JP6865663B2 - Information processing equipment, information processing methods, and programs - Google Patents

Description

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

In the field of web search, a method of providing information on various products or services (hereinafter referred to as "products, etc.") related to this query is performed according to a keyword entered by a user (hereinafter referred to as "query"). ing. In addition, in order to provide information on products, etc. intended by the user with high accuracy, categories such as products, etc. corresponding to the query are defined in advance as tree structure data between categories, and each category is defined in advance based on the tree structure data. A method for calculating the similarity between categories for a query is known (see, for example, Patent Document 1).

JP-A-2017-45196

However, in the conventional method, the degree of relevance to the query is low, and information such as a product not intended by the user may be provided. For example, when the query "TV" is provided with the product information of the "TV stand", or the query "beer" is provided with the product information of the "glass", the user intended. In some cases, information on products, etc. that belong to a category different from the products, etc. was provided.

The present invention has been made in consideration of such circumstances, and provides an information processing device, an information processing method, and a program capable of improving search accuracy by estimating a category more accurately. That is one of the purposes.

In one aspect of the present invention, the relationship between the acquisition unit that acquires the query input by the user and the query acquired by the acquisition unit is represented by a tree structure, and each of the queries and the search target. When the determination unit that determines whether or not the category of is present in the associated category dictionary data and the determination unit determines that the query acquired by the acquisition unit exists in the category dictionary data, the category dictionary When the first estimation unit that estimates the category corresponding to the query acquired by the acquisition unit and the determination unit determines that the query acquired by the acquisition unit does not exist in the category dictionary data based on the data. , The second estimation unit that estimates the category corresponding to the query acquired by the acquisition unit based on the model that learned the relationship between the query used for the search and the category, and the first estimation unit or the second estimation unit. It is an information processing device including a generation unit that generates information for displaying search results belonging to the category estimated by the unit.

According to one aspect of the present invention, the search accuracy can be improved by accurately estimating the category.

It is a functional block diagram of the search server 1 of 1st Embodiment. It is a figure which shows an example of the category tree data D2 of 1st Embodiment. It is a figure which shows an example of the word tree data D3 of 1st Embodiment. It is a flowchart which shows an example of the flow of the search process of the search server 1 of 1st Embodiment. It is a figure explaining the outline of an example of the search process of the search server 1 of 1st Embodiment. It is a figure which shows an example of the relationship between the query processed by the search server 1 of 1st Embodiment, and the estimated category. It is a figure which shows an example of the category tree data D4 of 1st Embodiment. It is a figure which shows an example of the word tree data D5 of 1st Embodiment. It is a functional block diagram of the search server 2 of the 2nd Embodiment. It is a flowchart which shows an example of the flow of the search process of the search server 2 of 2nd Embodiment.

Hereinafter, an information processing apparatus, an information processing method, and an embodiment of a program will be described with reference to the drawings.

<1. First Embodiment>
Hereinafter, the first embodiment of the present invention will be described. The information processing apparatus of the present embodiment separates the query used for the search into words (hereinafter, also referred to as "words"), and uses the query and the query based on the word tree data in which the predefined words and categories are associated with each other. The corresponding category is estimated, and information (search target) such as products belonging to this estimated category is output as a search result. The "category" in the present embodiment is information indicating a field to which the content (product or service) of the content provided as a search result for the query used for the search belongs. In the present embodiment, an example in which the information processing device is a search server that provides search results for contents such as products will be described.

<1-1. System configuration>
FIG. 1 is a functional block diagram of the search server 1 (information processing device) of the present embodiment. The search server 1 is connected to the terminal device T or the like via the network NW. The network NW includes, for example, WAN (Wide Area Network), LAN (Local Area Network), the Internet, a dedicated line, a wireless base station, a provider, and the like.

The terminal device T is operated by a user who uses the search service provided by the search server 1. The terminal device T is, for example, a personal computer, a mobile phone such as a smartphone, a tablet terminal, or a computer device such as a PDA (Personal Digital Assistant). In the terminal device T, a browser or application program that operates based on the user's operation sends, for example, a query requesting information provision to the search server 1, and receives information associated with the query from the search server 1.

The search server 1 transmits the page information of the search result associated with the query input from the terminal device T to the terminal device T. The search server 1 generates page information of the search result by using the search data in which the query and the reference information for referencing the page of the site associated with the query are associated with each other. The reference information includes, for example, a URL (Uniform Resource Locator).

The search server 1 includes, for example, a communication unit 10 (acquisition unit), a query separation unit 12 (separation unit), a determination unit 14, a first category estimation unit 16 (estimation unit), and a search result generation unit 18 (generation unit). ) And a storage unit 20. The communication unit 10 communicates with the terminal device T or the like via the network NW. The communication unit 10 receives a query from the terminal device T via the network NW, and transmits the page information of the search result to the terminal device T. The communication unit 10 includes, for example, a communication interface such as a NIC.

The storage unit 20 describes, for example, search data D1 including reference information for referencing a page of a site associated with a query, category tree data D2 representing a relationship between categories in a tree structure, and a relationship between queries. It stores word tree data D3 and the like, which are represented by a tree structure and include information on categories corresponding to each of the queries. The storage unit 20 is realized by, for example, a RAM (Random Access Memory), a ROM (Read Only Memory), an HDD (Hard Disk Drive), a flash memory, or a hybrid storage device in which a plurality of these is combined. Further, a part or all of the storage unit 20 may be an external device such as NAS or an external storage server that can be accessed by the search server 1.

The query separation unit 12 analyzes the character string of the query acquired from the terminal device T and separates it into words. The query separation unit 12 separates a query into words, for example, using a space, a comma, or the like included in the query as a delimiter.

The determination unit 14 determines whether or not the words separated by the query separation unit 12 are defined in the word tree data D3.

The first category estimation unit 16 sets the category corresponding to the word separated by the query separation unit 12 as the category corresponding to the query acquired from the terminal device T based on the word tree data D3 read from the storage unit 20. presume. The first category estimation unit 16 determines the query acquired from the terminal device T based on the result of tracing the query acquired from the terminal device T to the corresponding query in order from the query located at the highest level of the word tree data D3. Estimate the corresponding category. The details of the category estimation process by the first category estimation unit 16 will be described later.

The search result generation unit 18 performs a narrowing search using the categories estimated by the first category estimation unit 16 and generates a search result page which is a search result. For example, the search result generation unit 18 performs a narrowing search using the estimated category for the search data D1 stored in the storage unit 20, and displays a search result page containing information such as products belonging to the estimated category. Generate. The search result generation unit 18 transmits the generated search result page to the terminal device T via the communication unit 10.

Each functional unit of the search server 1 is realized, for example, by executing a program (software) by a hardware processor such as a CPU (Central Processing Unit) in a computer. In addition, some or all of these components are hardware (circuits) such as LSI (Large Scale Integration), ASIC (Application Specific Integrated Circuit), FPGA (Field-Programmable Gate Array), GPU (Graphics Processing Unit), etc. It may be realized by the part; including circuitry), or it may be realized by the cooperation of software and hardware.

<1-2. Word tree data>
The word tree data D3 will be described below. FIG. 2 is a diagram showing an example of category tree data D2, and FIG. 3 is a diagram showing an example of word tree data D3 defined based on category tree data D2. The word tree data D3 is defined manually, for example, by the designer of the search server 1.

In the category tree data D2 shown in FIG. 2, "home appliances (A)" is defined as the category of the first layer L1, and the category of the second layer L2 located below the first layer L1 "home appliances (A)". "Air conditioner (A: A)" is defined as "Air conditioner (A: A)", and "Air conditioner (A: A)" is defined as the category of the third layer L3 located below the second layer L2 "Air conditioner air conditioner (A: A)". : A) ”is defined, and as a category of the fourth layer L4 located below the third layer L3“ air conditioner (A: A: A) ”,“ air conditioner (for home use) (A: A: A: A) , "Air conditioner for windows (A: A: A: B)", "Air conditioner filter (A: A: A: C)", "Remote control for air conditioner (A: A: A: D)", and "Other air conditioners" Parts accessories (A: A: A: E) ”are defined. The symbols in parentheses indicate the category ID that identifies the category.

Further, in this category tree data D2, it is shown which of the categories of "main body", "quasi-main body", and "peripheral device" is associated with each of the categories defined in the fourth layer L4. The main body flag F is set. For example, the main body flag F "main body" is associated with the "air conditioner (for home use) (A: A: A: A)".

The word tree data D3 is defined based on the category tree data D2. For example, the word tree data D3 is data (category dictionary data) that associates a query that is expected to be used for a search (a query that is frequently used) with a category defined in the category tree data D2. is there. In the word tree data D3 shown in FIG. 3, the category ID “A: A: A” is associated with the query “air conditioner” as the first layer H1. This category ID "A: A: A" corresponds to the "air conditioner (A: A: A)" of the third layer L3 in the category tree data D2 shown in FIG.

Further, in this word tree data D3, the category ID "A: A: A: A" is associated with the query "main body" of the second layer H2 located below the first layer H1 "air conditioner". , The category ID "A: A: A: B" is associated with the query "for windows", and the category ID "A: A: A: C" is associated with the query "filter". The category ID "A: A: A: D" is associated with the query "remote control", and the category ID "A: A: A: E" is associated with the query "part".

In this word tree data D3, in addition to the above query, the category ID "A: A: A: B" is associated with the query "window air conditioner" as the first layer H1, and the first layer. The category ID "A: A: A: C" is associated with the query "air conditioner filter" as the first layer H1. Further, as including similar queries, the category ID "X: X" is associated with the query "automobile" as the first layer H1, and the second layer located below the first layer H1 "automobile". The category ID "X: X: X: Y" is associated with the H2 query "air conditioner filter".

<1-3. Search process>
The search process of the search server 1 will be described below. FIG. 4 is a diagram showing an example of the flow of the search process of the search server 1. FIG. 5 is a diagram illustrating an outline of an example of a search process of the search server 1.

First, the query separation unit 12 acquires the query input from the terminal device T via the communication unit 10 (S101). Next, the query separation unit 12 analyzes the query acquired from the terminal device T and separates it into words (S103). The query separation unit 12 separates a query into words, for example, using a space, a comma, or the like included in the query as a delimiter.

In the example shown in FIG. 5, the query separation unit 12 acquires the query “air conditioner for windows” entered by the user via the terminal device T on the search page, and uses this query as a space delimiter for the word 1 “for windows”. And the word 2 "air conditioner" are separated into two words.

Next, the determination unit 14 determines whether or not the words separated by the query separation unit 12 are defined in the first layer H1 (top layer) of the word tree data D3 (S105). In the example shown in FIG. 5, the word 1 "for window" does not exist in the first layer H1 of the word tree data D3 shown in FIG. 3, but the word 2 "air conditioner" is in the first layer H1 of the word tree data. Exists. Further, the word 1 "for windows" exists as the second layer H2 located below the word 2 "air conditioner". In this case, the determination unit 14 determines that the words separated by the query separation unit 12 are defined in the first layer H1 of the word tree data D3.

On the other hand, when all the words separated by the query separation unit 12 do not exist in the first layer H1 of the word tree data D3, the determination unit 14 determines that the words separated by the query separation unit 12 are the third word of the word tree data D3. It is determined that it is not defined in the first layer H1.

When the determination unit 14 determines that the words separated by the query separation unit 12 are defined in the first layer H1 of the word tree data D3, the first category estimation unit 16 is based on the word tree data D3. The category corresponding to the word separated by the query separation unit 12 is estimated as the category corresponding to the query acquired from the terminal device T (S107). In the example shown in FIG. 5, the category ID “A: A: A: B” associated with the word 1 “for windows” in the second layer H2 below the word 2 “air conditioner” is the “window air conditioner”. It is presumed to be the category intended by the user.

Next, the search result generation unit 18 performs a narrowing search in the category estimated by the first category estimation unit 16 with respect to the query acquired from the terminal device T (S109). The search result generation unit 18 deletes the words determined by the determination unit 14 in the word tree data D3 from the words included in the query acquired from the terminal device T, and searches for the remaining words. As a result, the search data D1 is searched for products and the like belonging to the estimated category.

In the example shown in FIG. 5, it is determined that both the word 1 "for window" and the word 2 "air conditioner" included in the query acquired from the terminal device T are defined in the word tree data D3. In this case, since the search result generation unit 18 deletes the word 1 and the word 2, the search query for the search data D1 does not exist. In this case, the search result generation unit 18 does not specify the search query, but specifies only the category ID "A: A: A: B (air conditioner for windows)" estimated to be the category intended by the user, and the search data. Search for D1.

Next, the search result generation unit 18 generates a search result page based on the search result using the estimated category, and transmits it to the terminal device T (S111). When the determination unit 14 determines that the words separated by the query separation unit 12 are not defined in the word tree data D3, the search result generation unit 18 does not perform the above-mentioned category estimation process, and the terminal device. A search result page is generated by performing a normal search process (search process that does not narrow down the category) using the query obtained from T, and is transmitted to the terminal device T. This completes the processing of this flowchart.

FIG. 6 is a diagram showing an example of the relationship between the query processed by the search server 1 and the estimated category. When the query obtained from the terminal device T is "air conditioner", it is assumed that the user who inputs this query intends to search for the "main body" of the air conditioner. Therefore, the flag of "main body narrowing down" is set in the "air conditioner (A: A: A)" of the first layer H1 in the word tree data D3 shown in FIG. This "main unit narrowing down" flag is the "main unit flag F" set in the category "air conditioner (household) (A: A: A: A)" of the fourth layer L4 of the category tree data D2 shown in FIG. It is associated with the "body". Therefore, when the query acquired from the terminal device T is "air conditioner", the first category estimation unit 16 sets "air conditioner (for home use) (A: A: A: A)" as the category intended by the user. presume.

Then, the search result generation unit 18 deletes the query "air conditioner" determined to be defined in the word tree data D3, and deletes the category "air conditioner (for home use)" determined to be the category intended by the user (A: A). : A: A) ”is searched for and a search result page is generated.

Further, when the query acquired from the terminal device T is the "air conditioner component remote controller", the query "air conditioner" exists as the first layer H1 in the word tree data D3, and the second query "air conditioner" is located below the query "air conditioner". There are both queries "parts" and "remote control" as hierarchy H2. In this case, the first category estimation unit 16 has the category ID "A: A: A: E" associated with the query "parts" of the second layer H2, "other air conditioner parts accessories", and the second layer H2. It is estimated that both the query "remote controller" and the category ID "A: A: A: D" associated with the "air conditioner remote controller" are the categories intended by the user.

Then, the search result generation unit 18 includes the query "air conditioner" determined to be defined as the first layer H1 and the queries "parts" and "remote controller" determined to be defined as the second layer H2. Information on products that belong to the category "Other air conditioner parts accessories (A: A: A: E)" or the category "Remote control for air conditioner (A: A: A: D)" that is determined to be the category intended by the user. Search (OR search) to generate a search result page.

Further, when the query acquired from the terminal device T is "window air conditioner company A", the query "air conditioner" exists as the first layer H1 in the word tree data D3, and is located below the query "air conditioner". There is a query "for windows" as the second layer H2. In this case, the first category estimation unit 16 sets the category ID "A: A: A: B" associated with the query "for windows" of the second layer H2 as "window air conditioner" as the category intended by the user. presume.

Then, the search result generation unit 18 deletes the word query "air conditioner" determined to be defined as the first layer H1 and the query "for windows" determined to be defined as the second layer H2. Then, using the remaining query "Company A" as a search query, information on products belonging to the "window air conditioner" category is searched, and a search result page is generated.

Further, when the query acquired from the terminal device T is a "filter", this query does not exist in the first layer H1 of the word tree data D3. In this case, the search result generation unit 18 performs a normal search process using the query "filter" without performing the above-mentioned category estimation process, and generates a search result page.

Further, when the query acquired from the terminal device T is "automobile air conditioner filter deodorization", this query includes the "air conditioner filter (A: A: A: C)" existing in the first layer H1 of the word tree data D3. ", And there are" automobiles (X: X) ". Further, the query "air conditioner filter (X: X: X: Y)" exists as the second layer H2 located below the query "automobile". In this case, the first category estimation unit 16 estimates that both the "air conditioner filter (A: A: A: C)" and the "air conditioner filter (X: X: X: Y)" are the categories intended by the user. ..

In this case, it is difficult to determine which of the two estimated categories "air conditioner filter (A: A: A: C)" and "air conditioner filter (X: X: X: Y)" should be prioritized. Not stick. Therefore, the search result generation unit 18 does not delete the query determined to be defined in the word tree data D3, but uses "automobile air conditioner filter deodorization" as the search query and "air conditioner filter (A: A: A)". : C) ”or“ Air conditioner filter (X: X: X: Y) ”is searched (OR search) to generate a search result page. That is, when the query acquired from the terminal device T includes two or more queries defined in the first layer H1 of the word tree data D3, the search result generation unit 18 uses the word tree. The search process for the search data D1 is performed using the query acquired from the terminal device T without deleting the query determined to be defined in the data D3.

Further, under the condition that the category tree data D4 as shown in FIG. 7 is defined, when the query acquired from the terminal device T is the "down jacket", this query "down jacket" is the men's "down jacket". I can't tell whether it's intended as a "jacket" or a ladies' "down jacket." In this case, as shown in the word tree data D5 shown in FIG. 8, the query "down jacket" is defined as the first layer H1, and "men's" and the query of the second layer H2 located below the "down jacket" are defined as "men's". Define "ladies". As a result, when the query obtained from the terminal device T is a "down jacket", the search result generation unit 18 is based on the word tree data D5, and is a men's "down jacket (B: A: A)" or a ladies'. You can search (OR search) products that belong to the "down jacket (B: B: A)" of.

According to the search server 1 of the present embodiment described above, the category corresponding to the query can be appropriately estimated, and the search accuracy can be improved. In addition, by deleting the words determined to be defined in the word tree data and performing a refined search using the estimated category, the provider of the product etc. can obtain the information of his / her own product etc. on the search result page. It is not necessary to add words that are expected to be used for search to the names of products, etc. for the purpose of being displayed at the top. In addition, by making it possible to narrow down the categories, when displaying products in ascending order of price, information on products in unintended categories will not be displayed, improving the convenience of the search site. be able to.

In the above, examples of shopping sites in the fields of home appliances and fashion have been described, but if the service involves information retrieval processing, such as shopping sites for cars and motorcycles, shopping sites for game machines, auction sites, etc. The embodiment is applicable. Further, in the word tree data, an example in which each query has one category corresponding to each query has been described, but one query may be associated with a plurality of categories.

<2. Second Embodiment>
Hereinafter, the second embodiment of the present invention will be described. Compared with the first embodiment, the search server of the present embodiment has a second category estimation unit that estimates a category based on machine learning, in addition to a first category estimation unit that estimates a category based on word tree data. The difference is that it is further provided. Therefore, for the configuration and the like, the drawings and related descriptions described in the first embodiment will be referred to, and detailed description will be omitted.

<2-1. System configuration>
FIG. 9 is a functional block diagram of the search server 2 (information processing device) of the present embodiment. The search server 2 further includes a second category estimation unit 22 (second estimation unit) as compared with the search server 1 of the first embodiment. The first category estimation unit 16 (first estimation unit) performs category estimation processing based on a word tree (category dictionary data) created for frequently used queries, so that the query defined in the word tree is used. The estimation accuracy is high, and it is possible to handle queries that specify multiple categories. However, the first category estimation unit 16 cannot be used for queries that are not defined in the word tree (that is, queries that are expected to be used moderately or infrequently, such as product names). Therefore, in the present embodiment, the second category estimation unit 22 performs category estimation processing based on machine learning for queries that are not defined in the word tree. The second category estimation unit 22 includes, for example, a terminal category estimation unit 24 and an intermediate category estimation unit 26.

The terminal category estimation unit 24 estimates the terminal (bottom layer) category in the category tree. For example, the terminal category estimation unit 24 has the query acquired from the terminal device T and the terminal category (intentionally intended by the user) having the largest number of clicks among the categories clicked by the user on the search result page provided in response to the query. By learning in advance the relationship with the category (correct answer category) considered to be, the first model (hereinafter referred to as “terminal category model M1”) is generated and stored in the storage unit 20. The terminal category estimation unit 24 estimates the terminal category corresponding to the query acquired from the terminal device T based on the terminal category model M1. In the estimation process by the terminal category estimation unit 24, the number of processes of the query acquired from the terminal device T (or the number of clicks on the search result page provided in response to the query acquired from the terminal device T) is sufficiently large. In some cases, it is assumed that the estimation accuracy is high.

The intermediate category estimation unit 26 estimates the category of the intermediate layer that can include a plurality of lower categories. For example, the intermediate category estimation unit 26 has a query obtained from the terminal device T and an intermediate category (intention of the user) having the largest number of clicks among the categories clicked by the user on the search result page provided in response to the query. A second model (hereinafter referred to as "intermediate category model M2") is generated by learning in advance the relationship with the category (correct answer category) that is considered to have been generated, and is stored in the storage unit 20. The intermediate category estimation unit 26 estimates an intermediate category corresponding to the query acquired from the terminal device T based on the intermediate category model M2.

The category estimation process by the terminal category estimation unit 24 described above may reduce the estimation accuracy for a query targeting a category that is divided into ladies and men. As described above, when the accuracy is expected to decrease when the terminal category estimation unit 24 described above is used, the intermediate category estimation unit 26 capable of including a plurality of lower categories is used. .. For example, when the query obtained from the terminal device T is "duffle coat", the intermediate category estimation unit 26 estimates a category that is "men's" and a category that is "ladies". Next, the search result generation unit 18 narrows down by the estimated category "men's" or "ladies" and uses the search query "duffle coat" for search processing (OR search for categories "men's" and "ladies"). I do.

The terminal category estimation unit 24 and the intermediate category estimation unit 26 include a multi-layered neural network (Deep Neural Network: DNN), a convolutional neural network (CNN), a recurrent neural network (RNN), and the like. The deep learning technique used may be adopted.

<2-2. Search process>
The search process of the search server 2 will be described below. FIG. 10 is a diagram showing an example of the flow of the search process of the search server 2.

First, the query separation unit 12 acquires the query input from the terminal device T via the communication unit 10 (S201). Next, the query separation unit 12 analyzes the query acquired from the terminal device T and separates it into words (S203).

Next, the determination unit 14 determines whether or not the words separated by the query separation unit 12 are defined in the word tree data D3 (S205).

When the determination unit 14 determines that the words separated by the query separation unit 12 are defined in the word tree data D3, the first category estimation unit 16 is determined by the query separation unit 12 based on the word tree data D3. The category corresponding to the separated words is estimated as the category corresponding to the query obtained from the terminal device T (S207).

On the other hand, when the determination unit 14 determines that the word separated by the query separation unit 12 is not defined in the word tree data D3, the determination unit 14 further processes the same number of queries in the past with respect to the query acquired from the terminal device T. Is equal to or greater than the threshold value (S213). This threshold value (an example of accuracy information) serves as a reference indicating the reliability of the estimation processing of the terminal category corresponding to the query acquired from the terminal device T. If the number of processed queries of the query acquired from the terminal device T is equal to or greater than the threshold value, it is assumed that the terminal category has been sufficiently learned for this query and the accuracy of the estimation processing by the terminal category estimation unit 24 is high. On the other hand, if the number of queries processed from the terminal device T is less than the threshold value, it is assumed that the learning of the terminal category for this query is insufficient and the accuracy of the estimation processing by the terminal category estimation unit 24 is low. The number of processes for each query in the past is included in, for example, the search log L stored in the storage unit 20.

When the determination unit 14 determines that the number of processed queries acquired from the terminal device T is equal to or greater than the threshold value, the terminal category estimation unit 24 determines the terminal category corresponding to the acquired query based on the terminal category estimation model M1. Estimate (S215). On the other hand, when the determination unit 14 determines that the number of processed queries acquired from the terminal device T is not equal to or greater than the threshold value, the intermediate category estimation unit 26 is a terminal corresponding to the acquired query based on the intermediate category estimation model M2. Estimate the category (S217).

Next, the search result generation unit 18 narrows down the query acquired from the terminal device T by the category estimated by any of the first category estimation unit 16, the terminal category estimation unit 24, and the intermediate category estimation unit 26. Do (S209). When performing a refined search using the categories estimated by the first category estimation unit 16, the search result generation unit 18 determines that the queries included in the acquired queries are defined in the word tree data D3. The executed query is deleted, and the remaining queries are used as search queries to search for information such as products belonging to the presumed category.

Next, the search result generation unit 18 generates a search result page based on the search result using the estimated category and transmits it to the terminal device T (S211). This completes the processing of this flowchart.

According to the search server 2 of the present embodiment described above, the category corresponding to the query can be appropriately estimated, and the search accuracy can be improved. When the category cannot be estimated based on the word tree data, the search accuracy can be further improved by estimating the category based on machine learning. When a plurality of categories are estimated by the intermediate category estimation unit 26, it is also possible to provide search results for each category.

According to the embodiment described above, the relationship between the acquisition unit that acquires the query input by the user and the query acquired by the acquisition unit is represented by a tree structure, and each of the queries. When the determination unit that determines whether or not the category to be searched and the category to be searched exists in the associated category dictionary data and the determination unit determines that the query acquired by the acquisition unit exists in the category dictionary data. If the first estimation unit that estimates the category corresponding to the query acquired by the acquisition unit and the query acquired by the acquisition unit by the determination unit do not exist in the category dictionary data based on the category dictionary data. When it is determined, the second estimation unit that estimates the category corresponding to the query acquired by the acquisition unit based on the model that learned the relationship between the query used for the search and the category, and the first estimation unit or By providing a generation unit that generates information for displaying search results belonging to the category estimated by the second estimation unit, it is possible to accurately estimate the category and improve the search accuracy.

Although the embodiments for carrying out the present invention have been described above using the embodiments, the present invention is not limited to these embodiments, and various modifications and substitutions are made without departing from the gist of the present invention. Can be added.

1, 2 ... Search server (information processing device)
10 ... Communication department (acquisition department)
12 ... Query separation part (separation part)
14 ... Judgment unit 16 ... First category estimation unit (first estimation unit)
18 ... Search result generation unit (generation unit)
20 ... Storage unit 22 ... Second category estimation unit (second estimation unit)
24 ... Terminal category estimation unit 26 ... Intermediate category estimation unit

Claims (9)

An acquisition unit that acquires the query entered by the user, and
Judgment of determining whether or not the query acquired by the acquisition unit has a tree structure in which the relationships between the queries are represented and each of the queries and the category to be searched are associated with each other in the category dictionary data. Department and
When the determination unit determines that the query acquired by the acquisition unit exists in the category dictionary data, the first estimation estimates the category corresponding to the query acquired by the acquisition unit based on the category dictionary data. Department and
When the determination unit determines that the query acquired by the acquisition unit does not exist in the category dictionary data, it is acquired by the acquisition unit based on the model that learned the relationship between the query used for the search and the category. A second estimation unit that estimates the query and the corresponding category,
An information processing device including a generation unit that generates information for displaying search results belonging to a category estimated by the first estimation unit or the second estimation unit. The second estimation unit
A terminal category estimation unit that corresponds to the query acquired by the acquisition unit and estimates the terminal category located at the end of the tree structure.
It includes an intermediate category estimation unit that corresponds to the query acquired by the acquisition unit and estimates an intermediate category located in the intermediate layer of the tree structure.
The information processing device according to claim 1. The determination unit determines whether the terminal category estimation unit or the intermediate category estimation unit is to perform the estimation processing based on the accuracy information of the first model used for the estimation processing of the terminal category estimation unit. ,
The information processing device according to claim 2. The determination unit determines whether or not the number of processed queries acquired by the acquisition unit is equal to or greater than the threshold value.
When the determination unit determines that the number of processed queries acquired by the acquisition unit is equal to or greater than the threshold value, the determination unit causes the terminal category estimation unit to perform estimation processing.
When the determination unit determines that the number of processed queries acquired by the acquisition unit is not equal to or greater than the threshold value, the determination unit causes the intermediate category estimation unit to perform estimation processing.
The information processing device according to claim 3. Further provided with a separator that separates the query acquired by the acquirer into words.
Based on the category dictionary data, the first estimation unit estimates the category corresponding to the word separated by the separation unit as the category corresponding to the query acquired by the acquisition unit.
The information processing device according to any one of claims 1 to 4. When the determination unit determines that the query acquired by the acquisition unit exists in the category dictionary data, the generation unit determines that the words separated by the separation unit exist in the category dictionary data. Is deleted, and information for displaying the search result is generated based on the result of searching the storage device that stores the information of the search target using the remaining words.
The information processing device according to claim 5. The category dictionary data is defined based on a category tree that represents the relationships between the categories to be searched in a tree structure.
The information processing device according to any one of claims 1 to 6. The computer
Gets the query entered by the user and
It is determined whether or not the acquired query represents the relationship between the queries in a tree structure and exists in the category dictionary data in which each of the queries and the category to be searched are associated with each other.
When it is determined that the acquired query exists in the category dictionary data, the category corresponding to the acquired query is estimated based on the category dictionary data.
When it is determined that the acquired query does not exist in the category dictionary data, the category corresponding to the acquired query is estimated based on the model learned the relationship between the query used for the search and the category. ,
An information processing method for generating information for displaying search results belonging to a category determined based on the category dictionary data or a category determined based on the model. On the computer
Get the query entered by the user
The acquired query is made to determine whether or not the relationship between the queries is represented in a tree structure and each of the queries and the category to be searched exist in the associated category dictionary data.
When it is determined that the acquired query exists in the category dictionary data, the category corresponding to the acquired query is estimated based on the category dictionary data.
When it is determined that the acquired query does not exist in the category dictionary data, the category corresponding to the acquired query is estimated based on the model that learned the relationship between the query used for the search and the category. ,
A program that generates information for displaying search results belonging to a category determined based on the category dictionary data or a category determined based on the model.

Images