JP2010049409A - Advertisement bidding system, advertisement bidding method, document retrieval system, document retrieval method, program, and control method for client system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an advertisement bidding system which allows bidding for effective advertisement. <P>SOLUTION: In the advertisement bidding system, a retrieval server 30 includes: a question database storing a plurality of pieces of question information including an additional keyword for narrowing down a retrieval result matching a keyword; and a bid database storing advertisements in association, to the question information. The retrieval server 30 receives a retrieval request, including the keyword from a user terminal 20 and retrieves a document matching the keyword. The retrieval server reads the question information corresponding to the keyword from the question database and transmits screen data for displaying the question information to the user terminal 20, together with a retrieval result of the document. The retrieval server receives designation of the question information, and narrows down the retrieval result of the document by using the additional keyword contained in the designated question information. The retrieval server acquires the advertisement, corresponding to the designated question information from the bid database and an advertisement database, and transmits screen data for displaying the advertisement to the user terminal 20, together with a narrowed-down retrieval result. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an advertisement bidding system, an advertisement bidding method, a document retrieval system, a document retrieval method, a program, and a client system control method.

In a system that searches a large amount of information, a large amount of information often matches a keyword specified by the user, and there is a need to narrow down a large amount of search results. On the other hand, for example, Patent Document 1 discloses a system for narrowing down by removing a result seen by a user from a search result.
JP 2006-285656 A

By the way, when displaying the search result of information, an advertisement is also displayed. In such a search-linked advertisement, an advertisement corresponding to a keyword specified by the user is often displayed.
However, when a keyword has a plurality of meanings, information unintended by the user is included in the search result, and accordingly, the advertisement may not be linked to the intention of the user. .

  The present invention has been made in view of such a background, and an advertisement bidding system, an advertisement bidding method, a document retrieval system, a document retrieval method, a program, and a client system capable of bidding effective advertisements. An object is to provide a control method.

  A main invention of the present invention for solving the above-described problem is a system for bidding an advertisement to be output together with a search result of a document corresponding to a keyword that has been narrowed based on a narrowing condition, Correspondingly, the condition-specific advertisement storage unit that stores the advertisement and the unit price of the advertisement, the bid condition receiving unit that receives input of the narrowing condition, the advertisement, and the unit price, and the received narrowing condition A bid registration unit for registering the received advertisement and the received unit price in the conditional advertisement storage unit.

  In the advertisement bidding system of the present invention, the narrowing-down condition is a keyword to be added to the keyword that constitutes a tree structure with the keyword as a root, and the search result is narrowed down based on the narrowing-down condition Is performed by searching the document corresponding to the narrowing condition, the keyword that is the root of the tree structure, and another narrowing condition that is an ancestor of the narrowing condition in the tree structure, The advertisement bidding system includes a narrowing condition storage unit that stores the narrowing condition of the tree structure having the keyword as a root in association with the keyword, and a first keyword included in the document to be searched. For each of the keywords different from the first keyword included in the document corresponding to the first keyword, A second keyword having a predetermined number or more is specified, the tree structure keyword is created so that the second keyword is a child of the first keyword, and the created tree structure keyword is designated as the first keyword. A narrowing condition registering unit that registers in the narrowing condition storage unit in association with one keyword.

  In the advertisement bidding system according to the present invention, the bid accepting unit accepts an input of a keyword, reads out the keyword of the tree structure corresponding to the accepted keyword from the refinement condition storage unit, and A keyword may be output, and one of the keywords in the output tree structure may be accepted as the narrowing condition.

  According to another aspect of the present invention, there is provided a document search system that displays the advertisement bid in the advertisement bidding system together with a search result of the document corresponding to the keyword that is narrowed down based on a filtering condition. A filtering condition storage unit that stores the filtering condition in association with the keyword, a keyword input unit that receives input of the keyword, a document search unit that searches for the document corresponding to the keyword, Of the narrow-down conditions, a narrow-down condition acquisition unit that reads the narrow-down conditions corresponding to keywords from the narrow-down condition storage unit, a search result display unit that displays the narrow-down conditions together with the search result of the document, A refinement condition input unit that accepts at least one designation, a refinement processing unit that refines the search result based on the designated refinement condition, and the finger And further comprising an advertisement acquisition unit that acquires the advertisement corresponding to been narrowing condition from the condition by ad storage unit, and a advertisement display unit that displays the advertisement together with the narrowed search results above.

  According to the document search system of the present invention, it is possible to present a condition for narrowing a search result according to a keyword to the user, and narrow the search result according to the user's selection. Therefore, the user does not have to come up with a new keyword to narrow down the search result, which is convenient. Further, according to the document search system of the present invention, when the search result is narrowed down based on the narrowing condition, an advertisement corresponding to the narrowing condition can be displayed together with the narrowed search result. As the search results are narrowed down, the intention of the user's search becomes clearer. Therefore, the advertiser can set the advertisement to be displayed when the search result is narrowed down to the extent that the user's intention is revealed. Therefore, it is possible to display an advertisement according to the user's intention, and a high advertising effect is expected.

  Further, in the document search system of the present invention, the narrowing-down condition storage unit manages the narrowing-down conditions in a tree structure with the keyword as a root, and the tree structure is divided into the narrowing-down conditions belonging to lower layers, You may make it comprise so that the number of the said documents contained in the said refined search result can be decreased.

  In the document search system of the present invention, the narrow-down condition acquisition unit reads out the child condition that is the narrow-down condition that is a child of the designated narrow-down condition from the narrow-down condition storage unit, and displays the advertisement The unit may display the child condition together with the narrowed search result and the advertisement. In this case, the document search system can narrow down the search results sequentially by presenting narrowing conditions in order from the root of the tree structure and narrowing down according to the narrowing conditions.

  In the document search system of the present invention, the advertisement storage unit by condition stores the advertisement and the amount of the advertisement in association with the narrowing-down condition, and the advertisement display unit The advertisements may be displayed in descending order of the amount of the advertisement. In this case, the advertiser can set to change the exposure of the advertisement according to the degree of narrowing down even for the same advertisement. Therefore, it is possible to flexibly set conditions for displaying advertisements.

  In the document search system of the present invention, the narrowing-down condition storage unit stores an additional keyword to be added to the keyword in association with the keyword, and the narrowing-down processing unit includes the keyword and The document that matches both the corresponding additional keyword may be searched.

  In the document search system of the present invention, the narrowing-down condition acquisition unit may read out a predetermined number of the narrowing-down conditions corresponding to the keyword from the narrowing-down condition storage unit.

  Further, in the document search system of the present invention, the narrowing-down condition input unit accepts designation of any of the predetermined number of narrowing conditions, or designation that is not any of the predetermined number of narrowing conditions, When it is specified that none of the predetermined number of narrow-down conditions is specified, the narrow-down condition acquisition unit receives the predetermined number of the narrow-down conditions corresponding to the keyword from the narrow-down condition storage unit. The second narrowing condition other than the first narrowing condition that is the narrowing condition is read, and the narrowing processing unit narrows down the search result based on the second narrowing condition. Also good.

  In the document search system of the present invention, the narrow-down condition storage unit stores the number of times the narrow-down condition is selected in association with the narrow-down condition for each keyword, and obtains the narrow-down condition The unit may read out only a predetermined number of the narrow-down conditions corresponding to the keywords from the narrow-down condition storage unit in the order of the number of times.

  Another aspect of the present invention is a method for bidding an advertisement that is output together with a search result of a document corresponding to a keyword that has been narrowed down based on a narrowing condition, in association with the narrowing condition, The computer that stores the advertisement and the unit price of the advertisement in a memory receives the narrowing condition, the input of the advertisement and the unit price of the advertisement, and associates the received advertisement and the reception with the received narrowing condition. The unit price is registered in the memory.

  In the advertisement bidding method of the present invention, the narrowing-down condition is a keyword to be added to the keyword that constitutes a tree structure with the keyword as a root, and the search result is narrowed down based on the narrowing-down condition Is performed by searching the document corresponding to the narrowing condition, the keyword that is the root of the tree structure, and another narrowing condition that is an ancestor of the narrowing condition in the tree structure, The computer stores, in the memory, the filtering condition of the tree structure having the keyword as a root in association with the keyword, and for each of the first keywords included in the document to be searched, the first keyword Of the keywords different from the first keyword included in the document corresponding to one keyword, the application frequency is a predetermined number or more. A certain second keyword is specified, the keyword of the tree structure is created so that the second keyword becomes a child of the first keyword, and the created keyword of the tree structure corresponds to the first keyword In addition, it may be registered in the memory.

  According to another aspect of the present invention, there is provided a document search method for displaying the advertisement bid by the advertisement bidding method together with a search result of the document corresponding to the keyword narrowed down based on a narrowing condition. A computer having a CPU and a memory stores the narrow-down condition in the memory in association with a keyword, accepts an input of the keyword, searches for the document corresponding to the keyword, and corresponds to the keyword The filtering condition is read from the memory, the filtering condition is displayed together with the search result of the document, designation of at least one of the filtering conditions is accepted, and the search result is displayed based on the filtering condition. Filter, acquire the advertisement corresponding to the filter condition from the memory, and display the advertisement together with the filtered search result And the.

  According to another aspect of the present invention, there is provided a program for bidding an advertisement that is output together with a search result of a document corresponding to a keyword that has been narrowed based on a narrowing condition, and is associated with the narrowing condition The computer storing the advertisement and the unit price of the advertisement in a memory receives the narrowing condition, the input of the advertisement and the unit price of the advertisement, and the received in association with the received narrowing condition And registering the advertisement and the accepted unit price in the memory.

  Further, in the program of the present invention, the narrowing-down condition is a keyword for adding to the keyword that constitutes a tree structure with the keyword as a root, and the narrowing of the search result based on the narrowing-down condition is as follows: The computer is performed by searching the document corresponding to the narrowing condition, the keyword that is a root of the tree structure, and another narrowing condition that is an ancestor of the narrowing condition in the tree structure, and the computer In addition, for each of the first keyword included in the document to be searched and the step of storing in the memory the filtering condition of the tree structure having the keyword as a root in association with the keyword Of the keywords different from the first keyword included in the document corresponding to one keyword, the application frequency is The second keyword that is greater than or equal to the number is specified, the tree structure keyword is created so that the second keyword is a child of the first keyword, and the created tree structure keyword is designated as the first keyword. The step of registering in the memory in association with the keyword may be further executed.

  According to another aspect of the present invention, the advertisement that is bid by causing the computer to execute the program is displayed together with a search result of the document corresponding to the keyword that is narrowed down based on the filtering condition. A program for storing the narrowing-down condition in the memory in association with a keyword in a computer including a CPU and a memory, receiving the keyword input, and the keyword corresponding to the keyword A step of searching for a document, a step of reading out the filtering condition corresponding to the keyword from the memory, a step of displaying the filtering condition together with a search result of the document, and at least one of the filtering condition A step of accepting designation, and filtering the search results based on the filtering condition And step, a step of acquiring the advertisement corresponding to the narrowing-down condition from the memory, and displaying the advertisement together with the narrowed search results above, and thereby the execution.

  According to another aspect of the present invention, there is provided a control method for a client system, wherein the server is communicably connected to the client system and generates screen data for displaying an advertisement on the client system together with a document search result. A system storing a plurality of search conditions for narrowing down the search results matching the keyword in association with a keyword; and storing the advertisement in the memory in association with the search condition Transmitting to the client system screen data for causing the client system to accept an input of the keyword from a user and causing the server system to transmit the accepted keyword. The keyword that receives the keyword Searching the document that matches the keyword; reading the narrowing condition corresponding to the keyword from the memory; and sending the narrowing condition together with the document search result to the client system. Displaying screen data for displaying at least one of the narrowing conditions, and receiving information designating the narrowing condition transmitted from the client system; and A step of narrowing down the search result based on the specified narrowing condition, a step of acquiring the advertisement corresponding to the designated narrowing condition from the memory, and displaying the advertisement together with the narrowed search result For transmitting screen data to the client system. And it is executed and up, the.

  Other problems and solutions to be disclosed by the present application will be made clear by the embodiments of the invention and the drawings.

  According to the present invention, an effective advertisement can be bid.

  Hereinafter, a document search system according to an embodiment of the present invention will be described. The document search system according to the present embodiment searches for a document such as a Web page according to a keyword, displays the search result, asks a question to narrow down the search result, and displays the search result according to an answer from the user. The advertisement is displayed according to the degree of narrowing and narrowing down. It is assumed that the document search system of this embodiment is realized as, for example, a Web site search engine. In the present embodiment, it is assumed that the document to be searched is a Web page written in HTML (HyperText Markup Language).

== System configuration ==
FIG. 1 is a diagram showing the overall configuration of the document search system of this embodiment. The document search system according to this embodiment includes an advertiser terminal 10, a user terminal 20, and a search server 30. The advertiser terminal 10, the user terminal 20, and the search server 30 are connected to a communication network 40 and can communicate with each other. The communication network 40 is, for example, the Internet, a LAN (Local Area Network), a VAN (Value Added Network), or the like. The communication network 40 is constructed by, for example, Ethernet (registered trademark), a public telephone line network, a wireless communication network, or the like. In the present embodiment, it is assumed that the communication network 40 is the Internet, and the advertiser terminal 10 and the user terminal 20 are accessing the search server 30 by HTTP (HyperText Transfer Protocol).

  The search server 30 (corresponding to the server system of the present invention) is a computer such as a personal computer or a workstation that searches documents. The search server 30 is assumed to be a search engine that searches a Web page on the Internet. As will be described later, the search server 30 displays an advertisement together with a document search result.

  The advertiser terminal 10 is a computer operated by the advertiser, such as a personal computer, a workstation, a PDA (Personal Digital Assistance), or a mobile phone terminal. In the advertiser terminal 10, a so-called browser is operating, and the advertiser operates the browser to access the search server 30 and performs settings related to the advertisement described later.

  The user terminal 20 (corresponding to the client system of the present invention) is a computer such as a personal computer, a workstation, a PDA, or a mobile phone terminal operated by a user who searches for a document. The browser also operates on the user terminal 20, and the user accesses the search server 30 by operating the browser.

== Configuration of Search Server 30 ==
FIG. 2 is a diagram illustrating a hardware configuration of the search server 30. The search server 30 includes a CPU 301, a memory 302, a storage device 303, a communication interface 304, an input device 305, and an output device 306. The storage device 303 stores various data and programs, for example, a hard disk drive, a flash memory, a CD-ROM drive, and the like. The communication interface 304 is an interface for connecting to the communication network 40. The communication interface 304 is, for example, an adapter for connecting to Ethernet (registered trademark), a modem for connecting to a telephone line network, a wireless communication device for connecting to a wireless communication network, or the like. The input device 305 is a keyboard, mouse, touch panel, or the like that accepts data input. The output device 306 is, for example, a display or a printer that outputs data.

  FIG. 3 is a diagram illustrating a software configuration of the search server 30. The search server 30 includes a crawl processing unit 311, a question information generation unit 321, a bid acceptance unit 322, a search request reception unit 331, a search processing unit 332, a question acquisition unit 333, a search result transmission unit 334, a question answer reception unit 335, A processing unit 336, an advertisement acquisition unit 337, an accounting processing unit 338, an index database 351, a question database 352, an advertisement database 353, and a bid database 354. Note that the crawl processing unit 311, question information generation unit 321, bid acceptance unit 322, search request reception unit 331, search processing unit 332, question acquisition unit 333, search result transmission unit 334, question answer reception unit 335, narrowing processing unit 336, the advertisement acquisition unit 337, and the billing processing unit 338 are realized by the CPU 301 included in the search server 30 reading out the program stored in the storage device 303 to the memory 302 and executing the program. Further, the index database 351, the question database 352, the advertisement database 353, and the bid database 354 are realized as storage areas provided by the memory 302 and the storage device 303.

  The index database 351 stores information related to documents to be searched (hereinafter referred to as index information). FIG. 4 is a diagram illustrating a configuration example of the index information stored in the index database 351. The index information includes a document URL, a document title, a document summary, and a keyword related to the document.

  The crawl processing unit 311 creates index information related to a document provided by another computer connected to the communication network 40. For example, the crawl processing unit 311 acquires a Web page indicated by a known URL, extracts URLs of other Web pages linked from the acquired Web page, and acquires a document indicated by the extracted URL. That is, the crawl processing unit 311 provides a function called a crawler, a robot, or a spider.

  The crawl processing unit 311 extracts the title of the acquired document, extracts words included in the document as keywords, and summarizes the content of the document. For example, keywords can be extracted from a document by morphological analysis. As for the keyword, an evaluation value may be calculated by TF-IDF (Term Frequency-Inverse Document Frequency), and only those having an evaluation value higher than a predetermined threshold may be selected.

  The crawl processing unit 311 creates index information including a title, a summary, and a keyword together with the URL of the document and registers the index information in the index database 351. In addition to the above, the index information can include items used in general search engines.

  The question database 352 (corresponding to the narrowing condition storage unit of the present invention) is information including a condition for further narrowing down the search result corresponding to a certain keyword (hereinafter referred to as question information. Corresponding to the narrowing condition of the present invention). Memorize). In the present embodiment, the condition for further narrowing down the search results corresponding to a certain keyword is a new keyword added to the keyword (hereinafter referred to as an additional keyword). The question database 352 manages additional keywords for each keyword. The additional keywords are managed in a tree structure as shown in FIG. In the example of FIG. 5, the tree structure has the keyword 41 “board” as a root. For example, when narrowing down using the additional keyword “shop” of the node 43, all the nodes on the path from the route 41 to the node 43, that is, all the keywords of the route 41, the node 42, and the node 43, that is, “ This is achieved by searching “Board Surfboard Shop” using AND conditions.

  A configuration example of the question information registered in the question database 352 is shown in FIG. The question information includes a hierarchy ID, a parent hierarchy ID, a keyword, an additional keyword, a question, and an appearance frequency. The hierarchy ID is identification information of the question information, and is also a number for identifying a node in the tree structure. The parent hierarchy ID is the hierarchy ID of the parent node. In the present embodiment, it is assumed that the question information about the route 41 is not registered in the question database 352, and only the question information about the additional keyword is registered. Therefore, the parent hierarchy ID is not set in the question information corresponding to the child node of the route 41 (null is set). Further, the question information includes a keyword as a route. A question is a message for explaining an additional keyword of question information to a user who performs a search. The appearance frequency is the number of index information including all keywords on the path from the root to the node.

  The question information generation unit 321 generates question information based on the index information stored in the index database 351, and registers the generated question information in the question database 352. The details of the question information generation process by the question information generation unit 321 will be described later.

  The advertisement database 353 stores information related to advertisements (hereinafter referred to as advertisement information). FIG. 7 is a diagram illustrating a configuration example of advertisement information stored in the advertisement database 353. The advertisement information includes an advertisement ID, an advertiser ID, a copy, a message, and a link destination. The advertisement ID is identification information that identifies advertisement information. The advertiser ID is identification information that identifies the advertiser. A copy and a message are information displayed as an advertisement. In this embodiment, the advertisement is displayed as text information. However, a still image, a moving image, audio, or the like is output instead of a copy and a message, or together with at least one of a copy and a message. In that case, a URL in which an image, sound, or the like is stored is included in the advertisement information. A link is set to the displayed advertisement, and the link destination is the link destination of the advertisement information. It is assumed that advertisement information is registered in the advertisement database 353 in advance.

  The bid database 354 stores information related to conditions for displaying advertisements (hereinafter referred to as bid information). The advertisement database 353 and the bid database 354 correspond to the condition-specific advertisement storage unit of the present invention. A configuration example of bid information stored in the bid database 354 is shown in FIG. The bid information includes a hierarchy ID, an advertisement ID, and a bid price. The bid price is an amount charged to the advertiser when the advertisement is displayed. As will be described later, when a search using an additional keyword indicated by the hierarchical ID is performed, the advertisement indicated by the advertisement ID is displayed, and the amount of the bid price is charged to the advertiser each time it is displayed.

  The bid reception unit 322 receives each item of bid information from the advertiser terminal 10 and registers the bid information including each received item in the bid database 354. The details of the bid information registration process by the bid acceptance unit 322 will be described later.

The search request receiving unit 331 receives a command (hereinafter referred to as a search request) for instructing to search for a document from the user terminal 20. The search request includes a keyword.
The search processing unit 332 searches for a document corresponding to the keyword specified by the user. In the present embodiment, the search processing unit 332 searches the index database 351 for index information corresponding to a keyword included in the search request.
The question acquisition unit 333 acquires question information corresponding to the keyword from the question database 352. The question information acquisition process by the question acquisition unit 333 will be described later.

A search result transmission unit 334 (corresponding to the advertisement display unit of the present invention) displays a list of index information searched by the search processing unit 332 and a list of question information acquired by the question acquisition unit 333. Data is transmitted to the user terminal 20.
The search result transmission unit 334 transmits the document search result to the user terminal 20. Screen data for displaying a list of index information searched by the search processing unit 332, a list of question information acquired by the question acquisition unit 333, and a list of advertisement information acquired by the advertisement acquisition unit 337 is displayed on the user terminal 20. Send.

  The question answer receiving unit 335 (corresponding to the narrowing condition input unit of the present invention) receives selection of question information from the user. As will be described later, the question answer receiving unit 335 receives the hierarchical ID of the question information selected by the user from the user terminal 20.

  The narrowing processing unit 336 narrows down the document search result. In the present embodiment, the narrowing processing unit 336 corresponds to the additional keyword of the question information corresponding to the hierarchical ID received by the question answer receiving unit 335 and the keyword or additional keyword of the ancestor node of the question information in the tree structure. Index information is searched from the index database 351.

  The advertisement acquisition unit 337 acquires an advertisement corresponding to the keyword from the advertisement database 353. As will be described later, the advertisement acquisition unit 337 reads the advertisement ID corresponding to the hierarchical ID of the question information selected by the user from the bid database 354 and acquires the advertisement information corresponding to the read advertisement ID from the advertisement database 353. The advertisement information acquisition process by the advertisement acquisition unit 337 will be described later.

  Each time the screen data is transmitted to the user terminal 20, the billing processing unit 338 acquires a bid price corresponding to the advertisement information included in the screen data from the bid database 354, and the acquired bid price is the advertiser ID of the advertisement information. Charge the advertiser indicated by Note that the billing processing of the advertisement by the billing processing unit 338 can be performed by a general billing processing technique.

  Hereinafter, processing in the search server 30 will be described.

== Question information generation process ==
First, the question information registration process will be described. FIG. 9 is a diagram showing the flow of the question information registration process. The question information registration process shown in FIG. 9 can be performed each time index information is registered by the crawl processing unit 311, for example.

  The question information generating unit 321 reads the index information keywords stored in the index database 351 without duplication (S501). The question information generation unit 321 performs the following processing for each read keyword.

  The question information generation unit 321 sets a null character to the variable parent question (S502), and sets null to the variable parent ID (S503). The question information generation unit 321 calls the question registration detailed process shown in FIG. 10 with the keyword as the first argument and the second argument, the parent question as the second argument, and the parent ID as the third argument (S504).

  The question information generation unit 321 uses the first argument as a root keyword, the second argument as a search keyword, the third argument as a parent question, the fourth argument as a parent ID (S521), and index information corresponding to the search keyword. Read from the index database 351 and set to the variable L (S522).

  The question information generation unit 321 calculates the appearance frequency of the keyword included in the index information in L, and sets the appearance frequency greater than a predetermined value to the variable K (S523). The question information generation unit 321 removes the search keyword from K (S524), and ends the process if K is empty (S525: YES).

  On the other hand, if K is not empty (S525: NO), the question information generation unit 321 performs the following processing using each keyword included in K as an additional keyword.

  If the parent question is an empty character (S526: YES), the question information generating unit 321 sets the additional keyword as a character string obtained by concatenating the character “no” and the root keyword (S527), and the parent question is empty. If not (S526: NO), a character string in which the character “no” and the additional keyword are connected to the parent question is used as the question (S528). For example, if the root keyword is “board” and the additional keyword is “surfboard”, when the parent question is empty, the question is “surfboard board” and when the parent question is “ultimate board”, the question is Become the ultimate board surfboard.

The question information generation unit 321 generates a unique hierarchical ID (S529), generates question information including a hierarchical ID, a parent ID, a root keyword, an additional keyword, a question, and an appearance frequency, and uses the generated question information as a question database. It is registered in 352 (S530). The question information generation unit 321 sets the search keyword added with an additional keyword as the next keyword (S531), sets the root keyword as the first argument, the next keyword as the second argument, the question as the third argument, and the hierarchy ID. As the fourth argument, the question registration detail process is recursively called (S532).
The question information generation unit 321 repeats the above processing and ends the processing.

Returning to FIG. 9, the question information generation unit 321 repeats the above processing for each keyword.
In this way, the question information is registered in the question database 352. According to the document search system of this embodiment, additional keywords can be managed in a tree structure based on the appearance frequency of keyword combinations included in a document to be searched. In the tree structure, the number of index information corresponding to the additional keyword and the keyword of its ancestor decreases as it goes down.

== Bid processing ==
Next, processing for receiving bid information from an advertiser will be described. FIG. 11 is a diagram showing a flow of bid information registration processing. Moreover, FIGS. 12-15 is a figure which shows an example of a screen as a user interface used for the registration process of bid information.

  The bid reception unit 322 transmits screen data for displaying the screen 61 for inputting a keyword to the advertiser terminal 10. FIG. 12 is a diagram illustrating an example of the screen 61. In the example of FIG. 12, the screen 61 includes a keyword input field 611. When a keyword is input in the input field 611 and the search button 612 is pressed, the advertiser terminal 10 transmits the keyword input in the input field 611 to the search server 30. The bid accepting unit 322 receives the keyword transmitted from the advertiser terminal 10 and accepts an input of a keyword to be advertised (hereinafter referred to as an advertising target keyword) from the advertiser (S541).

  The bid reception unit 322 reads the question information in which the advertising target keyword matches the keyword or the keyword of the question information in which the advertising target keyword is included in the additional keyword from the question database 352 without duplication (S542).

  The bid accepting unit 322 reads out from the question database 352 the keyword that matches the keyword of the question information for each of the read keywords (S543). The bid acceptance unit 322 creates screen data of the screen 62 that displays the question information in a tree shape for each keyword based on the hierarchical ID and the parent hierarchical ID, and transmits the screen data to the advertiser terminal 10 (S544).

  FIG. 13 is a diagram illustrating an example of the screen 62. In the example of FIG. 13, the screen 62 includes a keyword display field 621 and a question information selection field 622. In the selection field 622 of the screen 62, questions of question information are displayed in a tree shape, and each question information can be selected. When the question information is selected, the advertiser terminal 10 transmits the hierarchical ID of the selected question information to the search server 30. The bid reception unit 322 receives the hierarchy ID transmitted from the advertiser terminal 10 and receives the specification of the question information from the advertiser (S545).

  The bid reception unit 322 reads out the advertisement information corresponding to the advertiser ID indicating the advertiser (S546), creates screen data of the screen 63 that displays a list of the read advertisement information, and transmits the screen data to the advertiser terminal 10. (S547).

  FIG. 14 is a diagram illustrating an example of the screen 63. In the example of FIG. 14, the screen 63 includes a display field 631 for keywords input from the advertiser, a display field 632 for question information designated by the advertiser, and a selection field 633 for advertisements to be displayed. In the selection column 633, a list of advertisement information is displayed. When advertisement information is selected from the list, the advertiser terminal 10 transmits the advertisement ID of the selected advertisement information to the search server 30. The bid reception unit 322 receives the advertisement ID transmitted from the advertiser terminal 10 and receives the designation of the advertisement to be displayed from the advertiser (S548).

  The bid reception unit 322 transmits screen data for displaying the screen 64 for inputting the bid amount to the advertiser terminal 10. FIG. 15 is a diagram illustrating an example of the screen 64. In the example of FIG. 15, the screen 64 includes a keyword display field 641 input from the advertiser, a specified question information display field 642, a specified advertisement display field 643, and a bid price input field 644. ing. When the bid price is entered in the input field 644 and the bid button 645 is pressed, the advertiser terminal 10 transmits the bid price entered in the input field 644 to the search server 30. The bid accepting unit 322 receives the bid price transmitted from the advertiser terminal 10 and accepts the bid price input from the advertiser (S549).

  The bid receiving unit 322 generates bid information including the hierarchy ID, the advertisement ID, and the bid price received from the advertiser as described above, and registers the generated bid information in the bid database 354 (S550).

  As described above, bid information is registered. In the document search system according to the present embodiment, the advertiser can specify a bid amount that is a unit price of an advertisement and a combination of keywords to be advertised. Therefore, an advertiser's flexible advertising strategy becomes possible.

== Document search processing ==
Next, document search processing will be described. In the document search process, an advertisement is output together with the search result. FIG. 16 is a diagram showing the flow of document search processing.

When the search request receiving unit 331 receives a search request from the user terminal 20 (S701), the search processing unit 332 reads a predetermined number of index information including the keyword included in the search request (S702). The predetermined number can be an arbitrary number such as 10, 20, 100, for example.
The question acquisition unit 333 reads two pieces of question information corresponding to the search request keyword in descending order of appearance frequency (S703). Note that the number of pieces of question information to be read is not limited to two and may be an arbitrary number.
The search result transmission unit 334 transmits the screen data of the screen 81 displaying the link that can select the above two pieces of question information and “None” and the list of index information to the user terminal 20 (S704).

  An example of the screen 81 is shown in FIG. In the example of FIG. 17, on the screen 81, two question information links 811 and 812 and a “none” link 813 are displayed, and a document search result list 815 is displayed. When any one of the links 811 to 813 is clicked on the screen 81, the user terminal 20 displays the question information layer ID or information indicating “neither” (in this embodiment, the layer ID “−1”). To the search server 30.

  When the question answer receiving unit 335 receives the hierarchy ID from the user terminal 20, that is, when the above two pieces of question information or “None” is selected (S705: YES), the search result narrowing process shown in FIG. This is performed (S706).

  First, the narrowing processing unit 336 sets null to the parent ID, sets the excluded question list and the excluded keyword as empty lists, and sets the keyword included in the search request as the search keyword (S721).

  When the hierarchy ID received by the question answer receiving unit 335 is “−1”, that is, when “None” is selected (S722: YES), the narrowing processing unit 336 selects the above two pieces of question information. The hierarchical ID is added to the excluded question list (S723), and the additional keyword of the two question information is added to the excluded keyword (S724). The question acquisition unit 333 reads from the question database 352 the question information whose parent hierarchy ID is the parent ID and does not include the hierarchy ID in the excluded question list (S725), and the advertisement acquisition unit 337 reads from the bid database 354, Bidding information corresponding to the hierarchy ID of the question information read in step S725 is read (S726).

  On the other hand, when the hierarchy ID received by the question answer receiving unit 335 is one of the hierarchy IDs of the question information read out in step 703 in FIG. 16, that is, when the user selects any of the questions ( (S722: NO), the narrowing processing unit 336 sets the hierarchy ID as the parent ID (S727), and adds the additional keyword of the question information corresponding to the hierarchical ID to the search keyword (S728). The advertisement acquisition unit 337 reads bid information corresponding to the parent ID from the bid database 354 (S729).

Next, the narrowing screen transmission process shown in FIG. 19 is performed (S730).
The narrowing processing unit 336 reads a predetermined number of pieces of index information stored in the index database 351 that include the search keyword and do not include the excluded keyword (S741). The advertisement acquisition unit 337 reads a predetermined number of advertisement information corresponding to the advertisement IDs of the read bid information from the advertisement database 353 in descending order of bid price (S742). The predetermined number can be an arbitrary number such as 3, 5, 10, or the like. The question acquisition unit 333 reads, from the question database 352, two pieces of question information whose parent hierarchical ID is the parent ID and whose hierarchical ID is not included in the excluded question list in descending order of appearance frequency (S743).

  The search result transmitting unit 334 selects the two question information read in step S743 and a link that can select “None”, the list of index information read in step S741, and the list of advertisement information read in step S742. Is transmitted to the user terminal 20 (S744). The user terminal 20 outputs a screen 82 that displays question information for further narrowing down the search results, document search results, and advertisements according to the narrowing down based on the received screen data.

  An example of the screen 82 is shown in FIG. In the example of FIG. 20, on the screen 82, two question information links 821 and 822 and an “neither” link 823 are displayed, an index information list 824, an advertisement information list 825, Is displayed. In the list 825, advertisements corresponding to the question information selected by the user are displayed.

  The keyword specified by the user often includes a plurality of meanings. For example, the search result corresponding to the keyword “board” includes all documents related to “surfboard” and “snowboard”. According to the document search system of this embodiment, the search result of the document corresponding to the keyword specified by the user is displayed and a question for narrowing down the search result is presented to the user. The search result can be narrowed down by searching the document again by adding an additional keyword corresponding to the question selected by the user. Therefore, the user can narrow down the search result by simply answering the question from the document search system without thinking of a new keyword himself. Therefore, the user can search for a document efficiently.

  Further, in the document search system of the present embodiment, the additional keywords are managed in a tree structure, and the number of search result documents decreases as the search is performed using the lower additional keywords. Therefore, the search server 30 can narrow down the search results sequentially by presenting the questions of the nodes below the question selected by the user for the next narrowing down. Therefore, the search server 30 can efficiently search for a question for narrowing down, and the search processing efficiency can be improved.

  In the document search system of this embodiment, the advertiser can specify an advertisement in association with a favorite additional keyword hierarchy. In the document search system of this embodiment, an advertisement is displayed when the search result is narrowed down using corresponding question information. When the user narrows down the search results, the intention of the user's search becomes clearer. Therefore, the advertiser can set the advertisement to be displayed when the search result is narrowed down to the extent that the user's intention is revealed. Therefore, it is possible to display an advertisement according to the user's intention, and a high advertising effect is expected.

  For example, the search result corresponding to the keyword “board” includes all documents related to “surfboard”, documents related to “snowboard”, and the like, but the user is searching for “surfboard”. Advertisements related to “Snowboard” can be prevented from being displayed, and when a user answers (clicks) a question, it becomes clear that “Surfboard” is intended. Can be displayed.

  In the document search system of this embodiment, the advertiser can set the bid price of the advertisement. In the document search system of this embodiment, a predetermined number of advertisements are displayed in the list 825 of the screen 82 in descending order of the bid price of advertisements. Therefore, since the advertiser can set a bid according to the degree of narrowing down the search result, it is possible to flexibly set conditions for displaying the advertisement.

  In the present embodiment, the additional keyword is adopted as a condition for narrowing down the search results. However, the present invention is not limited to this. For example, each index information is assigned to one or more categories in advance, and It is also possible to narrow down using one.

  In the present embodiment, the question information generation unit 321 automatically registers the question information based on the index information. However, the present invention is not limited to this. For example, for the advertiser to narrow down for each keyword. An additional keyword or question may be input, and question information including the input additional keyword or question may be registered in the question database 352.

  In the present embodiment, the question information is generated so as to form a tree structure according to the frequency of the combination of keywords included in the index information. However, the present invention is not limited to this. For example, the concept of the meaning of each keyword May be managed so that the question information forms a tree structure in accordance with the concept. Also, the advertiser may specify a tree structure for all additional keywords, or the advertiser may be able to edit the tree structure automatically generated by the search server 30.

  In the present embodiment, the pay per view method is charged every time an advertisement is displayed. However, the present invention is not limited to this. For example, the advertisement may be charged each time it is clicked or clicked. After that, you may be charged when sales actually increase.

  In the present embodiment, two pieces of question information are presented in order of appearance frequency. For example, the number of times the user clicked the question information is stored in association with the question information, and the number of times is large. You may make it show in order.

  Although the present embodiment has been described above, the above embodiment is intended to facilitate understanding of the present invention and is not intended to limit the present invention. The present invention can be changed and improved without departing from the gist thereof, and the present invention includes equivalents thereof.

It is a figure which shows the whole structure of the document search system of this embodiment. 2 is a diagram illustrating a hardware configuration of a search server 30. FIG. 2 is a diagram illustrating a software configuration of a search server 30. FIG. It is a figure which shows the structural example of index information. It is a figure which shows a mode that an additional keyword is managed by a tree structure. It is a figure which shows the structural example of question information. It is a figure which shows the structural example of advertisement information. It is a figure which shows the structural example of bid information. It is a figure which shows the flow of a registration process of question information. It is a figure which shows the flow of a question registration detailed process. It is a figure which shows the flow of a registration process of bid information. It is a figure which shows an example of the screen 61 used for the registration process of bid information. It is a figure which shows an example of the screen 62 used for the registration process of bid information. It is a figure which shows an example of the screen 63 used for the registration process of bid information. It is a figure which shows an example of the screen 64 used for the registration process of bid information. It is a figure which shows the flow of the search process of a document. It is a figure which shows an example of the screen 81 which displays the search result of a document. It is a figure which shows the flow of a search result narrowing-down process. It is a figure which shows the flow of the transmission process of a narrowing-down screen. It is a figure which shows an example of the screen 82 which displays an advertisement with the search result narrowed down.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Advertiser terminal 20 User terminal 30 Search server 40 Communication network 301 CPU
302 Memory 303 Storage Device 304 Communication Interface 305 Input Device 306 Output Device 311 Crawl Processing Unit 321 Question Information Generating Unit 322 Bid Accepting Unit 331 Search Request Receiving Unit 332 Search Processing Unit 333 Question Obtaining Unit 334 Search Result Transmitting Unit 335 Question Answer Receiving Unit 336 Narrow down processing unit 337 Advertisement acquisition unit 338 Charge processing unit 351 Index database 352 Question database 353 Advertisement database 354 Bid database

Claims (18)

A system that bids advertisements that are output together with the search results of documents corresponding to keywords, based on the search criteria,
In association with the narrowing-down conditions, a condition-specific advertisement storage unit that stores the advertisement and the unit price of the advertisement;
A bid accepting unit that accepts input of the narrowing conditions, the advertisement, and the unit price;
A bid registration unit that registers the received advertisement and the received unit price in the conditional advertisement storage unit in association with the received narrowing condition;
An advertising bidding system comprising: The advertising bidding system according to claim 1,
The narrowing-down condition is a keyword for adding to the keyword, constituting a tree structure with the keyword as a root,
The search result filtering based on the filtering condition corresponds to the filtering condition, the keyword that is the root of the tree structure, and another filtering condition that is an ancestor of the filtering condition in the tree structure. By searching the document to be
The advertising bidding system includes:
A narrow-down condition storage unit that stores the narrow-down condition of the tree structure with the keyword as a root in association with the keyword;
For each of the first keywords included in the document to be searched, the application frequency is a predetermined number or more among keywords different from the first keyword included in the document corresponding to the first keyword. A second keyword is specified, the tree structure keyword is created such that the second keyword is a child of the first keyword, and the created tree structure keyword is associated with the first keyword. A refining condition registration unit to be registered in the refining condition storage unit,
An advertisement bidding apparatus comprising: The advertising bidding system according to claim 2,
The bid accepting unit accepts an input of a keyword, reads out the keyword of the tree structure corresponding to the accepted keyword from the refinement condition storage unit, outputs the keyword of the read tree structure, and outputs the output tree structure Accepting one of the keywords as the filtering condition,
An advertising bidding system characterized by The document search system for displaying the advertisement bid in the advertisement bidding system according to claim 1, together with a result of narrowing down a search result of the document corresponding to the keyword based on a filtering condition,
A refinement condition storage unit that stores the refinement condition in association with the keyword;
A keyword input unit for receiving input of the keyword;
A document search unit for searching for the document corresponding to the keyword;
A narrowing condition acquisition unit that reads the narrowing condition corresponding to the keyword from the narrowing condition storage unit;
A search result display unit for displaying the narrowing condition together with the search result of the document;
A refinement condition input unit that accepts at least one designation of the refinement conditions;
A narrowing processing unit that narrows down the search results based on the designated narrowing conditions;
An advertisement acquisition unit that acquires the advertisement corresponding to the specified narrowing condition from the conditional advertisement storage unit;
An advertisement display unit that displays the advertisement together with the refined search results;
A document retrieval system comprising: The document search system according to claim 4,
The refinement condition storage unit manages the refinement condition in a tree structure with the keyword as a root,
The tree structure is configured to reduce the number of documents included in the refined search results as the refinement conditions belonging to a lower layer are;
Document search system characterized by The document search system according to claim 5,
The narrow-down condition acquisition unit reads out the child condition that is the narrow-down condition to be a child of the specified narrow-down condition from the narrow-down condition storage unit,
The advertisement display unit displays the child condition together with the narrowed search result and the advertisement.
Document search system characterized by The document search system according to claim 4,
The condition-specific advertisement storage unit stores the advertisement and the amount of the advertisement in association with the narrowing-down condition,
The advertisement display unit displays the advertisement in descending order of the amount of the advertisement;
Document search system characterized by The document search system according to claim 4,
The narrow-down condition storage unit stores an additional keyword to be added to the keyword in association with the keyword,
The narrowing-down processing unit searches for the document that matches both the keyword and the corresponding additional keyword;
Information retrieval system characterized by The document search system according to claim 4,
The narrowing condition acquisition unit reads a predetermined number of the narrowing conditions corresponding to the keyword from the narrowing condition storage unit,
Document search system characterized by The document search system according to claim 9,
The refinement condition input unit accepts designation of any of the predetermined number of refinement conditions, or designation that is not any of the prescribed number of refinement conditions,
When it is specified that none of the predetermined number of narrow-down conditions is specified, the narrow-down condition acquisition unit receives the predetermined number of the narrow-down conditions corresponding to the keyword from the narrow-down condition storage unit. Reading a second narrowing condition other than the first narrowing condition, which is a narrowing condition, and the narrowing processing unit narrows down the search result based on the second narrowing condition;
Document search system characterized by The document search system according to claim 4,
The filtering condition storage unit stores the number of times the filtering condition is selected in association with the filtering condition for each keyword.
The narrowing-down condition acquisition unit reads out the narrowing-down conditions corresponding to the keywords from the narrowing-down condition storage unit only in a predetermined number in the order of the number of times,
Document search system characterized by A method for bidding an advertisement that outputs a search result of documents corresponding to a keyword together with a search result based on a filtering condition,
A computer that stores the advertisement and the unit price of the advertisement in a memory in association with the filtering condition,
Accepting the filtering conditions, the advertisement and the unit price of the advertisement,
Registering the accepted advertisement and the accepted unit price in the memory in association with the accepted filtering condition;
A bid method for advertising. The advertising bidding method according to claim 12,
The narrowing-down condition is a keyword for adding to the keyword, constituting a tree structure with the keyword as a root,
The search result filtering based on the filtering condition corresponds to the filtering condition, the keyword that is the root of the tree structure, and another filtering condition that is an ancestor of the filtering condition in the tree structure. By searching the document to be
The computer
In association with the keyword, the filtering condition of the tree structure having the keyword as a root is stored in the memory,
For each of the first keywords included in the document to be searched, the application frequency is a predetermined number or more among keywords different from the first keyword included in the document corresponding to the first keyword. A second keyword is specified, the tree structure keyword is created such that the second keyword is a child of the first keyword, and the created tree structure keyword is associated with the first keyword. Registering in the memory
A bid method for advertising. A document search method for displaying the advertisement bid by the advertisement bidding method according to claim 12 together with a search result of the document corresponding to the keyword narrowed down based on a filtering condition.
A computer comprising a CPU and a memory
In association with a keyword, the filtering condition is stored in the memory,
Accept input of the keyword,
Search for the document corresponding to the keyword,
Read the refinement condition corresponding to the keyword from the memory;
Display the filtering condition together with the search result of the document,
Accepts at least one designation of the refinement conditions;
Narrow the search results based on the filtering criteria,
Obtaining the advertisement corresponding to the filtering condition from the memory;
Displaying the advertisement with the refined search results;
Document search method characterized by A program for bidding advertisements that are output together with the search results of documents corresponding to keywords, based on the search conditions,
A computer that stores the advertisement and the unit price of the advertisement in a memory in association with the filtering condition,
Receiving the narrowing conditions, the advertisement, and the unit price of the advertisement;
Registering the received advertisement and the received unit price in the memory in association with the received narrowing condition;
A program for running The program according to claim 15,
The narrowing-down condition is a keyword for adding to the keyword, constituting a tree structure with the keyword as a root,
The search result filtering based on the filtering condition corresponds to the filtering condition, the keyword that is the root of the tree structure, and another filtering condition that is an ancestor of the filtering condition in the tree structure. By searching the document to be
In the computer,
Storing the narrow-down condition of the tree structure with the keyword as a root in the memory in association with the keyword;
For each of the first keywords included in the document to be searched, the application frequency is a predetermined number or more among keywords different from the first keyword included in the document corresponding to the first keyword. A second keyword is specified, the tree structure keyword is created such that the second keyword is a child of the first keyword, and the created tree structure keyword is associated with the first keyword. Registering in said memory;
A program for further execution. A program for displaying the advertisement bid by causing a computer to execute the program according to claim 15 together with a search result of the document corresponding to the keyword, together with a result narrowed down based on the filtering condition. And
In a computer equipped with a CPU and memory,
Storing the narrow-down condition in the memory in association with a keyword;
Receiving an input of the keyword;
Retrieving the document corresponding to the keyword;
Reading the refinement condition corresponding to the keyword from the memory;
Displaying the filtering condition together with a search result of the document;
Receiving a designation of at least one of the narrowing conditions;
Narrowing the search results based on the filtering condition;
Obtaining the advertisement corresponding to the filtering condition from the memory;
Displaying the advertisement with the refined search results;
A program for running A client system control method comprising:
A server system that is communicably connected to the client system and creates screen data for displaying an advertisement on the client system together with a document search result;
Storing in a memory a plurality of narrowing-down conditions for narrowing down the search results matching the keyword in association with the keyword;
Storing the advertisement in the memory in association with the filtering condition;
Causing the client system to accept input of the keyword from a user, and sending screen data to the client system for sending the accepted keyword to the server system;
Receiving the keyword transmitted from the client system;
Searching for the document matching the keyword;
Reading the refinement condition corresponding to the keyword from the memory;
Transmitting the screen data for causing the client system to display the narrowing condition together with the search result of the document and for accepting designation of at least one of the narrowing conditions to the client system;
Receiving information specifying the filtering condition transmitted from the client system;
Narrowing the search results based on the specified filtering condition;
Obtaining the advertisement corresponding to the specified filtering condition from the memory;
Transmitting screen data for displaying the advertisement together with the narrowed search results to the client system;
Performing the
A method for controlling a client system.

Images