, and obtaining a second query including the generated query subsets;
calculating a similarity of the second query to the first query based on a user's behavior history; and calculating a score of the query subset based on the similarity of the second query. and identifying a query subset with a higher calculated score among the plurality of query subsets.

One aspect of an information processing program according to the present invention is an information processing program for causing a computer to execute information processing, the program comprising: a first query acquisition process for acquiring a first query; a query subset generation process that divides the first query acquired by the first query acquisition process into a plurality of tokens and generates a plurality of query subsets each composed of one or more divided tokens; and the query subset generation. a second query acquisition process for acquiring a second query including the query subset generated by the process; and calculating a degree of similarity of the second query to the first query based on a user's action history. A score calculation process for calculating the score of the query subset based on the similarity of the second query calculated by the similarity calculation process and the similarity calculation process, and among the plurality of query subsets, and query subset identification processing for identifying a query subset having a higher score calculated by the score calculation unit.

Advantageous Effects of Invention According to the present invention, it is possible to specify with high accuracy a component that more closely matches the user's intention from among the components of a query without predefining each of the components included in the query.
The objects, aspects and effects of the present invention described above and the objects, aspects and effects of the present invention not described above can be understood by a person skilled in the art to carry out the following invention by referring to the accompanying drawings and the description of the claims. can be understood from the form of

FIG. 1 is a block diagram showing an example of the functional configuration of a search system including a keyword specifying device according to an embodiment of the present invention. FIG. 2 is a flowchart showing an example of a schematic processing procedure of keyword identification processing executed by the keyword identification device according to the present embodiment. FIG. 3 is a simplified example of a query model that defines the relationship between queries and attributes and their weights, generated by the query model generation unit of the keyword identification device according to the present embodiment and stored in the query model storage unit. 1 is a conceptual diagram for explanation. FIG. 4 is a conceptual diagram illustrating an example of calculating similarity for a certain query in FIG. FIG. 5 is a conceptual diagram illustrating an example of calculating similarities for other queries in FIG. FIG. 6 is a flowchart showing an example of a detailed processing procedure of single evaluation processing executed by the single evaluation unit of the keyword identification device according to this embodiment. FIG. 7 is a conceptual diagram illustrating an example of calculating the similarity of other queries that share a certain single with respect to the original query, and calculating the score of the single based on the calculated similarity. FIG. 8 is a block diagram showing an example of the hardware configuration of the keyword specifying device according to this embodiment.

Embodiments for carrying out the present invention will be described in detail below with reference to the accompanying drawings. Among the constituent elements disclosed below, those having the same functions are denoted by the same reference numerals, and descriptions thereof are omitted. The embodiments disclosed below are examples of means for realizing the present invention, and should be appropriately modified or changed according to the configuration of the device to which the present invention is applied and various conditions. is not limited to the embodiment of Also, not all combinations of features described in the present embodiment are essential for the solution means of the present invention.

The keyword identification device according to this embodiment divides a query to be submitted to a search engine into individual tokens that make up the query, generates a single consisting of one or more tokens, and generates a single , generate a query containing that single.
The keyword identification device according to the present embodiment further refers to the query model generated based on the user's behavior history, calculates the similarity of each generated query to the original query, and from the calculated similarity , the score of the single is calculated, and the single with the higher calculated score is specified as a keyword that more closely matches the user's intention.

In the following, for example, this embodiment is applied to a search engine implemented in EC (Electronic Commerce), and in a use case where a user inputs a multi-word query to purchase an item, based on the user's action history Although an example applied to the use of selecting a keyword that more closely matches the user's intention in a query will be described, this embodiment is not limited to this, and can be applied to searches for any use.

<Functional Configuration of Keyword Identification Device>
FIG. 1 is a block diagram showing an example of the functional configuration of a keyword identification device 1 according to this embodiment.
The keyword identification device 1 shown in FIG.
The keyword identification device 1 may be communicably connected to a client device (not shown) such as a PC (Personal Computer) via a network. In this case, the keyword identification device 1 is mounted on a server, and the client device may provide a user interface when the keyword identification device 1 inputs/outputs information to/from the outside. A part or all of 11 to 15 may be provided.

Referring to FIG. 1, a keyword identification device 1, a user action history DB (database) 2, a search engine 3, and a query model storage unit 4 constitute a search system, and the keyword identification device 1 accesses the user action history DB 2. Then, a query model is generated, stored in the query model storage unit 4, and the processing result of the keyword identification processing is supplied to the search engine 3, but the present embodiment is not limited to this. For example, the keyword identification device 1 may be installed inside the search engine 3 or may directly acquire the user's action history from the search engine 3 .

The query input unit 11 inputs a query, which is a search request for causing the search engine 3 to execute a search, and supplies the query to the query division unit 12 . The query input unit 11 may receive a query in real time from a client device connected via a network, or may acquire a query stored in advance in a storage device.
The query input unit 11 also receives input of various parameters necessary for executing keyword identification processing in the keyword identification device 1 . The query input unit 11 may receive input of various parameters via a user interface of a client device communicably connected to the keyword identification device 1 .

The query division unit 12 divides the query supplied from the query input unit 11 into multiple tokens. To split a multi-word query containing multiple keywords into multiple tokens, you can detect the delimiters (space, comma, semicolon, etc.) between keywords written as a string in the query, or can be morphologically analyzed and decomposed into parts of speech.
In this embodiment, the query dividing unit 12 further generates multiple singles to be evaluated (score calculation) by the single evaluation unit 14 described later from the divided tokens, and divides the generated singles into , to the single evaluation unit 14 .

A token is a semantic minimum unit that constitutes a key phrase or keyword included in a query, and one word corresponds to one token.
A shingle is a keyphrase or keyword used for partial match search of a query and is composed of one or more tokens (words).

As a non-limiting example, assume that a query as a search request is "travel coffee mug". In this case, the query "travel coffee mug" is split into three tokens "travel", "coffee" and "mug". Each token consists of one word, which is the minimum unit.
From these three tokens "travel", "coffee", and "mug", a single consisting of two tokens (a cluster of tokens) can be written as "travel coffee", "travel mug", "coffee mug", one As singles composed of tokens, "travel", "coffee" and "mug" are generated respectively.

In this embodiment, as can be understood from the single "travel mug", not only combining consecutive words in the query, but also combining non-consecutive (adjacent) words in the query character string, the single generated.
Thus, a single is a keyword or keyphrase that consists of one token or any combination of multiple tokens. That is, a single is any subset of the set of tokens (tuple) and any subset of queries (query subset). Multiple singles, or subsets of multiple queries, are generated from multi-word queries.
Hereinafter, keywords and key phrases are collectively referred to as "keywords". In other words, a keyword consists of one or more words that serve as search keys. By generating singles, the search range can be expanded and more search results can be output.

The query model generation unit 13 generates a query model based on the user's action history, and outputs the query model to the query model storage unit 4 . The query model storage unit 4 is composed of a non-volatile storage device such as a HDD (Hard Disk Drive) or an SSD (Solid State Drive), and stores the query model generated by the query model generation unit.
A query model is a graph that models the semantic relationship between a query and attributes, consisting of a query, one or more attributes associated with the query, and links connecting the query and each attribute. is. A link that connects a query and each attribute is given a weight that indicates the degree of relevance between the query and the attribute that are connected to both ends of the link. A query model may be a graph connecting multiple queries via one or more attributes associated with both of the multiple queries.

In this embodiment, the attributes associated with the query are, for example, item genre (category), vendor (brand), item name or item ID, price range, condition (new, new and old) that characterize already purchased items. , used items, etc.), tags (sale items, non-sale items), etc., but is not limited thereto, and may be any information that can be acquired in association with the query.

In this embodiment, the query model generation unit 13 stores the behavior history of the user stored in the user behavior history DB 2, for example, a Generate a query model based on the operations, the frequency of those operations, and so on.
Operations performed by the user on items include, for example, an operation of clicking an image of an item or a link, an operation of tapping an image of an item to view it, an operation of tabbing "Favorites", and a purchase operation. includes, but is not limited to, any operation that may occur in the process of purchasing an item. Hereinafter, these operations are simply referred to as "operations".

Specifically, the query model generation unit 13 connects the input query and the item purchased corresponding to the query with a link, and weights each link according to the number of purchases. A model graph is generated, and the generated graph is stored in the query model storage unit 4 . By tracking actions such as user actions, the attributes of the purchased item can be linked to the query that was originally run for the purchase, and the weight that should be given to the link can be determined. The query model thus generated models the user's intent associated with purchasing an item.
The query model generation unit 13 refers to the user behavior history DB 2, and generates a query model in advance from queries input to the search engine 3 in the past and operations performed by the user in response to the queries. good. A query model may be generated individually for each user, or may be generated for a group of users grouped by age group, gender, occupation, past purchase amount, or the like. Details of this query model will be described later with reference to FIGS.

The single evaluation unit 14 refers to the query models stored in the query model storage unit 4 and evaluates each of the singles supplied from the query division unit 12 .
Specifically, the single evaluation unit 14 inputs a query (hereinafter, also referred to as an “original query”) to be processed for keyword identification processing input to the query input unit 11, and a plurality of queries generated from the original query. A plurality of singles are evaluated by calculating a score for the original query for each single based on the singles.
The score calculated by the single evaluation unit 14 serves as an index indicating the extent to which each single matches the user's intention in the original query. The details of this single evaluation process will be described later with reference to FIG.

The output unit 15 outputs one or a plurality of singles for which a high score is calculated by the single evaluation unit 14 to an output device such as a display device, and supplies the singles to the search engine 3 . The output unit 15 also outputs various input data and various processing results of the keyword identification process executed by the keyword identification device 1, query models stored in the query model storage unit 13, and the like as appropriate via an output device such as a display device. you can

<Processing procedure for keyword identification processing>
FIG. 2 is a flowchart showing an example of the procedure of keyword identification processing executed by the keyword identification device 1 according to this embodiment.
Each step in FIG. 2 is implemented by the CPU reading out and executing a program stored in a storage device such as an HDD of the keyword identification device 1 . Also, at least part of the flowchart shown in FIG. 2 may be realized by hardware. When implemented by hardware, for example, by using a predetermined compiler, a dedicated circuit may be automatically generated on an FPGA (Field Programmable Gate Array) from a program for implementing each step. Also, a Gate Array circuit may be formed in the same manner as the FPGA and implemented as hardware. Also, it may be realized by an ASIC (Application Specific Integrated Circuit). The same applies to each step in FIG. 3, which will be described later.

In S1, the query input unit 11 of the keyword identification device 1 receives input of a query, which is a search request for causing the search engine 3 to perform a search. The input query may be a multi-word query containing multiple keywords, but may also be a single-word query containing a single keyword. The query input unit 11 may receive a query from a client device connected via a network, or acquire a query stored in advance in a storage device.
In S2, the query division unit 12 of the keyword identification device 1 divides the query supplied from the query input unit 11 into a plurality of tokens. Each of the multiple tokens is a single keyword.

At S3, the query division unit 12 of the keyword identification device 1 generates a plurality of singles from the plurality of tokens divided at S2. From the query input at S1, the single generated at S2 is a subset of the query, consisting of one or more tokens. Each single is evaluated by the single evaluator 14 and given a score.

In S4, the single evaluation unit 14 of the keyword identification device 1 evaluates each single by calculating a score for each of the multiple singles generated in S3.
Specifically, the single evaluation unit 14 refers to the query model stored in the query model storage unit 4, and for each single, based on the action history of the user who entered the query in S1, inputs the query in S1. Calculates the score for the given query. The single score calculated here is an indicator of how well the query input in S1 matches the user's intention in purchasing the item, that is, how significant it is in expressing the user's intention. becomes. The details of the single evaluation process executed by the single evaluation unit 14 will be described later with reference to FIG.

In S5, the output unit 15 of the keyword identification device 1 selects one or a plurality of singles for which a high score was calculated among the plurality of singles for which the score was calculated in S4 as a keyword that most conspicuously expresses the user's intention. can be determined to be At S5, the output unit 15 may supply the search engine 3 with one or more singles to be output.
The search engine 3 may, for example, sort search results by query so that search results corresponding to singles supplied from the output unit 15 are preferentially presented to the user. Alternatively, if the search results using the query input in S1 as a search key are few or if the search range is too narrow, the search engine 3 uses other queries including singles supplied from the output unit 15 as search keys. , the search may be performed again. As a result, search accuracy is improved, and search results that more closely match the user's intention can be presented.

<Generation of query model and calculation of similarity between queries>
Generating a query model and calculating similarity between queries will be described with reference to FIGS. In addition, in FIGS. 3 to 5, in order to simplify the explanation, an example in which each query is a single-word query consisting of a single keyword will be explained. can be processed.

FIG. 3 is a conceptual diagram for simplifying and explaining an example of a query model that defines the relationship between queries and attributes and their weights. Referring to FIG. 3, the query "pants" is linked to the attributes "kids", "womens", and "mens". The link between the query "pants" and the attribute "kids" has a weight of 0.2, the link between the query "pants" and the attribute "womens" has a weight of 0.3, and the link between the query "pants" and the attribute "kids" has a weight of 0.3. A weight of 0.5 is given to each link between "mens".
The query model of the query "pants" shown in FIG. It shows that the items in the category have action histories such as operations. That is, each attribute is an attribute obtained by categorizing the search results for the input query, and the similarity between queries is the degree of similarity between the search result set of the original query and the search result set of another query. indicates
The weight assigned to each link correlates with the frequency of operations. Depending on the type of operation, the weight may be set such that, for example, a higher weight is given to the purchase operation than to the click operation.

Referring to FIG. 3, the queries 'dress', 'jeans' and 'shirts' are linked to the attributes 'kids', 'womens' and 'mens' respectively. These queries 'dress', 'jeans' and 'shirts' are the other queries that share the attributes 'kids', 'womens' and 'mens' with the query 'pants' respectively. In FIG. 3, the query model storage unit 4 uses the query model storage unit 4 to calculate the single score for each of the original query input in S1 of FIG. be another query read from

A link between the query "dress" and the attribute "kids" is given a weight of 0.4, and a link between the query "dress" and the attribute "womens" is given a weight of 0.6. On the other hand, the query "dress" has no link to the attribute "mens" because it is not associated with the attribute "mens".
The link between the query "jeans" and the attribute "kids" has a weight of 0.3, the link between the query "jeans" and the attribute "womens" has a weight of 0.2, and the link between the query "jeans" and the attribute "womens" has a weight of 0.2. A weight of 0.5 is given to each link between "mens".
The link between the query "shirts" and the attribute "kids" has a weight of 0.3; the link between the query "shirts" and the attribute "womens" has a weight of 0.4; A weight of 0.3 is assigned to each link to "mens".

In FIG. 3, the query models of the other queries “dress”, “jeans”, and “shirts” share attributes with the query model of the original query “pants”. similar to each other.
FIG. 4 is a conceptual diagram of the query model of the original query "pants" and the query model of another query "shirts" extracted from FIG.

In this embodiment, the single evaluation unit 14 calculates the similarity by calculating the weight given to the link from each query to the attribute shared by the original query and other queries. do. For example, the single evaluation unit 14 may calculate the degree of similarity by adding the minimum value of the weights of the paired links (pairwise minimum).
Referring to FIG. 4, the minimum weight of paired links to attribute “women” is 0.3, the minimum weight of paired links to attribute “kids” is 0.2, and the minimum weight of paired links to attribute “mens” is is 0.3. Therefore, the sum of these three minimum values, 0.8, is the degree of similarity of the query "shirts" to the original query "pants".

On the other hand, FIG. 5 is a conceptual diagram in which the query model of the original query "pants" and the query model of another query "jeans" are extracted from FIG.
Referring to FIG. 5, the minimum weight of paired links to attribute “womens” is 0.2, the minimum weight of paired links to attribute “kids” is 0.2, and the minimum weight of paired links to attribute “mens” is is 0.5. Therefore, the sum of these three minimum values, 0.9, is the degree of similarity of the query “jeans” to the original query “pants”.
In the examples shown in FIGS. 3 to 5, it can be evaluated that the query "jeans" has a higher degree of similarity to the original query "pants" than the query "shirts".

<Detailed processing procedure for single evaluation processing>
FIG. 6 is a flowchart showing an example of a detailed processing procedure of single evaluation processing executed by the single evaluation unit 14 of the keyword identification device 1 according to this embodiment.
In S41, the single evaluation unit 14 of the keyword identification device 1 determines attributes and weights of the original query input in S1 of FIG.
Specifically, the single evaluation unit 14 refers to the query model storage unit 4, and if the query model of the original query input in S1 is already stored in the query model storage unit 4, the query model storage unit 4 The query model of the original query is read from the unit 4, skipping S42, and proceeding to S43.

On the other hand, if the query model of the original query is not stored in the query model storage unit 4, the single evaluation unit 14 refers to the user behavior history DB 2 and refers to the items purchased in the past by the user who entered the original query in S1. may be determined as the attribute of the original query, or the attribute of the item to be purchased may be estimated from the information on the EC site page displayed when the original query was entered. may Also, a predetermined initial value may be set for the weight for each attribute. Alternatively, the single evaluation unit 14, for example, reads from the query model storage unit 13 the query model of the query that shares the most tokens with the original query, and assigns the attributes and weights of the read query model to those of the original query. It may be determined or inferred as attributes and weights.
In the case of a new user who has no past action history, the initial query of the original query is obtained from the action history and query model of other users and user groups in categories close to the user's information that can be obtained when registering as a member. A model may be inferred.

In S42, the single evaluation unit 14 of the keyword identification device 1 causes the query model generation unit 13 to generate a query model by associating the attribute and weight determined in S41 with the original query input in S1. The query model generation unit 13 may store the query model generated in S<b>42 in the query model storage unit 4 .

In S43, the single evaluation unit 14 of the keyword identification device 1, for each single generated in S3 of FIG. Acquired from the query model storage unit 4 .
The single evaluator 14 may obtain other queries that share the single with the original query and share at least one attribute with the original query. This is because the degree of similarity between queries that do not share attributes is 0 according to the degree of similarity calculation in the examples of FIGS.

In S44, the single evaluation unit 14 of the keyword identification device 1 calculates the degree of similarity to the original query for each of the other queries acquired in S43.
Specifically, the single evaluation unit 14 calculates the weight given to the pair of links connected from both queries for the attribute shared by the query model of the original query and the query model of the other query. , to calculate the similarity of other queries to the original query. As already described with reference to FIGS. 4 and 5, the single evaluation unit 14 may calculate the degree of similarity by, for example, adding the minimum value of the weights of the paired links of the attributes.

In S45, the single evaluation unit 14 of the keyword identification device 1 calculates the score of the evaluation target single based on the similarities calculated in S44 with respect to all other queries acquired in S43.
Specifically, the single evaluation unit 14 calculates the average of the plurality of similarities calculated in S44 for all of the plurality of queries acquired in S43, for example, the geometric mean, as the evaluation target single. may be calculated as a score of

In S46, the single evaluation unit 14 of the keyword identification device 1 determines whether or not all of the singles generated in S3 of FIG. 2 have been evaluated.
If there is an unprocessed single (S46: N), the process returns to S43 and repeats the processes from S43 to S46. On the other hand, if there are no unprocessed singles (S46: Y), the process proceeds to S5 in FIG. 2, and one or more singles for which a higher score is calculated are output as keywords that match the user's intention.
In this way, the keyword identification device 1 according to this embodiment does not evaluate the individual tokens themselves that make up the original query, but evaluates the user's action history associated with each other query containing these tokens. allows you to specify keywords without predefining tokens or singles.

FIG. 7 is a conceptual diagram illustrating an example of calculating the similarity of other queries that share a certain single with respect to the original query, and calculating the score of the single based on the calculated similarity.
Referring to FIG. 7, it is assumed that the original query input in S1 of FIG. 2 is {“a”, “b”, “c”}. In this case, at S2 of FIG. 2, the original query is split into three tokens {“a”}, {“b”}, {“c”}, and at S3 of FIG. , "b"}, {"a", "c"}, {"b", "c"}, {"a"}, {"b"}, {"c"} are generated.

For each of these six singles, all other queries containing that single are obtained in S43 of FIG. For example, for single {“a”, “b”}, queries {“a”, “b”, “d”} and {“a”, “b”, “c”} are obtained. Query {“a”, “b”, “c”} is the same query as the original query.
For each of these other queries {“a”, “b”, “d”} and {“a”, “b”, “c”} obtained for the single {“a”, “b”}, In S44 of FIG. 6, the degree of similarity with respect to the original query {“a”, “b”, “c”} is calculated.

Referring to FIG. 7, query {“a”, “b”, “d”} shares attributes α ₁ and α ₂ with original query {“a”, “b”, “c”}, The minimum value of the weight given to the pair of links connected to the attribute _α1 is 0.2, and the minimum value of the weight given to the pair of links connected to the attribute _α2 is 0.3. Therefore, the similarity of the query {“a”, “b”, “d”} to the original query {“a”, “b”, “c”} is 0 as the sum of 0.2 and 0.3. .5.
On the other hand, the query {“a”, “b”, “c”}, which is the same as the original query, contains the original query {“a”, “b”, “c”} and all attributes α ₁ , α ₂ , and _α3 , and the minimum value of the weight given to the paired links connecting each attribute is 0.3 respectively. Therefore, the similarity of the query {“a”, “b”, “c”} to the original query {“a”, “b”, “c”} is calculated as 0.9.
Therefore, in S45 of FIG. 6, the score of single {“a”, “b”} is calculated as 0.67 as the geometric mean of 0.5 and 0.9.
By repeating the above process for all of the singles generated in S3 of FIG. 2, it is possible to identify one or more singles that originally had a high score.

<Hardware Configuration of Keyword Identification Device>
FIG. 8 is a diagram showing a non-limiting example of the hardware configuration of the keyword identification device 1 according to this embodiment.
The keyword identification device 1 according to this embodiment can be implemented on any single or multiple computers, mobile devices, or any other processing platform.
With reference to FIG. 8, an example in which the keyword identification device 1 is implemented in a single computer is shown, but the keyword identification device 1 according to this embodiment is implemented in a computer system including a plurality of computers. you can A plurality of computers may be interconnectably connected by a wired or wireless network.

As shown in FIG. 8, the keyword identification device 1 may include a CPU 81, a ROM 82, a RAM 83, an HDD 84, an input section 85, a display section 86, a communication I/F 87, and a system bus 88. The keyword identification device 1 may also comprise an external memory.
A CPU (Central Processing Unit) 81 comprehensively controls operations in the keyword identification device 1, and controls each component (82 to 87) via a system bus 88, which is a data transmission line.

A ROM (Read Only Memory) 82 is a non-volatile memory that stores control programs and the like necessary for the CPU 81 to execute processing. The program may be stored in a non-volatile memory such as a HDD (Hard Disk Drive) 84 or an SSD (Solid State Drive) or an external memory such as a removable storage medium (not shown).
A RAM (Random Access Memory) 83 is a volatile memory and functions as a main memory, a work area, and the like for the CPU 81 . That is, the CPU 81 loads necessary programs and the like from the ROM 82 to the RAM 83 when executing processing, and executes the programs and the like to realize various functional operations.

The HDD 84 stores, for example, various data and information necessary for the CPU 81 to perform processing using programs. The HDD 84 also stores various data, information, and the like obtained by the CPU 81 performing processing using programs and the like, for example.
The input unit 85 is composed of a pointing device such as a keyboard and a mouse.
The display unit 86 is configured by a monitor such as a liquid crystal display (LCD). The display unit 86 is a GUI (Graphical User Interface) that is a user interface for inputting instructions to the keyword identification apparatus 1 for various parameters used in the keyword identification process, communication parameters used in communication with other apparatuses, and the like. may be provided.

The communication I/F 87 is an interface that controls communication between the keyword identification device 1 and an external device.
A communication I/F 87 provides an interface with a network and performs communication with an external device via the network. Various data, various parameters, etc. are transmitted/received to/from an external device via the communication I/F 87 . In this embodiment, the communication I/F 87 may perform communication via a wired LAN (Local Area Network) conforming to a communication standard such as Ethernet (registered trademark) or a dedicated line. However, the network that can be used in this embodiment is not limited to this, and may be configured as a wireless network. This wireless network includes a wireless PAN (Personal Area Network) such as Bluetooth (registered trademark), ZigBee (registered trademark), and UWB (Ultra Wide Band). It also includes a wireless LAN (Local Area Network) such as Wi-Fi (Wireless Fidelity) (registered trademark) and a wireless MAN (Metropolitan Area Network) such as WiMAX (registered trademark). Furthermore, wireless WANs (Wide Area Networks) such as LTE/3G, 4G, and 5G are included. It should be noted that the network connects each device so as to be able to communicate with each other, and the communication standard, scale, and configuration are not limited to those described above.

At least some of the functions of the elements of the keyword identification device 1 shown in FIG. 1 can be realized by the CPU 81 executing a program. However, at least some of the functions of the elements of the keyword identification device 1 shown in FIG. 1 may operate as dedicated hardware. In this case, the dedicated hardware operates under the control of the CPU 81 .

As described above, according to the present embodiment, the keyword identification device divides a query to be input to a search engine into individual tokens that constitute the query, and creates a single token consisting of one or more tokens. is generated as a subset of the query, and for each generated single, a query containing the single is generated.
The keyword identification device according to the present embodiment further refers to the query model generated based on the user's behavior history, calculates the similarity of each generated query to the original query, and from the calculated similarity , the score of the single is calculated, and the single with the higher calculated score is specified as a keyword that more closely matches the user's intention.

Therefore, without predefining each of the tokens and singles included in the query, to flexibly and easily identify keywords that more represent the user's intentions from among the singles (keyphrases and keywords) that make up the query. can be done.
As a result, a desired search result is presented in response to the query, contributing to improvement in search accuracy.

It should be noted that although specific embodiments are described above, the embodiments are merely examples and are not intended to limit the scope of the invention. The apparatus and methods described herein may be embodied in forms other than those described above. Also, appropriate omissions, substitutions, and modifications may be made to the above-described embodiments without departing from the scope of the invention. Forms with such omissions, substitutions and modifications are included in the scope of what is described in the claims and their equivalents, and belong to the technical scope of the present invention.

1 keyword identification device 2 user action history DB 3 search engine 4 query model storage unit 11 query input unit 12 query division unit 13 query model generation unit 14 single evaluation unit 15... Output unit, 81... CPU, 82... ROM, 83... RAM, 84... HDD, 85... Input unit, 86... Display unit, 87... Communication I/F, 88... Bus

Claims (13)

a first query acquisition unit that acquires a first query;
a query subset generation unit that divides the first query acquired by the first query acquisition unit into a plurality of tokens and generates a plurality of query subsets each composed of one or more divided tokens;
a second query acquisition unit that acquires a second query that shares the query subset generated by the query subset generation unit with the first query ;
Similarity for calculating a degree of similarity of the second query to the first query based on an operation history of items on which a user has performed operations in the past in response to the first query or the second query a degree calculation unit;
a score calculation unit that calculates a score of the query subset based on the similarity of the second query calculated by the similarity calculation unit;
and a query subset identifying unit that identifies, as a keyword , a query subset having a higher score calculated by the score calculating unit among the plurality of query subsets. The information according to claim 1, wherein the similarity calculation unit calculates the similarity as a degree of similarity between the search result set of the second query and the search result set of the first query. processing equipment. Based on the user's action history, the first query or the second query is connected to the attribute of the query with a link, and a weight is given to the link to generate a query model, and the generated further comprising a query model generation unit that stores the query model in a storage device,
The information processing according to claim 1 or 2, wherein the similarity calculation unit calculates the similarity of the second query by referring to the query model stored in the storage device. Device. The similarity calculation unit calculates the weight given to the link connected to the attribute shared between the query model of the first query and the query model of the second query , calculating the similarity of the second query. The similarity calculation unit calculates a plurality of the weights respectively given to the pairs of the links connected to the attributes shared between the query model of the first query and the query model of the second query. The information processing apparatus according to claim 4, wherein the similarity of the second query is calculated by comparing and adding the weight having a small value for a plurality of the attributes. The similarity calculation unit calculates the similarity of the second query by comparing information associated with an item on which the user has performed an operation in the past, between the first query and the second query. 6. The information processing apparatus according to any one of claims 1 to 5, wherein a is calculated. 3. The query model generation unit generates the query model by setting information associated with the item as the attribute and assigning the weight based on the frequency of the operation. 6. The information processing device according to any one of 3 to 5. 8. The score of the query subset according to any one of claims 1 to 7, wherein the score calculation unit calculates the score of the query subset by calculating the geometric mean of the similarities of all the second queries. The information processing device described. The second query acquisition unit acquires the second query for each of the plurality of query subsets generated by the query subset generation unit;
The information processing apparatus according to any one of claims 1 to 8, wherein the score calculation unit calculates the score for each of the plurality of query subsets. The information processing according to any one of claims 1 to 9, wherein the query subset generation unit generates the query subset from a plurality of discontinuous tokens in the character string of the first query. Device. 11. Any one of claims 1 to 10, further comprising an output control unit that controls output so that search results corresponding to the query subset identified by the query subset identification unit are preferentially presented. The information processing device according to . An information processing method executed by an information processing device,
obtaining a first query;
dividing the obtained first query into a plurality of tokens and generating a plurality of query subsets composed of one or more divided tokens;
obtaining a second query that shares the generated query subset with the first query ;
calculating the degree of similarity of the second query to the first query based on an operation history of items that the user has operated in the past in response to the first query or the second query; and,
calculating a score for the query subset based on the similarity of the second query;
and specifying , as a keyword , a query subset having a higher calculated score among the plurality of query subsets. An information processing program for causing a computer to execute information processing, the program causing the computer to:
a first query acquisition process for acquiring a first query;
a query subset generation process for dividing the first query acquired by the first query acquisition process into a plurality of tokens and generating a plurality of query subsets each composed of one or more divided tokens;
a second query acquisition process for acquiring a second query that shares the query subset generated by the query subset generation process with the first query ;
A degree of similarity for calculating the degree of similarity of the second query to the first query based on an operation history of items on which a user has performed operations in the past in response to the first query or the second query. a calculation process;
A score calculation process for calculating a score of the query subset based on the similarity of the second query calculated by the similarity calculation process;
and a query subset identification process for identifying, as a keyword , a query subset with a higher score calculated by the score calculation process among the plurality of query subsets. processing program.

Images