JP2013196315A - Information processing apparatus and method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable a user to obtain desired retrieval results by enlarging a range of keywords matched with the retrieval results.SOLUTION: Click logs of second data retrieval is retrieved with a keyword of which the number of hits of retrieval results or the number of selecting operations of the retrieval results is zero in first data retrieval. Feature words corresponding to retrieval targets of which the click rates are equal to or higher than a predetermined rate among the retrieval results are estimated as suitable other keywords. Thus, with respect to the keyword of which the number of hits or the number of selecting operations is zero, on the basis of the click logs of the data retrieval in different classes, the feature words of the retrieval targets which are selected many in the case where the keyword is retrieved in the second data retrieval in different target fields or storage data, can be estimated as the suitable other keywords. Therefore, when the keyword of which the number of hits or the number of selecting operations has been zero is designated in the first data retrieval, a range of keywords matched with retrieval results is enlarged so as to enables the user to obtain the desired retrieval results in the future.

Description

  The present invention relates to information retrieval.

  In recent years, various types of data search, such as Web search, product search at an electronic commerce site, product search at an Internet auction, and map search, have been used on the Internet and others.

  For example, the present applicant proposes a technique for associating a page with a high click rate in a search result with a query (search request specifying a keyword) (see, for example, Patent Document 1), This is useful for optimizing search results, such as positioning it higher.

JP 2008-181186 A

  In relation to a database that defines the correspondence between a keyword and a search result, the number of hits in the search result may be zero when the input keyword is not an official name but an abbreviation or common name. In addition, even when several hits are found as a search result, there is a case where the selection operation such as click is not led. In these cases, the range of keywords that match the search results is narrow, and as a result, the user cannot obtain the desired search results.

  In view of the above problems, an object of the present invention is to expand a range of keywords that match a search result so that a user can obtain a desired search result.

  Based on the above object, one aspect (1) of the present invention is an information processing apparatus that performs a first data search based on a keyword, and is based on the number of search results hits or a search result in the first data search. And zero search means for detecting the keyword for which the number of selection operations performed is zero, and a search using the zero keyword detected by the zero detection means in the click log of the second data search A log extracting unit that extracts a click log as a result, a target selecting unit that selects a search target having a click rate of a predetermined value or more from the extracted click log, and a feature word that acquires a feature word corresponding to the selected search target Word acquisition means, association means for associating the acquired feature word with the keyword that was zero and storing it in the correspondence storage means as an estimated keyword; Search means for performing the first data search based on a keyword to be searched in the first data search and the estimated keyword stored in the correspondence storage means in association with the keyword. It is characterized by having.

  Another aspect (6) of the present invention is the information processing method in the information processing apparatus in which the computer performs the first data search based on the keyword, in which the above aspect is captured in the category of the method. In the data search, the zero detection process for detecting the keyword in which the number of hits in the search result or the number of selection operations performed on the search result is zero, and the zero detection process in the click log of the second data search A log extraction process that extracts a click log of a search result using the keyword that was detected by zero, and a target selection process that selects a search target having a click rate of a predetermined value or more from the extracted click log In addition, a feature word acquisition process for acquiring a feature word corresponding to the search target, and the acquired feature word in association with the keyword that is zero Based on the association processing to be stored as a keyword in the correspondence storage means, the keyword to be searched in the first data search, and the estimated keyword stored in the correspondence storage means in association with this keyword A computer executes a search process for performing the first data search.

  According to another aspect (2) of the present invention, in any one of the above aspects, the second data search is a Web search using a Web page as a search target, and the feature word acquisition unit (“unit” is a method). In this case, the term “processing” or “step” is used, and the same applies hereinafter) in which the feature word corresponding to the web page selected as the search target by the target selection unit is associated with the web page and the feature word. It is obtained by referring to index data for Web search.

  According to another aspect (3) of the present invention, in any one of the above aspects, the feature word acquisition unit is a feature word corresponding to the search target selected by the target selection unit, and the keyword is zero It is characterized by acquiring feature words other than.

  According to another aspect (4) of the present invention, in any one of the above aspects, the search means corresponds to the keyword from the correspondence storage means when the number of hits in the search result for the input keyword is zero. The estimated keyword stored in addition is acquired.

  According to another aspect (5) of the present invention, in any one of the above aspects, the search means includes a search result for the input keyword that is greater than or equal to a predetermined value and the number of selection operations performed on the search result. When it is zero, the estimated keyword stored in association with the keyword is acquired from the correspondence storage means.

  According to the present invention, it is possible to expand a range of keywords that match a search result so that a user can obtain a desired search result.

The functional block diagram which shows a structure about embodiment of this invention. The figure which shows the example of data in embodiment of this invention. The figure which shows the example of data in embodiment of this invention. The figure which shows the example of data in embodiment of this invention. The flowchart which shows the process sequence in embodiment of this invention. The figure which shows the example of a screen display in embodiment of this invention. The figure which shows the example of a screen display in embodiment of this invention. The figure which shows the example of a screen display in embodiment of this invention.

  Next, a mode for carrying out the present invention (referred to as “embodiment”) will be illustrated along the drawings. In addition, the assumptions common to the contents already described in the background art and problems are omitted as appropriate.

[1. Constitution〕
The present embodiment relates to the information processing apparatus 1 (also referred to as “the present apparatus 1” or “this apparatus”) shown in the configuration diagram of FIG. 1, and the present apparatus 1 is based on keywords according to access from the terminal T. It is a server device that performs a first data search.

  The first data search is specialized in a specific genre, such as searching for items sold in Internet auctions, searching for products sold on e-commerce shopping sites, and searching for place names and facility names on map sites. Search. Hereinafter, a spot search for searching for a facility such as a store will be mainly taken as an example, and an example of data related to a product search will be shown as needed.

  The Web search server 2 is a server device that provides a Web search for a Web page as a second data search, and is also simply referred to as “server 2” or “server” below. The server 2 includes an index data storage unit 21 that stores in advance Web search index data (for example, FIG. 2) in which Web pages and feature words are associated with each other, and a Web search unit that performs Web search using the index data. 22.

  In addition, the server 2 includes a click log storage unit 23 that stores a click log of Web search. The click log may directly represent the rate of selection operations (also referred to as “click rate”) performed for each individual web page for a certain keyword for the entire past period or the most recent predetermined period (for example, FIG. 3) may represent the number of such selection operations (also referred to as “selection number”), or may be a history indicating a Web page on which a selection operation has been performed for each search result based on a keyword.

  The selection operation for the search target in the search result is that any one of the hit stores included in the search result is selected by a mouse click or a touch panel tap operation. The apparatus 1, the terminal T, and the server 2 are connected via a communication network N (for example, a mobile communication network such as the Internet, a mobile phone, a PHS, and a public wireless LAN).

  The apparatus 1 has at least an arithmetic control unit 6 such as a CPU, a storage unit 7 such as a main memory and an auxiliary storage device, and a communication unit 8 (for example, various wireless and wired communications) as a computer configuration. Adapters, other communication circuits, communication devices, and the like). The same applies to the server 2 (not shown).

  In the present apparatus 1, elements (20, 30 and others) such as each unit shown in FIG. 1 are realized by the arithmetic control unit 6 executing the computer program stored in the storage unit 7. The server 2 conforms to this. Of the elements to be realized, the mode of the information storage means is free and can be realized in any data format such as a file on the storage unit 7, or may be remote storage by network computing (cloud).

  The storage means may include not only a data storage area but also functions such as data input / output and management. In addition, the unit of the storage means shown in this application is for convenience of explanation, and the configuration can be divided or integrated as appropriate, and the processing data and processing results of each means are stored in addition to the explicit storage means. A storage means is used as appropriate.

  The arrows in the figure (for example, FIG. 1) supplementarily indicate the main direction of the flow of data, control, etc., and do not deny other flows or imply direction limitations. For example, before or after data acquisition in a certain direction, a data request or an acknowledgment (ACK) can occur in the reverse direction.

  In addition, each means other than the storage means is a processing means for realizing and executing the information processing functions and operations (for example, FIG. 1) described below, but these are functional units arranged for explanation. It doesn't matter if it matches the actual hardware elements or software modules.

[2. Action)
Of the processing procedure in the apparatus 1 configured as described above, a part particularly related to the present embodiment is shown in a flowchart of FIG.

[2-1. Search and storage of zero words)
First, FIG. 6 is a screen example that also serves as a result display example of spot search. In the present apparatus 1, the search means 74 receives a keyword for spot search from the terminal T as illustrated in the input area A1 ( Step S11) A spot search is performed (Step S12), and the search result is output to the terminal T (Step S20). In the spot search, the spot index data in which the information for each facility is associated with the keyword is stored in the spot index storage means 73 in advance and used.

  Then, among the keywords accepted by the search means 74, the zero detection means 20 determines the keyword (step S13: “YES”) in which the number of hits in the search result is zero in the first data search and the search result. A keyword (step S15: “YES”) in which the number of selections made is zero is detected.

  The keyword detected here is hereinafter also referred to as “zero word”. The search result A2 in the screen example of FIG. 6 indicates that the number of hits in the search result is zero. In this case, the input keyword “Koichi” is detected as a zero word.

  While the search means 74 outputs the search result to the terminal T, it records a history (not shown) in which at least the keyword is associated with the selection number (step S20), and the zero detection means 20 indicates the number of keyword selections in the history. Judge from. A keyword selected as one of the items in the search result even once in the past or once in the past predetermined period does not become a zero word (step S15: “NO”).

  Here, the zero word is detected as both a keyword for which the number of hits is zero (step S13: “YES”) and a keyword for which the number of selections for the search result is zero (step S15: “YES”). Any one of them may be used. The zero detection means 20 stores the detected zero word in the zero word storage means 25 (steps S14 and S16).

  The estimated keyword is acquired from the stored zero word as follows (steps S21 to S25). Steps S17 to S19 will be described later because they are processing related to the use of the acquired estimated keyword.

[2-2. (Get estimated keywords)
In order to acquire the estimated keyword, first, the log extracting unit 30 at a predetermined timing (step S21: “YES”), the zero detecting unit 20 in the click log of the second data search (here, Web search). The click log of the search result using the zero word detected by is extracted (step S22).

  The timing (step S21) for performing the processing from the click log extraction (steps S22 to S25) is determined by appropriate conditions such as real-time processing, background processing, and batch processing, in addition to the fixed period. The click log is read from the click log storage means 23 and necessary portions are extracted.

  Subsequently, the target selection unit 40 selects a search target (here, a Web page; the same applies hereinafter) having a click rate of a predetermined value or more from the extracted click log (step S23). If the click log does not directly represent the click rate, but represents the number of selections or Web pages that have been selected for each search result, the click logs are aggregated and the click rate is calculated and used for Web page selection. .

  The threshold for determining that the click rate is equal to or greater than a predetermined value is arbitrary. For example, when the click rate is 10% or 30%, the click log (FIG. 3) of the zero word “Koichi” has a high click rate (URL is “ http: //ichi...co.jp ") is selected, and a Web page with a low click rate (URL is" http: //fashion....jp ") is not selected. A display example of the Web search result corresponding to the click log of FIG. 3 is shown in FIG.

  Further, the feature word acquisition unit 52 uses the index data for Web search (FIG. 2) stored in the index data storage unit 21 as the feature word corresponding to the Web page selected as the search target by the target selection unit 40. It acquires by referring (step S24). The feature word is, for example, a word included in the Web page title or a word extracted from the Web page linked by tf-idf.

  At this time, the feature word acquisition unit 52 acquires a feature word corresponding to the Web page selected by the target selection unit 40 and other than zero words (step S24). For example (FIG. 2), “Curry House Co ○ Ichibanya” and “Koichi” are associated with the selected Web page (URL is “http: // ichi... Co.jp”). However, since “Koichi” is equal to zero word, “Curry House Coxxx Ichibanya” is acquired.

  Then, the associating means 60 associates it with the zero word and stores it in the correspondence storing means 65 as an estimated keyword (estimated key) (step S25). For example, in the example of FIG. 4, “Curry House Co ○ Ichibanya” is stored as an estimated keyword in association with the zero word “Koichi”.

[2-3. Use of estimated keywords)
Then, the search means 74 performs a first data search based on the keyword to be searched in the first data search and the estimated keyword stored in the correspondence storage means 65 in association with the keyword ( Steps S12, S18, S19). However, since the estimated keyword is used when stored in association with the zero word, for example, before the estimated keyword is stored, it is displayed as a search result as shown in FIG. The estimated keyword is not always used but is used when necessary.

  Specifically, the search means 74, when the number of hits in the search result obtained by performing the spot search (step S12) based on the input keyword (step S11) is zero (step S13: “YES”), The estimated keyword stored in association with the keyword is acquired from the correspondence storage means 65 (step S18).

  Further, the search means 74 has zero selection for the search result with the input keyword (step S15: “YES”), and the search result with the input keyword is greater than or equal to a predetermined value (for example, predetermined within the most recent predetermined period). If it exists (step S17: “YES”), the estimated keyword stored in association with the keyword is acquired from the correspondence storage means 65 (step S18).

  For example, when the number of hits in the search result of the spot search with the keyword “Koichi” is zero as shown in FIG. 6, it is associated with the word “koichi” as described above. Then, the Web page of “Curry House CoOO Ichibanya” (URL is “http: // ichi... Co.jp”) is stored in association with the estimated keyword (FIG. 4).

  As a result, at the time of the re-search, even if the number of hits is zero in the spot search based on “Koichi”, the URL is “http” based on the estimated keyword “Curry House Coxxx Ichibanya”. </ ichi ... co.jp "is displayed (FIG. 8). The time of re-search includes both the case where a search is performed with the same keyword within a predetermined time, such as immediately after obtaining a search result with zero hits, and the case where the search is performed with the same keyword after the predetermined time has elapsed, such as the next day or one week later. .

  Similarly, in the product search, for example, if the number of hits or selection is zero and the estimated keyword “Final OO Taji” is associated with the input keyword “FF” (FIG. 4), “FF” will be used for the next and subsequent times. The search results based on the estimated keyword “Final OO Taji” are displayed (not shown).

[3. effect〕
(1) As described above, in this embodiment (for example, FIG. 5), the number of search result hits (step S13: “YES”) and the number of search result selection operations in the first data search (for example, spot search). A click log of the second data search (for example, Web search) is searched with the keyword that is zero (step S15) (step S22). Among the search results, a feature word corresponding to a search target (step S23) having a click rate of a predetermined value or more is estimated as another appropriate keyword (steps S24 and S25).

  As a result, for the keyword that was zero, based on the click log of the data search with different types, the keyword was searched in the second data search with different target fields or accumulated data, etc. Feature words can be estimated as other appropriate keywords. For this reason, after that, when the keyword that was zero in the first data search is specified, the range of keywords that match the search result is expanded so that the user can obtain the desired search result. it can.

(2) In particular, in the present embodiment, the first data search is a specialized search specialized for a specific genre (for example, spot search in steps S12 and S19), and various keywords such as abbreviations and slang terms are necessarily set. Even if the search is not performed, a Web search using a wide Web page as a search target is used as the second data search, and a Web page having a click rate of a predetermined value or more is selected as the search target.

  Thereby, based on examples of many people including abbreviations, slang terms, and the like, the range of keywords that match the search results can be expanded so that the user can obtain the desired search results. In addition, in order to acquire a feature word corresponding to the selected Web page, an existing information resource called Web search index data (for example, FIG. 2) in which the Web page and the feature word are associated with each other is obtained. The burden of re-applying language processing technology such as -idf can be avoided.

(3) Further, in the click log based on the search result of the data search using the keyword that was zero, it is considered that there are many search targets that include the keyword in the feature word (for example, FIG. 2). On the other hand, in this embodiment, by excluding the keyword that was zero from the feature words acquired as corresponding to the selected search target (step S24), the same keyword becomes an estimated keyword and the search result is You can avoid situations that do not improve.

(4) If the number of hits in the search result for the keyword specified by the user is even one, the hit result is likely to reflect the user's intention, and the estimated keyword is acquired in all cases. If the search is performed again, there is a possibility that noise (that is, a search result deviating from the user's intention) is included in the search result. On the other hand, in this embodiment, when the number of hits is zero (step S13: “YES”), the estimated keyword is acquired and used (step S14), so that the accuracy of the search result can be maintained.

(5) Also, if there is even one selection operation for the search result with the keyword specified by the user, it is highly likely that the hit result reflects the user's intention, and the estimated keyword is used in all cases. When it is acquired and re-searched, there is a possibility that noise (that is, a search result deviating from the user's intention) is included in the search result. The same applies to the case where an estimated keyword is used when there is no selection operation by chance in a small number of search results.

  On the other hand, in the present embodiment, when there are more than a predetermined number of search results for the input keyword (step S17: “YES”) and the number of selection operations is zero (step S15: “YES”), estimation is performed. Since keywords are acquired and used, the accuracy of search results can be maintained.

[4. Other embodiments]
In addition, the said embodiment is only an illustration and this invention includes what is illustrated below and other embodiment other than that. For example, in the above embodiment, the first data search is a spot search, and the second data search is a Web search, but the type of each data search can be arbitrarily selected. For example, the reverse combination may be used, and the first data search may be a product search or an image search.

  Moreover, each aspect can be grasped | ascertained as another category (a method, a program, a system containing a terminal, etc.) which is not specified. Regarding the category of method and program, “means” shown in the category of apparatus is appropriately read as “process” or “step”. Further, the order of processes and steps is not limited to the one directly specified in the present application, and can be changed such as changing the order or executing some processes collectively or partly at any time.

  In addition, computers such as terminals that implement and execute individual means, processes, and steps may be common, or may differ for each means, process, step, or timing. Further, all or any part of the “means” can be read as “part” (unit, section, module, etc.).

  In addition, the configuration diagram, data diagram, flowchart, and the like in the present application are merely examples, and the presence / absence of each element, the order of arrangement and processing execution, specific contents, and the like can be changed as appropriate. For example, the apparatus 1 and the server 2 can be configured integrally.

  In addition, the mode for realizing the individual means constituting the apparatus 1 is free, and functions provided by an external server are called by API (application program interface) or network computing (so-called cloud). For example, the configuration of the present invention can be flexibly changed. Furthermore, each element such as means relating to the present invention may be realized by other information processing mechanisms such as a physical electronic circuit as well as a computer control unit.

DESCRIPTION OF SYMBOLS 1 Information processing apparatus 2 Web search server 6 Operation control part 7 Storage part 8 Communication part 20 Zero detection means 21 Index data storage means 22 Web search means 23 Click log storage means 25 Zero word storage means 30 Log extraction means 40 Target selection means 52 Features Word acquisition means 60 Association means 65 Correspondence storage means 73 Spot etc. index storage means 74 Search means T Terminal

Claims (6)

An information processing apparatus for performing a first data search based on a keyword,
Zero detection means for detecting the keyword in which the number of hits in the search result or the number of selection operations performed on the search result in the first data search is zero;
Log extraction means for extracting a click log of a search result using the keyword that was zero detected by the zero detection means among the click logs of the second data search;
Target selection means for selecting a search target having a click rate of a predetermined value or higher from the extracted click log;
Feature word acquisition means for acquiring a feature word corresponding to the selected search target;
Association means for associating the acquired feature word with the keyword that was zero and storing it in the correspondence storage means as an estimated keyword;
Search means for performing the first data search based on a keyword to be searched in the first data search and the estimated keyword stored in the correspondence storage means in association with the keyword;
An information processing apparatus comprising: The second data search is a Web search that searches a Web page.
The feature word acquisition unit refers to the feature word corresponding to the Web page selected as the search target by the target selection unit with reference to index data for Web search in which a Web page and a feature word are associated with each other. The information processing apparatus according to claim 1, wherein the information processing apparatus is acquired.   3. The feature word acquisition unit acquires a feature word corresponding to the search target selected by the target selection unit and other than the zero keyword. Information processing device.   The search means acquires the estimated keyword stored in association with the keyword from the correspondence storage means when the number of hits in the search result for the input keyword is zero. Item 4. The information processing device according to any one of Items 1 to 3.   The search means stores the search result for the input keyword in association with the keyword from the correspondence storage means when there is a predetermined search result or more and the number of selection operations performed on the search result is zero. 5. The information processing apparatus according to claim 1, wherein the estimated keyword is acquired. An information processing method in an information processing apparatus in which a computer performs first data search based on a keyword,
A zero detection process for detecting the keyword in which the number of hits in the search result or the number of selection operations performed on the search result in the first data search is zero;
A log extraction process for extracting a click log of a search result using the keyword that was zero detected by the zero detection process among the click logs of the second data search;
A target selection process for selecting a search target having a click rate equal to or higher than a predetermined value from the extracted click log;
A feature word acquisition process for acquiring a feature word corresponding to the selected search target;
An association process in which the acquired feature word is associated with the keyword that was zero and stored in the correspondence storage unit as an estimated keyword;
A search process for performing the first data search based on a keyword to be searched in the first data search and the estimated keyword stored in the correspondence storage means in association with the keyword;
An information processing method characterized in that a computer executes.

Images