JP6853009B2 - Aggregation device, aggregation method, and aggregation program - Google Patents

Description

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

Conventionally, a technique for collecting information about a search target is provided based on a search query (hereinafter, also simply referred to as “query”) used in a search system. For example, a query may include a plurality of character strings (hereinafter, also referred to as "tokens") separated by a predetermined delimiter. Further, for example, a technique for classifying a predetermined object into a class is provided. For example, by the technique of classifying a predetermined object into a class, a label related to the content (meaning) of the token is attached.

C.N. Silla Jr., A.A. Freitas, "A survey of hierarchical classification across different application domains", Data Mining and Knowledge Discovery, 22 (1-2): 182-196, January 2011

However, in the above-mentioned prior art, when the search query includes tokens with a plurality of labels, it is not always possible to appropriately total the number of searches. For example, simply counting the number of searches based on the entire query may not be able to properly count how many searches were performed on which target when multiple labeled tokens are included. ..

The present application has been made in view of the above, and provides an aggregation device, an aggregation method, and an aggregation program that appropriately aggregates the number of searches even when tokens with a plurality of labels are included. The purpose is.

The aggregation device according to the present application includes an acquisition unit for acquiring information on each token included in a search query and having a label related to a place name or a label related to a category other than the place name, and each token. It is characterized by including an aggregation unit that aggregates information on the number of searches by the search query for each combination of attached labels.

According to one aspect of the embodiment, there is an effect that the number of searches can be appropriately totaled even when tokens with a plurality of labels are included.

FIG. 1 is a diagram showing an example of aggregation processing according to the embodiment. FIG. 2 is a diagram showing a configuration example of the aggregation device according to the embodiment. FIG. 3 is a diagram showing an example of the search history information storage unit according to the embodiment. FIG. 4 is a diagram showing an example of the token information storage unit according to the embodiment. FIG. 5 is a diagram showing an example of the aggregated information storage unit according to the embodiment. FIG. 6 is a flowchart showing the aggregation processing procedure according to the embodiment. FIG. 7 is a diagram showing a configuration example of the aggregation device according to the modified example. FIG. 8 is a diagram showing an example of the aggregated information storage unit according to the modified example. FIG. 9 is a diagram showing an example of a user information storage unit according to a modified example. FIG. 10 is a hardware configuration diagram showing an example of a computer that realizes the function of the tabulation device.

Hereinafter, the tabulation apparatus, the tabulation method, and the mode for carrying out the tabulation program (hereinafter, referred to as “embodiment”) according to the present application will be described in detail with reference to the drawings. It should be noted that this embodiment does not limit the aggregation device, aggregation method, and aggregation program according to the present application. Further, in each of the following embodiments, the same parts are designated by the same reference numerals, and duplicate description is omitted.

(Embodiment)
[1. Aggregation process]
First, an example of the aggregation process according to the embodiment will be described with reference to FIG. FIG. 1 is a diagram showing an example of aggregation processing according to the embodiment. In FIG. 1, the aggregation device 100 calculates a score (number of searches) for each combination of labels (hereinafter, also referred to as “tags”) attached to tokens for search queries (queries) used in a predetermined search system. The case where the score of the combination of each label is totaled is shown. The "token" here means a character string separated by a predetermined delimiter included in the query used by the user for the search. For example, "token" means each string separated by spaces in the query used by the user for the search. For example, if the search query is "Roppongi gourmet", the character string "Roppongi" or the character string "gourmet" becomes a token.

As shown in FIG. 1, the aggregation system 1 includes a search device 50 and an aggregation device 100. The search device 50 and the tabulation device 100 are communicably connected by wire or wirelessly via a predetermined network. The estimation system 1 shown in FIG. 1 may include a plurality of search devices 50 and a plurality of totaling devices 100.

The search device 50 is an information processing device that provides a predetermined search service. For example, the search device 50 acquires a search query from the terminal device used by the user and provides the terminal device with the search result corresponding to the search query. The search device 50 may be integrated with the totaling device 100.

In the example of FIG. 1, the aggregation device 100 acquires the search query Q11 from the search device 50 (step S10). The aggregation device 100 may acquire the search query Q11 from the terminal device used by the user. Further, in the example of FIG. 1, the search query Q11 is assumed to be the character string "Tomato, Aoba-ku, Yokohama".

First, the aggregation device 100 acquires the token included in the search query Q11 by dividing the search query Q11 into tokens. Specifically, the aggregation device 100 acquires each character string separated by a space (“”) from the search query Q11 as a token. In the example of FIG. 1, the aggregation device 100 acquires the token TK11 which is the character string “Yokohama” from the search query Q11 (step S11-1). Further, in the example of FIG. 1, the aggregation device 100 acquires the token TK12 which is the character string “Aoba Ward” from the search query Q11 (step S11-2). Further, in the example of FIG. 1, the aggregation device 100 acquires the token TK13 which is the character string “tomato” from the search query Q11 (step S11-3).

Then, the aggregation device 100 acquires information about the tags attached to the tokens TK11 to TK13 from the token information storage unit 122 (see FIG. 4). For example, since the tag "Kanagawa" whose category is "place name" is attached to the token TK11, the aggregation device 100 has the tag "Kanagawa" attached to "Yokohama" as shown in the tag TG11-1. To get. As described above, the token related to the place name is tagged with the category "place name", but the token related to the place name may be tagged with the place name by various methods. For example, using the information about the place name extracted from the data such as registration, the token related to the place name may be tagged with the category "place name".

Further, the aggregation device 100 acquires the weight of the tag “Kanagawa” attached to the token TK11 from the token information storage unit 122 (see FIG. 4). For example, the tabulation device 100 acquires the weight "1" of the tag "Kanagawa" attached to "Yokohama" as shown in the weight WT11-1.

Further, in the example of FIG. 1, the aggregation device 100 acquires the tag "Kanagawa" attached to "Aoba Ward" as shown in the tag TG12-1. Further, the tabulation device 100 acquires the tag "Miyagi" attached to "Aoba Ward" as shown in the tag TG12-2.

Further, the aggregation device 100 acquires the weights of the tag “Kanagawa” and the tag “Miyagi” attached to the token TK12 from the token information storage unit 122 (see FIG. 4). For example, the aggregation device 100 acquires the weight "0.5" of the tag "Kanagawa" attached to "Aoba Ward" as shown in the weight WT12-1. Further, for example, the aggregation device 100 acquires the weight "0.5" of the tag "Miyagi" attached to "Aoba Ward" as shown in the weight WT12-2. In the example of FIG. 1, in order to simplify the explanation, the weight of the tag "Kanagawa" and the tag "Miyagi" attached to the token TK12 is set to "0.5", but "Aoba Ward" The weights of the tag "Kanagawa" and the tag "Miyagi" attached to "Aoba Ward" may be changed based on various information such as the relationship with and the relationship with other tokens.

For example, since the search query Q11 has a token TK11 indicating the place name "Kanagawa", the weight of the tag "Kanagawa" attached to "Aoba Ward" may be larger than the weight of the tag "Miyagi". For example, the aggregation device 100 may change the weight of the tag "Kanagawa" and the weight of the tag "Miyagi" attached to "Aoba Ward" in the search query Q11 according to the population of Kanagawa and Miyagi prefectures. Good. For example, when the ratio of the population of Kanagawa prefecture to the population of Miyagi prefecture is "3: 2", the counting device 100 sets the weight of the tag "Kanagawa" attached to "Aoba Ward" of the search query Q11 to "0". It may be set to "0.3" and the weight of the tag "Miyagi" may be set to "0.2". For example, the aggregation device 100 is attached to "Aoba Ward" in the search query Q11 according to the difference between the degree of similarity between "Aoba Ward" and "Kanagawa" and the degree of similarity between "Aoba Ward" and "Miyagi". The weight of the tag "Kanagawa" and the weight of the tag "Miyagi" may be changed. In this case, the counting device 100 calculates the similarity between "Aoba Ward" and "Kanagawa" and the similarity between "Aoba Ward" and "Miyagi" by appropriately using various conventional techniques such as Word2Vec. You may.

Further, for example, since the token TK13 is attached with the tag "restaurant" and the tag "agriculture" whose category is "industry", the totaling device 100 is attached to "tomato" as shown in tag TG13-1. Acquire the attached tag "restaurant business". Further, the tabulation device 100 acquires the tag "agriculture" attached to the "tomato" as shown in the tag TG13-2.

Further, the aggregation device 100 acquires the weights of the tag “restaurant” and the tag “agriculture” attached to the token TK13 from the token information storage unit 122 (see FIG. 4). For example, the counting device 100 acquires the weight "0.3" of the tag "restaurant business" attached to the "tomato" as shown in the weight WT13-1. Further, for example, the tabulation device 100 acquires the weight "0.2" of the tag "agriculture" attached to the "tomato" as shown in the weight WT13-2. For example, in the example of FIG. 1, the aggregation device 100 specifies the weight of the tag “restaurant” and the tag “agriculture” attached to the token TK13 as the keyword of the search-linked advertisement, the industry of the advertiser. It may be calculated based on the ratio of. In this case, it indicates that the ratio of the industry "food and drink" to the industry "agriculture" in the advertiser who specified the character string "tomato" as the keyword of the search-linked advertisement is "3: 2".

The weight of the tag related to the type of business is not limited to the information related to the keyword of the search-linked advertisement, and may be determined by appropriately using various information. For example, the aggregation device 100 uses the weight of the tag "restaurant" and the weight of the tag "agriculture" attached to "tomato" in the search query Q11 as the market size (sales) in the tomato restaurant business and the market size in tomato agriculture (sales). It may be changed according to the amount of sales). For example, the tabulation device 100 is attached to "tomato" in the search query Q11 when the ratio of the market size (sales) in the tomato restaurant industry to the market size (sales) in the tomato agriculture is "4: 1". The weight of the tag "restaurant" may be "0.4" and the weight of the tag "agriculture" may be "0.1".

After that, the aggregation device 100 calculates the ratio of each tag in the category "place name" based on the tokens TK11 and 12 with the tags related to the category "place name" (step S12-1). In the example of FIG. 1, as shown in the ratio information RT11-1, the totaling device 100 calculates the ratio between the tag “Kanagawa” of the category “place name” and the tag “Miyagi” based on the information regarding the tokens TK11 and 12. To do. In the example of FIG. 1, the aggregation device 100 sets the ratio of the tag “Kanagawa” of the category “place name” to the tag “Miyagi” of the tags TG11-1, TG12-1, and TG12-2 of the tokens TK11 and TK12. It may be calculated as the ratio of the values to which the weights are added "1 + 0.5: 0.5". In this case, the totaling device 100 calculates the ratio between the tag "Kanagawa" of the category "place name" and the tag "Miyagi" as "1.5: 0.5", and the ratio in the category "place name" of the search query Q11. (Hereinafter, also referred to as "index value") calculates that "Kanagawa" is "3/4" and "Miyagi" is "1/4".

The calculation of the ratio is not limited to the above, and may be calculated by various methods depending on the purpose. For example, the aggregation device 100 has a ratio of the ratio of the tag “Kanagawa” of the category “place name” and the tag “Miyagi” multiplied by the weights of the tokens TK11 and TK12, “1 × 0.5: 0 × 0. It may be calculated as "5". In this case, the aggregation device 100 calculates the ratio between the tag "Kanagawa" of the category "place name" and the tag "Miyagi" as "1: 0", and sets the search query Q11 only for the tag "Kanagawa" of the category "place name". It may be a query that targets.

Further, the aggregation device 100 calculates the ratio of each tag in the category "industry" based on the token TK13 with the tag related to the category "industry" (step S12-2). In the example of FIG. 1, as shown in the ratio information RT12-1, the aggregation device 100 calculates the ratio between the tag “food and drink” of the category “industry” and the tag “agriculture” based on the information regarding the token TK13. In the example of FIG. 1, the aggregation device 100 sets the ratio of the tag “food and drink” of the category “industry” to the tag “agriculture”, and the weight ratio “0.3” of each tag TG13-1 and TG13-2 of each token TK13. : 0.2 "may be calculated. In this case, in the aggregation device 100, the ratio of the tag "restaurant" of the category "industry" to the tag "agriculture" is "3: 2", and the index value in the category "industry" of the search query Q11 is "restaurant". Is calculated as "3/5" and "agriculture" is calculated as "2/5".

After that, the aggregation device 100 calculates the score (number of searches) for each combination of the tags TG11-1 to TG13-2 attached to the tokens TK11 to TK13 based on the ratio information RT11-1 and RT12-1. The scores of the combinations of tags TG11-1 to TG13-2 are totaled (step S13).

In the example of FIG. 1, the totaling device 100 calculates and totals the scores for each of the four combinations as shown in the total list LT11. Here, the "combination ID" shown in the aggregation list LT11 indicates identification information for identifying the combination of tags that is the target of aggregation of the number of searches. Further, "category 1 (place name)" is included in the combination identified by the corresponding combination ID, and indicates a tag whose category is "place name". Further, "category 2 (industry)" indicates a tag included in the combination identified by the corresponding combination ID and whose category is "industry". For example, the combination identified by the combination ID "CN11" (hereinafter, may be referred to as "combination CN11") has a tag "Kanagawa" whose category is "place name" and a tag "food and beverage industry" whose category is "industry". Indicates a combination of.

Further, for example, the combination identified by the combination ID "CN12" (hereinafter, may be described as "combination CN12") has a tag "Kanagawa" whose category is "place name" and a tag "agriculture" whose category is "industry". It is shown that it is a combination with. Further, for example, the combination identified by the combination ID "CN13" (hereinafter, may be referred to as "combination CN13") has a tag "Miyagi" whose category is "place name" and a tag "restaurant business" whose category is "industry". It is shown that it is a combination with. Further, for example, the combination identified by the combination ID "CN14" (hereinafter, may be referred to as "combination CN14") has a tag "Miyagi" whose category is "place name" and a tag "agriculture" whose category is "industry". It is shown that it is a combination with.

That is, in the example of FIG. 1, the totaling device 100 calculates a score for each of the combinations CN11 to CN14. For example, the aggregation device 100 calculates a score for each of the combinations CN11 to CN14 as shown in the calculation information FM11. For example, the counting device 100 calculates the score of the combination CN11 by multiplying the index value of "Kanagawa" in the category "place name" and the index value of "restaurant business" in the category "industry". Specifically, the totaling device 100 sets the score of the combined CN11 to "0." by the formula "CN11 (score) = (1.5 / 2) x (0.3 / 0.5)" in the calculation information FM11. 45 ”is calculated.

Further, for example, the aggregation device 100 calculates the score of the combination CN12 by multiplying the index value of "Kanagawa" in the category "place name" and the index value of "agriculture" in the category "industry". Specifically, the totaling device 100 sets the score of the combined CN12 to "0." by the formula "CN12 (score) = (1.5 / 2) x (0.2 / 0.5)" in the calculation information FM11. 3 "is calculated.

Further, for example, the aggregation device 100 calculates the score of the combination CN13 by multiplying the index value of "Miyagi" in the category "place name" and the index value of "restaurant business" in the category "industry". Specifically, the totaling device 100 sets the score of the combined CN13 to "0." by the formula "CN13 (score) = (0.5 / 2) x (0.3 / 0.5)" in the calculation information FM11. It is calculated as "15".

Further, for example, the aggregation device 100 calculates the score of the combination CN14 by multiplying the index value of "Miyagi" in the category "place name" and the index value of "agriculture" in the category "industry". Specifically, the totaling device 100 sets the score of the combined CN12 to "0." by the formula "CN14 (score) = (0.5 / 2) x (0.2 / 0.5)" in the calculation information FM11. 1 "is calculated.

The total score of each combination CN11 to CN14 described above is "0.45 + 0.3 + 0.15 + 0.1 = 1". Therefore, the aggregation device 100 can use the scores of each combination CN11 to CN14 as the number of times when one search of the search query Q11 is distributed to each combination of tags attached to each token.

In the example of FIG. 1, as shown in the aggregation list LT11, the aggregation device 100 aggregates the scores of the combinations CN11 to CN14 as the number of searches for one search of the search query Q11. For example, as shown in the aggregation list LT11, the aggregation device 100 aggregates the number of searches of the combination CN11 as “0.45” times for one search of the search query Q11. Further, for example, the aggregation device 100 aggregates the number of searches of the combination CN12 as "0.3" times for one search of the search query Q11. Further, for example, the aggregation device 100 aggregates the number of searches of the combination CN13 as "0.15" times for one search of the search query Q11. Further, for example, the aggregation device 100 aggregates the number of searches of the combination CN14 as "0.1" times for one search of the search query Q11. After that, the aggregation device 100 transmits the aggregation information as shown in the aggregation list LT11 to the search device 50 (step S14).

As described above, by calculating the score for each combination of tags attached to each token, the aggregation device 100 is attached with a plurality of labels (tags) including labels related to place names and labels related to categories other than place names. Even if the specified token is included, the number of searches can be appropriately aggregated. In this way, the totaling device 100 can appropriately grasp the search amount for each region or industry by calculating the score for each combination of tags and totaling the score. In addition, the aggregation device 100 can grasp the business demand for each region or industry based on the aggregated information. Such information on business demand for each region or industry may be used for advertising sales and the like. In addition, the aggregation device 100 may provide various services for the number of searches for each combination that has been aggregated.

For example, the aggregation device 100 may provide the service provider in the region with information indicating what kind of industry the search is performed in combination with the name of each region (for example, prefecture). For example, the aggregation device 100 provides information indicating what kind of industry the search is performed in combination with, based on the score (number of searches) for each combination stored in the aggregation information storage unit 123 shown in FIG. It may be provided to businesses that provide services in Kanagawa and Miyagi prefectures. For example, the aggregation device 100 provides information indicating that there are many searches related to the restaurant business in Kanagawa Prefecture based on the comparison between the score of the combination CN11 and the score of the combination CN12 stored in the aggregation information storage unit 123 shown in FIG. , May be provided to businesses that provide services in Kanagawa Prefecture. Further, for example, the aggregation device 100 shows that there are many searches related to agriculture in Miyagi prefecture based on the comparison between the score of the combination CN13 and the score of the combination CN14 stored in the aggregation information storage unit 123 shown in FIG. Information may be provided to businesses that provide services in Miyagi Prefecture.

Further, in the above example, the case where the tag categories are two, "place name" and "industry" is shown, but various categories may be applied to the tag categories depending on the purpose. For example, the tag category may be a category related to user attributes. For example, the tag category may be "gender", "age", or the like. In this case, for example, the weight of each tag may be calculated based on the ratio. For example, if the token "car" is tagged with the category "gender" tag "male" and the tag "female", it is attached to the token "car" according to the percentage of the token "car" used for the search. The weight of the tag "man" and the weight of the tag "woman" may be calculated. For example, when the ratio of using the token "car" for search is "90%" for men and "10%" for women, the weight of the tag "man" attached to the token "car" is "0. It is calculated as "9", and the weight of the tag "woman" may be calculated as "0.1". For example, when the tag "male" and the tag "female" of the category "gender" are attached to the token "car", the token is according to the gender ratio of users who are interested in the token "car". The weight of the tag "male" attached to the "car" and the weight of the tag "female" may be calculated. By aggregating the scores of the tag combinations related to the user attributes in this way, the aggregating device 100 can aggregate the scores for each user corresponding to the detailed user attributes. The above is an example, and the score for each combination of tags aggregated by the aggregation device 100 may be used for any service as long as the score is available.

[2. Aggregator configuration]
Next, the configuration of the tabulation device 100 according to the embodiment will be described with reference to FIG. FIG. 2 is a diagram showing a configuration example of the aggregation device according to the embodiment. As shown in FIG. 2, the aggregation device 100 includes a communication unit 110, a storage unit 120, and a control unit 130. The totaling device 100 may have a display unit for displaying various types of information and an input unit for inputting various types of information.

(Communication unit 110)
The communication unit 110 is realized by, for example, a NIC (Network Interface Card) or the like. Then, the communication unit 110 is connected to a predetermined network by wire or wirelessly, and transmits / receives information to / from an external information processing device.

(Memory unit 120)
The storage unit 120 is realized by, for example, a semiconductor memory element such as a RAM (Random Access Memory) or a flash memory (Flash Memory), or a storage device such as a hard disk or an optical disk. As shown in FIG. 2, the storage unit 120 according to the embodiment includes a search history information storage unit 121, a token information storage unit 122, and an aggregate information storage unit 123.

(Search history information storage unit 121)
The search history information storage unit 121 according to the embodiment stores various information related to the search. For example, in FIG. 3, the search history information storage unit 121 stores information about a query used in a predetermined search system. FIG. 3 is a diagram showing an example of the search history information storage unit according to the embodiment. The search history information storage unit 121 shown in FIG. 3 has items such as "search query ID", "search query", "input user", and "date and time". The "search query" includes items such as "token 1", "token 2", and "token 3".

The "search query ID" indicates identification information for identifying the query. "Token 1", "token 2", and "token 3" of the "search query" indicate tokens included in the corresponding query. "Input user" indicates a user who has input (used) the corresponding query. Further, "date and time" indicates the date and time when the corresponding query was input (used).

In the example of FIG. 3, it is shown that the query identified by the search query ID "Q11" includes tokens such as "Yokohama", "Aoba Ward", and "tomato". It also indicates that the user who used the query identified by the search query ID "Q11" is the user "U1". It also indicates that the search using the query identified by the search query ID "Q11" was performed on the "date and time DA". In the example of FIG. 3, the date and time are expressed by abstract symbols such as the date and time DA, but even if it is a specific date and time such as "July 27, 2016 21:28:54". Good.

Further, in the example of FIG. 3, it is shown that the query identified by the search query ID "Q12" includes tokens such as "Chuo-ku", "Honmachi", and "cucumber". It also indicates that the user who used the query identified by the search query ID "Q12" is the user "U2". It also indicates that the search using the query identified by the search query ID "Q12" was performed in the "date and time DB".

The search history information storage unit 121 is not limited to the above, and various information may be stored depending on the purpose. The token ID described later may be stored in the "token 1", the "token 2", and the "token 3".

(Token information storage unit 122)
The token information storage unit 122 according to the embodiment stores various information related to the feature cluster. For example, in FIG. 4, the token information storage unit 122 stores information about the feature cluster. FIG. 4 is a diagram showing an example of the token information storage unit according to the embodiment. The token information storage unit 122 shown in FIG. 4 has items such as "token ID", "token", "tag 1", and "tag 2". The "tag 1" and "tag 2" include items such as "tag", "category", and "weight".

The "token ID" indicates identification information for identifying the token. “Token” refers to a token identified by the corresponding token ID. The "tag" of the "tag 1" indicates a tag attached to the corresponding token. Further, the "category" of the "tag 1" indicates the category of the tag attached to the corresponding token. Further, the "weight" of the "tag 1" indicates the weight of the tag attached to the corresponding token. Further, the "tag" of the "tag 2" indicates a tag attached to the corresponding token. Further, the "category" of the "tag 2" indicates the category of the tag attached to the corresponding token. Further, the "weight" of the "tag 2" indicates the weight of the tag attached to the corresponding token.

In the example of FIG. 4, it is shown that one tag "Kanagawa" is attached to the token "Yokohama" identified by the token ID "TK11". In addition, the category of the tag "Kanagawa" attached to the token "Yokohama" is "place name", and its weight is "1".

Further, in the example of FIG. 4, it is shown that the token "Aoba Ward" identified by the token ID "TK12" is attached with two tags, the tag "Kanagawa" and the tag "Miyagi". In addition, the category of the tag "Kanagawa" attached to the token "Aoba Ward" is "place name", and its weight is "0.5". In addition, the category of the tag "Miyagi" attached to the token "Aoba Ward" is "place name", and its weight is "0.5".

The token information storage unit 122 is not limited to the above, and may store various information depending on the purpose. For example, the token information storage unit 122 may store information regarding the date and time when the token was acquired.

(Aggregate information storage unit 123)
The aggregated information storage unit 123 according to the embodiment stores various information related to the aggregated number of searches. FIG. 5 is a diagram showing an example of the aggregated information storage unit according to the embodiment. The aggregated information storage unit 123 shown in FIG. 5 has items such as "combination ID", "category 1 (place name)", "category 2 (industry)", and "score (number of searches)".

The "combination ID" indicates identification information for identifying a combination of tags to be counted for the number of searches. Further, "category 1 (place name)" is included in the combination identified by the corresponding combination ID, and indicates a tag whose category is "place name". Further, "category 2 (industry)" indicates a tag included in the combination identified by the corresponding combination ID and whose category is "industry". The “score (number of searches)” indicates the total score (number of searches) for the combination identified by the corresponding combination ID.

For example, in the example shown in FIG. 5, the combination identified by the combination ID "CN11" indicates that the combination is a combination of the tag "Kanagawa" whose category is "place name" and the tag "restaurant business" whose category is "industry". .. Further, for example, in the example shown in FIG. 5, the score (number of searches) of the combination identified by the combination ID "CN11" is "1000".

The aggregated information storage unit 123 is not limited to the above, and may store various information depending on the purpose.

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

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

(Acquisition unit 131)
For example, the acquisition unit 131 acquires various types of information. For example, the acquisition unit 131 acquires various information from the search history information storage unit 121, the token information storage unit 122, and the aggregated information storage unit 123. For example, the acquisition unit 131 acquires various information from an external information processing device. For example, the acquisition unit 131 acquires information about a search query that is the target of totaling the number of searches from an external information processing device. For example, the acquisition unit 131 acquires information about each token included in the search query and having a label related to the place name or a label related to a category other than the place name.

In addition, the acquisition unit 131 acquires information about each token included in the search query and labeled with a different category. For example, the acquisition unit 131 acquires information about one token labeled with a place name and information about a token with a label with a category other than the place name. For example, the acquisition unit 131 acquires information about tokens labeled with the industry as another category.

In the example of FIG. 1, the acquisition unit 131 acquires the token included in the search query Q11 by dividing the search query Q11 into tokens. For example, the acquisition unit 131 acquires each character string separated by a space (“”) from the search query Q11 as a token. For example, the acquisition unit 131 acquires the token TK11 which is the character string “Yokohama” from the search query Q11. Further, for example, the acquisition unit 131 acquires the token TK12 which is the character string "Aoba Ward" from the search query Q11. Further, for example, the acquisition unit 131 acquires the token TK13 which is the character string "tomato" from the search query Q11.

For example, the acquisition unit 131 acquires information about one token included in the search query entered by the related user related to the place name and labeled with the place name. Further, for example, the acquisition unit 131 acquires information about one token included in the search query input by a user other than the related user and having a label related to the place name. Further, for example, the acquisition unit 131 acquires information about one token included in the search query entered by the related user who resides in the area corresponding to the place name, and one token included in the search query entered by another user. Get information about.

(Calculation unit 132)
The calculation unit 132 calculates various information. The calculation unit 132 calculates the tag ratio for each category. In the example of FIG. 1, the calculation unit 132 calculates the ratio of each tag in the category “place name” based on the tokens TK11 and 12 with the tags related to the category “place name” (step S12-1). For example, as shown in the ratio information RT11-1, the calculation unit 132 calculates the ratio between the tag “Kanagawa” of the category “place name” and the tag “Miyagi” based on the information regarding the tokens TK11 and 12. For example, the calculation unit 132 adds the weights of the tags TG11-1, TG12-1, and TG12-2 of the tokens TK11 and TK12 to the ratio of the tag "Kanagawa" and the tag "Miyagi" of the category "place name". It may be calculated as a ratio of values "1 + 0.5: 0 + 0.5". In this case, the calculation unit 132 calculates the ratio between the tag "Kanagawa" of the category "place name" and the tag "Miyagi" as "1.5: 0.5", and the index in the category "place name" of the search query Q11. The values are calculated as "3/4" for "Kanagawa" and "1/4" for "Miyagi".

Further, for example, the calculation unit 132 calculates the ratio of each tag in the category "industry" based on the token TK13 with the tag related to the category "industry". In the example of FIG. 1, as shown in the ratio information RT12-1, the calculation unit 132 calculates the ratio between the tag “food and drink” of the category “industry” and the tag “agriculture” based on the information regarding the token TK13. For example, the calculation unit 132 sets the ratio of the tag "food and drink" of the category "industry" to the tag "agriculture" to the weight ratio of each tag TG13-1 and TG13-2 of each token TK13 "0.3: 0.2". May be calculated as. In this case, the calculation unit 132 has a ratio of the tag "restaurant" of the category "industry" to the tag "agriculture" of "3: 2", and the index value in the category "industry" of the search query Q11 is "restaurant". Is calculated as "3/5" and "agriculture" is calculated as "2/5".

For example, the calculation unit 132 calculates the score (number of searches) for each combination of the tags TG11-1 to TG13-2 attached to the tokens TK11 to TK13 based on the ratio information RT11-1 and RT12-1. In the example of FIG. 1, the calculation unit 132 calculates the score for each of the four combinations as shown in the summary list LT11.

In the example of FIG. 1, the calculation unit 132 calculates the score for each of the combinations CN11 to CN14. For example, the calculation unit 132 calculates a score for each of the combinations CN11 to CN14 as shown in the calculation information FM11. For example, the calculation unit 132 calculates the score of the combination CN11 by multiplying the index value of "Kanagawa" in the category "place name" and the index value of "restaurant business" in the category "industry". Specifically, the calculation unit 132 uses the formula “CN11 (score) = (1.5 / 2) × (0.3 / 0.5)” in the calculation information FM11 to set the score of the combination CN11 to “0. 45 ”is calculated.

Further, for example, the calculation unit 132 calculates the score of the combination CN12 by multiplying the index value of "Kanagawa" in the category "place name" and the index value of "agriculture" in the category "industry". Specifically, the calculation unit 132 uses the formula “CN12 (score) = (1.5 / 2) × (0.2 / 0.5)” in the calculation information FM11 to set the score of the combination CN12 to “0. 3 "is calculated.

Further, for example, the calculation unit 132 calculates the score of the combination CN13 by multiplying the index value of "Miyagi" in the category "place name" and the index value of "restaurant business" in the category "industry". Specifically, the calculation unit 132 uses the formula “CN13 (score) = (0.5 / 2) × (0.3 / 0.5)” in the calculation information FM11 to set the score of the combination CN13 to “0. It is calculated as "15".

Further, for example, the calculation unit 132 calculates the score of the combination CN14 by multiplying the index value of "Miyagi" in the category "place name" and the index value of "agriculture" in the category "industry". Specifically, the calculation unit 132 uses the formula “CN14 (score) = (0.5 / 2) × (0.2 / 0.5)” in the calculation information FM11 to set the score of the combination CN12 to “0. 1 "is calculated.

(Aggregation section 133)
The aggregation unit 133 aggregates various information. For example, the aggregation unit 133 aggregates information on the number of searches by the search query for each combination of labels attached to each token. For example, the aggregation unit 133 aggregates information on the number of searches by the search query for each label combination based on the weight of the label attached to each token. For example, the aggregation unit 133 aggregates information on the number of searches by the search query for each label combination for each of the related user and the other user.

In the example of FIG. 1, the aggregation unit 133 aggregates the scores of the combinations of the tags TG11-1 to TG13-2 based on the scores calculated by the calculation unit 132. In the example of FIG. 1, as shown in the aggregation list LT11, the aggregation unit 133 aggregates the scores of the combinations CN11 to CN14 as the number of searches for one search of the search query Q11. For example, as shown in the aggregation list LT11, the aggregation unit 133 aggregates the number of searches of the combination CN11 to "0.45" for one search of the search query Q11. Further, for example, the aggregation unit 133 aggregates the number of searches of the combination CN12 as "0.3" times for one search of the search query Q11. Further, for example, the aggregation unit 133 aggregates the number of searches of the combination CN13 to "0.15" for one search of the search query Q11. Further, for example, the aggregation unit 133 aggregates the number of searches of the combination CN14 to "0.1" for one search of the search query Q11.

(Transmission unit 134)
The transmission unit 134 transmits various information to an external information processing device. For example, the transmission unit 134 transmits the number of searches aggregated by the aggregation unit 133 to an external information processing device.

[3. Aggregation processing flow]
Next, the procedure of the aggregation process by the aggregation device 100 according to the embodiment will be described with reference to FIG. FIG. 6 is a flowchart showing the aggregation processing procedure according to the embodiment.

As shown in FIG. 6, the aggregation device 100 acquires a search query (step S101). For example, the counting device 100 acquires the search query Q11 (see FIG. 1) which is the character string "Yokohama Aoba-ku tomato".

Then, the aggregation device 100 divides the search query into tokens (step S102). In the example of FIG. 1, the aggregation device 100 acquires the token included in the search query Q11 by dividing the search query Q11 into tokens.

In addition, the aggregation device 100 calculates the tag ratio for each category (step S103). In the example of FIG. 1, the aggregation device 100 calculates the ratio of each tag in the category “place name” based on the tokens TK11 and TK12 with the tags related to the category “place name”. Further, in the example of FIG. 1, the aggregation device 100 calculates the ratio of each tag in the category "industry" based on the token TK13 with the tag related to the category "industry".

After that, the totaling device 100 totals the number of searches based on the tag ratio (step S104). In the example of FIG. 1, as shown in the aggregation list LT11, the aggregation device 100 aggregates the scores of the combinations CN11 to CN14 as the number of searches for one search of the search query Q11. For example, as shown in the aggregation list LT11, the aggregation device 100 aggregates the number of searches of the combination CN11 as “0.45” times for one search of the search query Q11. Further, for example, the aggregation device 100 aggregates the number of searches of the combination CN12 as "0.3" times for one search of the search query Q11. Further, for example, the aggregation device 100 aggregates the number of searches of the combination CN13 as "0.15" times for one search of the search query Q11. Further, for example, the aggregation device 100 aggregates the number of searches of the combination CN14 as "0.1" times for one search of the search query Q11.

[4. Modification example]
In the above example, the case where the search count of the search query is totaled for each combination of tags regardless of the user who input the search query is shown, but the totaling device 100A counts the number of searches according to the classification of the users. It may be totaled for each combination of tags. This point will be described with reference to FIGS. 7 to 9.

[5. Aggregator configuration]
Next, the configuration of the totaling device 100A according to the modified example will be described with reference to FIG. 7. FIG. 7 is a diagram showing a configuration example of the aggregation device according to the modified example. As shown in FIG. 7, the totaling device 100A includes a communication unit 110, a storage unit 120A, and a control unit 130A. In the totaling device 100A, the same points as those of the totaling device 100 according to the embodiment are designated by the same reference numerals, and the description thereof will be omitted as appropriate.

(Storage unit 120A)
The storage unit 120A is realized by, for example, a semiconductor memory element such as a RAM or a flash memory, or a storage device such as a hard disk or an optical disk. As shown in FIG. 7, the storage unit 120A according to the modified example includes a search history information storage unit 121, a token information storage unit 122, an aggregate information storage unit 123A, and a user information storage unit 124.

(Aggregate information storage unit 123A)
The aggregated information storage unit 123A according to the modified example stores various information related to the aggregated number of searches. FIG. 8 is a diagram showing an example of the aggregated information storage unit according to the modified example. The aggregated information storage unit 123A shown in FIG. 8 has items such as "combination ID", "user classification", "category 1 (place name)", "category 2 (industry)", and "score (number of searches)". As described above, in the aggregated information storage unit 123A shown in FIG. 8, the item “user classification” is different from the aggregated information storage unit 123 according to the embodiment.

The "user category" indicates a user category in which the number of searches is totaled for the combination identified by the combination ID.

For example, in the example shown in FIG. 8, the combination identified by the combination ID "CN111" indicates the total number of searches for the search query entered by the user classified in the user category "local". Further, the combination identified by the combination ID "CN111" indicates that the combination is a combination of the tag "Kanagawa" whose category is "place name" and the tag "restaurant business" whose category is "industry". Further, for example, in the example shown in FIG. 8, the score (number of searches) of the combination identified by the combination ID "CN111" is "800".

For example, in the example shown in FIG. 8, the combination identified by the combination ID "CN112" indicates the total number of searches for the search query entered by the user classified in the user category "far". Further, the combination identified by the combination ID "CN112" indicates that the combination is a combination of the tag "Kanagawa" whose category is "place name" and the tag "restaurant business" whose category is "industry". Further, for example, in the example shown in FIG. 8, the score (number of searches) of the combination identified by the combination ID "CN112" is "200".

The aggregated information storage unit 123A is not limited to the above, and may store various information depending on the purpose.

(User information storage unit 124)
The user information storage unit 124 according to the modified example stores various information related to user attributes. For example, the user information storage unit 124 stores user attribute information. FIG. 9 is a diagram showing an example of a user information storage unit according to a modified example. The user information storage unit 124 shown in FIG. 9 includes items such as "user ID", "age", "gender", "residential place", and "interest".

The "user ID" indicates identification information for identifying a user. Further, "age" indicates the age of the user identified by the user ID. The "age" may be the specific age of the user identified by the user ID, such as 35 years old. Further, "gender" indicates the gender of the user identified by the user ID. Further, the "residential place" indicates the area where the user resides, which is identified by the user ID. The "residential place" may be a specific address of the user identified by the user ID. Further, "interest" indicates an object of interest to the user identified by the user ID. A plurality of "interests" may be registered.

For example, in the example shown in FIG. 9, the age of the user identified by the user ID "U1" is "30's", and the gender is "male". Further, for example, in the example shown in FIG. 9, the user identified by the user ID "U1" indicates that the place of residence is "Kanagawa" and the object of interest is "sports".

The user information storage unit 124 is not limited to the above, and may store various information depending on the purpose. For example, the user information storage unit 124 may store information regarding the demographic attribute and information regarding the psychographic attribute of the user. For example, the user information storage unit 124 may store information such as a name, family structure, income, and lifestyle.

(Control unit 130A)
Returning to the description of FIG. 7, the control unit 130A is a controller, and various programs (corresponding to an example of an extraction program) stored in a storage device inside the totaling device 100A by, for example, a CPU or an MPU. Is realized by executing RAM as a work area. Further, the control unit 130A is a controller, and is realized by, for example, an integrated circuit such as an ASIC or FPGA.

As shown in FIG. 7, the control unit 130A includes an acquisition unit 131, a calculation unit 132, an aggregation unit 133A, and a transmission unit 134, and realizes or executes an information processing function or operation described below. ..

For example, the acquisition unit 131 according to the modified example acquires various information. For example, the acquisition unit 131 acquires information about a search query and a user who has input the search query. For example, the acquisition unit 131 acquires a search query and user attribute information from a terminal device used by the user. For example, the acquisition unit 131 acquires information about the user's place of residence. For example, the acquisition unit 131 acquires information about one token included in the search query entered by the related user related to the place name and labeled with the place name. Further, for example, the acquisition unit 131 acquires information about other tokens included in the search query entered by a user other than the related user and having a label related to the place name. Further, for example, the acquisition unit 131 acquires information about one token included in the search query entered by the related user who resides in the area corresponding to the place name, and other tokens included in the search query entered by the other user. Get information about.

(Aggregation unit 133A)
The aggregation unit 133A aggregates various information. For example, the aggregation unit 133A aggregates various information in the same manner as the aggregation unit 133. For example, the aggregation unit 133A aggregates information on the number of searches by the search query for each label combination for each of the related user and the other user. For example, the aggregation unit 133A relates to a user whose attribute includes a place of residence corresponding to a tag related to the category "place name" as a related user, and other users as other users, regarding the number of searches by a search query for each label combination. Aggregate information.

For example, when the user who input the search query Q11 is the user U1 including the attribute of the place of residence "Kanagawa", the aggregation unit 133A is input by the related user, that is, the user classified into the user category "local" shown in FIG. As a search query, the number of searches is totaled. Further, for example, when the user who has input the search query Q11 is the user U2 including the attribute of the place of residence "Aichi", the aggregation unit 133A is a related user, that is, a user classified into the user category "far" shown in FIG. Aggregate the number of searches as a search query entered by. In this way, by classifying the users according to the address (residential place) of the user searched by the aggregation unit 133 and performing aggregation, the aggregation device 100A sets the business demand to the demand from within the region and the region. It is possible to grasp the demand separately from the outside demand.

[6. effect〕
As described above, the totaling device 100 according to the embodiment and the totaling device 100A according to the modified example have an acquisition unit 131 and a totaling unit 133 and 133A. The acquisition unit 131 acquires information about each token included in the search query and having a label related to the place name or a label related to a category other than the place name. The aggregation unit 133, 133A aggregates the information regarding the number of searches by the search query for each combination of labels attached to each token.

As a result, even when the totaling device 100 according to the embodiment and the totaling device 100A according to the modified example include tokens with a plurality of labels such as a label related to a place name and a label related to a category other than the place name. , The number of searches can be aggregated appropriately.

Further, in the aggregation device 100 according to the embodiment and the aggregation device 100A according to the modified example, the aggregation unit 133, 133A relates to the number of searches by the search query for each label combination based on the weight of the label attached to each token. Aggregate information.

As a result, the aggregation device 100 according to the embodiment and the aggregation device 100A according to the modified example aggregate information on the number of searches by the search query for each label combination based on the weight of the label attached to each token. , Even when tokens with multiple labels are included, the number of searches can be appropriately aggregated.

Further, in the aggregation device 100 according to the embodiment and the aggregation device 100A according to the modified example, the acquisition unit 131 acquires information on tokens labeled with the industry as another category.

As a result, the aggregation device 100 according to the embodiment and the aggregation device 100A according to the modified example are labeled with a plurality of labels including the industry by acquiring information on tokens labeled with the industry as other categories. Even if tokens are included, the number of searches can be calculated appropriately.

Further, in the aggregation device 100A according to the modified example, the acquisition unit 131 is one token included in the search query input by the related user related to the place name, and information about one token with a label related to the place name. Acquires information about other tokens that are acquired and included in the search query entered by a user other than the related user and that are labeled with the place name. The aggregation unit 133A aggregates information on the number of searches by the search query for each label combination for each of the related user and the other user.

As a result, the totaling device 100A according to the modified example has a plurality of labels attached to each of the related user and the other user by totaling the information regarding the number of searches by the search query for each combination of labels. Even when tokens are included, the number of searches can be appropriately aggregated according to the classification of users.

Further, in the aggregation device 100A according to the modified example, the acquisition unit 131 acquires information about one token included in the search query input by the related user who resides in the area corresponding to the place name, and the search input by the other user. Get information about other tokens contained in the query.

As a result, the aggregation device 100A according to the modified example acquires information on one token included in the search query entered by the related user residing in the area corresponding to the place name, and is included in the search query entered by the other user. By acquiring information about other tokens, the number of searches can be appropriately aggregated according to the classification of users based on the area of residence, even if tokens with multiple labels are included. ..

[7. Hardware configuration]
The tabulation device 100 according to the above-described embodiment and the tabulation device 100A according to a modified example are realized by, for example, a computer 1000 having a configuration as shown in FIG. FIG. 10 is a hardware configuration diagram showing an example of a computer that realizes the function of the tabulation device. The computer 1000 has a CPU 1100, a RAM 1200, a ROM 1300, an HDD (Hard Disk Drive) 1400, a communication interface (I / F) 1500, an input / output interface (I / F) 1600, and a media interface (I / F) 1700.

The CPU 1100 operates based on a program stored in the ROM 1300 or the HDD 1400, and controls each part. The ROM 1300 stores a boot program executed by the CPU 1100 when the computer 1000 is started, a program depending on the hardware of the computer 1000, and the like.

The HDD 1400 stores a program executed by the CPU 1100, data used by such a program, and the like. The communication interface 1500 receives data from another device via a predetermined network N and sends the data to the CPU 1100, and transmits the data generated by the CPU 1100 to the other device via the predetermined network N.

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

The media interface 1700 reads a program or data stored in the recording medium 1800 and provides the program or data to the CPU 1100 via the RAM 1200. The CPU 1100 loads the program from the recording medium 1800 onto the RAM 1200 via the media interface 1700, and executes the loaded program. The recording medium 1800 is, for example, an optical recording medium such as a DVD (Digital Versatile Disc) or PD (Phase change rewritable Disk), a magneto-optical recording medium such as an MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory. And so on.

For example, when the computer 1000 functions as the totaling device 100 according to the embodiment and the totaling device 100A according to the modified example, the CPU 1100 of the computer 1000 executes the program loaded on the RAM 1200 to control the control units 130 and 130A. Realize the function. The CPU 1100 of the computer 1000 reads and executes these programs from the recording medium 1800, but as another example, these programs may be acquired from another device via a predetermined network N.

Although some of the embodiments of the present application have been described in detail with reference to the drawings, these are examples, and various modifications are made based on the knowledge of those skilled in the art, including the embodiments described in the disclosure line of the invention. It is possible to practice the present invention in other improved forms.

[8. Others]
Further, among the processes described in the above-described embodiment, all or a part of the processes described as being automatically performed can be manually performed, or the processes described as being manually performed can be performed. All or part of it can be done automatically by a known method. In addition, the processing procedure, specific name, and information including various data and parameters shown in the above document and drawings can be arbitrarily changed unless otherwise specified. For example, the various information shown in each figure is not limited to the illustrated information. Further, each component of each of the illustrated devices is a functional concept, and does not necessarily have to be physically configured as shown in the figure. That is, the specific form of distribution / integration of each device is not limited to the one shown in the figure, and all or part of them may be functionally or physically distributed / physically in arbitrary units according to various loads and usage conditions. Can be integrated and configured.

Further, the above-described embodiments can be appropriately combined as long as the processing contents do not contradict each other.

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

100 Aggregation device 121 Search history information storage unit 122 Token information storage unit 123 Aggregation information storage unit 130 Control unit 131 Acquisition unit 132 Calculation unit 133 Aggregation unit 134 Transmission unit

Claims (7)

Information about each token included in a search query, each with a label for a place name or a label for a category other than the place name, and each token including a multi-labeled token with multiple labels. The acquisition department to acquire and
A plurality of combinations of labels attached to each token, and among the plurality of labels attached to the plurality of label tokens, each of the plurality of combinations not including combinations between labels of the same category is targeted. A tabulation section that aggregates information about the number of searches by search queries, and
A counting device characterized by being provided with. The tabulation unit
The aggregation device according to claim 1, wherein information on the number of searches by the search query is aggregated for each of the plurality of combinations based on the weight of the label attached to each token. The acquisition unit
The aggregation device according to claim 1 or 2, wherein information about a token labeled with an industry as the other category is acquired. The acquisition unit
One token included in the search query entered by the related user related to the place name, the information about the one token labeled with the place name is acquired, and the information is entered by a user other than the related user. Obtain information about other tokens included in the search query and labeled with the place name.
The tabulation unit
The invention according to any one of claims 1 to 3, wherein information on the number of searches by the search query is aggregated for each of the related users and the other users for each of the plurality of combinations. Aggregator. The acquisition unit
Acquires information about the one token included in the search query entered by the related user who resides in the area corresponding to the place name, and acquires information about the other token included in the search query entered by the other user. The tabulation device according to claim 4, wherein the tabulation device is characterized in that. It ’s a computer-executed aggregation method.
Information about each token included in a search query, each with a label for a place name or a label for a category other than the place name, and each token including a multi-labeled token with multiple labels. Acquisition process to acquire and
A plurality of combinations of labels attached to each token, and among the plurality of labels attached to the plurality of label tokens, each of the plurality of combinations not including combinations between labels of the same category is targeted. Aggregation process that aggregates information about the number of searches by search query and
An aggregation method characterized by including. Information about each token included in a search query, each with a label for a place name or a label for a category other than the place name, and each token including a multi-labeled token with multiple labels. The acquisition procedure to acquire and
A plurality of combinations of labels attached to each token, and among the plurality of labels attached to the plurality of label tokens, each of the plurality of combinations not including combinations between labels of the same category is targeted. Aggregation procedure to aggregate information about the number of searches by search query and
A tabulation program characterized by having a computer execute.

Images