JP2010198350A - Retrieval apparatus and retrieval method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a retrieval apparatus capable of presenting a suitable retrieval result to a user. <P>SOLUTION: A retrieval result generation unit 22 in the retrieval apparatus 20 includes: a retrieval word group obtaining part 22a for outputting a retrieval word included in a retrieval request to a retrieval word analysis device 30 and obtaining a retrieval word group as an analytical result by retrieval word analysis algorithm 31; a Web page information obtaining part 22b for referring to an index database 52 based on the retrieval word group and obtaining Web page information in which the retrieval word group is included; and an arrayed retrieval result obtaining part 22c for outputting the obtained Web page information to a display order determination device 60 and obtaining arrayed retrieval results generated by display order determination algorithm 61. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a search device and a search method for outputting a search result in response to a search request.

  Search engines that can be used through many search service sites on the Internet crawl web pages, store information on each web page collected by crawling, assist users in entering search terms, and search web pages in search results. A search service is provided by integrally using a plurality of functions such as display order determination.

  Non-Patent Document 1 discloses a study on unfairness (ranking bias) caused by the difference in the display order of Web pages in search results for each search service. Non-Patent Document 2 introduces the difference between the search engines of the two companies.

Yuu Hirate, Yasuaki Yoshida, Hayato Yamana, “Research Trends on Web Search Engine Ranking Bias”, [online], 18th Data Engineering Workshop, [Search February 12, 2009], Internet <URL: http://www.ieice.org/~de/DEWS/DEWS2007/pdf/c7-7.pdf> Takahiro Watanabe, “Six differences between Google and Yahoo! that SEO personnel should know” (Part 1), [online], [February 12, 2009 search], Internet <URL: http: // www. alphaseo.jp/seo-column/071019_101432.html>

  Thus, since the conventional search engine uses a unique algorithm in each function, the search results obtained differ depending on which search engine the user uses. Here, the user can select a search engine according to the preference or purpose, but cannot receive a search service other than the functions combined with the search engine provided in advance. Therefore, there is a problem that the user cannot always obtain a good search result.

  The present invention has been developed to solve the above-described problems, and an object of the present invention is to provide a search device and a search method capable of presenting good search results to a user.

  In order to solve the above-described problem, a search apparatus according to the present invention combines a search request acquisition unit that acquires a search request from a client apparatus, and a plurality of search algorithms and an index database based on the acquired search request. And a search result output unit that outputs the generated search result to the client device, and the index database relates to a search index and a Web page including the search index. A database associated with Web page information, wherein the plurality of search algorithms determine a search word analysis algorithm for analyzing a search word included in the search request and a display order of the Web page information, and Arranged based on the display order A display order determination algorithm for generating a result, and the search result generation unit outputs a search word included in the search request to a search word analysis unit in which the search word analysis algorithm is stored, and A search word group acquisition unit that acquires a search word group as an analysis result by the search word analysis algorithm from a search word analysis unit, and the index database is referred to based on the search word group, and the search word group is included A Web page information acquisition unit that acquires Web page information, and outputs the acquired Web page information to a display order determination unit in which the display order determination algorithm is stored, and from the display order determination unit to the display order determination algorithm An arrayed search result acquisition unit that acquires the generated arrayed search result, and the search result output unit includes And outputs a column already search results to the client device.

  Here, the search word analysis unit, the index database, and the display order determination unit may be provided integrally with the search device, or may be provided separately from each other so as to be communicable with the search device. . According to such a configuration, it is possible to present a good search result to a user of a client apparatus that has conventionally only received a search service that integrates a plurality of search functions.

  The search device further includes a characteristic storage unit that stores characteristics of a plurality of search word analysis algorithms and characteristics of a plurality of display order determination algorithms, and the search word group acquisition unit includes the search request and the search request One search word analysis algorithm is selected based on characteristics of a plurality of search word analysis algorithms, the search request is output to the search word analysis unit in which the selected search word analysis algorithm is stored, and the arranged search The result acquisition unit selects one display order determination algorithm based on the search request and the characteristics of the plurality of display order determination algorithms, and the display order determination unit stores the selected display order determination algorithm. Web page information is output.

  According to this configuration, one search word analysis algorithm and one display order determination algorithm suitable for the search are selected and used from the plurality of search word analysis algorithms and the plurality of display order determination algorithms. The result can be generated and output to the client device.

  The search device further includes a trend analysis unit that performs a trend analysis on the search word based on the Web page information, and the search result output unit displays the result of the trend analysis together with the arranged search result in the client. The structure which outputs to an apparatus may be sufficient.

  In addition, the search device further includes a recommendation information generation unit that generates recommendation information related to a news article or a product related to the search word based on the Web page information, and the search result output unit arranges the recommendation information as the arrayed The configuration may be such that the search result is output to the client device.

  The search request includes a user's browsing history, and the search word group acquisition unit generates an additional search word based on the browsing history, and the generated additional search word is searched together with the search word. It may be configured to output to the word analysis unit.

  The search term group acquisition unit transmits / receives data to / from the search term analysis unit using a predefined Web API, and the Web page acquisition unit uses the predefined Web API to transmit the index database and data The arranged search result acquisition unit may transmit / receive data to / from the display order determination unit using a predefined Web API.

  In order to solve the above problem, a search method according to the present invention combines a search request acquisition unit that acquires a search request from a client device, and a plurality of search algorithms and an index database based on the acquired search request. A search method by a search device comprising: a search result generation unit that generates a search result; and a search result output unit that outputs the generated search result to the client device, wherein the index database includes a search index And a search page analysis algorithm for analyzing a search term included in the search request, and display of a search result A display order determination algorithm for determining the order, and A search request acquisition step in which a search request acquisition unit acquires the search request from the client device, and a search term in which the search term analysis algorithm stores a search term included in the search request. A search word group acquisition step of outputting to the analysis unit and acquiring a search word group as an analysis result by the search word analysis algorithm, and the search result generation unit refers to the index database based on the search word group And a Web page information acquisition step for acquiring Web page information including the search term group, and the search result generation unit sends the acquired Web page information to the display order determination unit in which the display order determination algorithm is stored. A display order acquisition step of outputting and acquiring a display order by the display order determination algorithm; and the search The result generation unit arranges the Web page information based on the acquired display order, thereby generating an arranged search result generation step for generating an arranged search result, and the search result output unit generates the arranged search A search result output step of outputting a result to the client device.

  ADVANTAGE OF THE INVENTION According to this invention, a favorable search result can be shown with respect to a user.

It is a figure showing typically a search system provided with a search device concerning an embodiment of the present invention. It is a flowchart for demonstrating the search method by the search device which concerns on embodiment of this invention.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings as appropriate. Similar parts are denoted by the same reference numerals, and redundant description is omitted. FIG. 1 is a diagram schematically illustrating a search system including a search device according to an embodiment of the present invention.

  A search system 1 according to an embodiment of the present invention is a system that provides a customized search service by combining a plurality of search functions by mashup based on a search request from a client device 10. Search word analyzing device 30 (30A, 30B), crawl device 40, index generating device 50, and a plurality of display order determining devices 60 (60A, 60B).

<Client device 10>
The client device 10 is, for example, a computer device such as a personal computer, a workstation, a portable information terminal, a mobile phone, or the like, and includes an input unit 11 including a keyboard and a mouse, an output unit 12 including a display and a printer, and an input unit. 11 and a control unit 13 that controls the output unit 12. The control unit 13 transmits a search request input by the user operating the input unit 11 to the search device 20, receives a search result transmitted from the search device 20, and causes the output unit 12 to display the search result. The search request is data including a search word and auxiliary information used for the search. The auxiliary information is information related to the user's preference or purpose, such as past search history, browsing history (information related to the web page browsed immediately before), and the like. The search request includes a search term search condition (AND, OR, NOT condition), specification of the language, region, domain, etc. of the search target Web page, last update date, search target location in the Web page, search target Web page file Information such as format and site format may be included. The control unit 13 stores auxiliary information such as a search history, a browsing history, and the like by using a cookie or the like, generates a search request from the stored auxiliary information and the input search word, and searches the search device. 20 can be output.

<Search device 20>
The search device 20 includes a CPU (Central Processing Unit), a RAM (Random Access Memory), a ROM (Read-Only Memory), and an input / output circuit. As a functional unit, a search request acquisition unit 21 and a search result generation Unit 22, search result output unit 23, and characteristic storage unit 24.

  The search request acquisition unit 21 acquires the search request transmitted from the client device 10 and outputs the acquired search request to the search result generation unit 22.

  The search result generation unit 22 generates a search result by combining a plurality of search functions based on the search result acquired by the search request acquisition unit 21. The search result group acquisition unit 22a and the Web page information acquisition A unit 22b and an arranged search result acquisition unit 22c. The search term group acquisition unit 22a, the Web page information acquisition unit 22b, and the arranged search result acquisition unit 22c will be described later.

  The search result output unit 23 outputs the search result generated by the search result generation unit 22, that is, the arranged search result to the client device 10.

  The characteristic storage unit 24 stores characteristics of search word analysis algorithms 31A and 31B described later and characteristics of display order determination algorithms 61A and 61B described later.

The search term analysis algorithm 31 is an algorithm that analyzes a search term and generates a search term group. Examples of the characteristics of the search word analysis algorithm 31 include the following functional characteristics.
(1) Response to fluctuations in notation Response to fluctuations in katakana notation of foreign words: When the search word is “computer”, a search word group “computer computer” is generated. If the search term is “computer”, a search term group “computer” is generated.
Correspondence to Hiragana notation / Katakana notation / Kanji notation / Romaji notation / English notation: When the search word is “Tamago”, a search term group “Tamago egg egg egg” is generated. When the search term is “egg”, a search term group “egg egg egg egg tamago” can be generated. Similarly, when the search word is in katakana notation, kanji notation, or romaji notation, a search word group including other notations corresponding to the search word can be generated. With regard to Roman notation and English notation, it is also possible to generate a search term group corresponding to a difference between full-width / half-width characters and a difference between uppercase / lowercase characters.
Responding to fluctuations in sending kana: When the search term is “do”, a search term group “perform” is generated. If the search term is “perform”, a search term group “perform” is generated.
(2) Dealing with typing errors When the search term is “simulation”, a search term group called “simulation simulation” is generated. When the search term is “simulation”, a search term group “simulation simulation” is generated.
(3) Correspondence to synonyms When the search term is “sampling”, a search term group “sampling purchase” is generated. If the search term is “purchase”, a search term group “purchase trade-in” is generated.
(4) Correspondence to notation of play mixture peculiar to the net When the search word is “Itofuyu”, a search word group “End of Itofu” is generated. If the search term is “end”, a search term group “end thread winter” is generated.
The characteristic storage unit 24 includes which functional characteristics (1) to (4) each of the search word analysis algorithms 31A and 31B has, and which search word analysis unit 31 includes the search word analysis algorithms 31A and 31B. , 32 is stored.

The display order determination algorithm 61 is an algorithm for determining the display order of Web pages. Examples of the characteristics of the display order determination algorithm 61 include the following functional characteristics.
(11) A page ranking is generated based on the number of linked web pages, and the display order is mechanically determined based on the page ranking.
(12) The importance of the Web page is generated based on the category registration information of the Web page in addition to the number of linked pages of the Web page, and the display order is determined based on the importance.
(13) The importance of the web page is generated based on the last update date of the web page in addition to the number of linked pages of the web page, and the display order is determined so that the new web page is displayed at the top.
(14) The display order is determined so that the word-of-mouth site and the BBS site are displayed at the top.
(15) The importance of the web page is generated based on the browsing frequency of the web page, and the display order is determined so that the web page with the high browsing frequency is displayed at the top (the web page with the unknown browsing frequency is at the bottom) To be displayed).

<Search Word Analysis Device 30 and Search Word Group Acquisition Unit 22a>
The search word analysis device 30A stores a search word analysis algorithm 31A, analyzes the search word using the search word analysis algorithm 31A, generates a search word group as an analysis result, and generates the generated search word group. It outputs to the search word group acquisition part 22a. Here, the search word analysis algorithm 31A is an algorithm having the functional characteristics (1) to (3) described above.

  The search word analysis device 30B stores a search word analysis algorithm 31B different from the search word analysis algorithm 31A, analyzes the search word using the search word analysis algorithm 31B, and generates a search word group as an analysis result. Then, the generated search word group is output to the search word group acquiring unit 22a. Here, the search word analysis algorithm 31B is an algorithm having the functional characteristics (1) to (4) described above.

  The search word group acquisition unit 22a refers to the characteristics of the search word analysis algorithms 31A and 31B stored in the characteristic storage unit 24, and based on these characteristics and information (auxiliary information) included in the search request, The search word analysis algorithm 31 to be used is selected, and the search word is output to the search word analysis device 30 in which the selected search word analysis algorithm 31 is stored. The search word group acquisition unit 22a stores which search word analysis algorithm 31 is stored in which search word analysis device 30, and the search word analysis algorithm 31 to be used is stored based on the stored contents. A search word can be output to the search word analysis device 30. If the search word analysis algorithm 31 is specified in the search request, the search word group acquisition unit 22a does not refer to the characteristic storage unit 24 and stores the specified search word analysis algorithm 31 stored therein. A search word can be output to the word analysis device 30. The Web API (Web Application Programming Interface) of the search word analysis algorithms 31A and 31B is publicly disclosed in advance, and the search word group acquisition unit 22a can transmit and receive the search word and the search word group using the WebAPI publicly disclosed in advance. it can. The search word group acquisition unit 22a outputs the acquired search word group and a search request (auxiliary information) other than the search word to the Web page information acquisition unit 22b.

  More specifically, the search word group acquisition unit 22a stores the search word analysis algorithm 31 when information specifying the search word analysis algorithm 31 to be used for the search request is included in the search request. The search word may be output to the search word analysis device 30. In this case, when the user of the client device 10 operates the input unit 11 to input a search word, the user can input the search word analysis algorithm 31 to be used as auxiliary information. Alternatively, the user of the client device 10 can operate the input unit 11 in advance and input the search term analysis algorithm 31 to be used as a default. In these cases, the control unit 13 outputs a search term and information specifying the search term analysis algorithm 31 to be used to the search request acquisition unit 21 as a search request.

  In addition, when the search request includes a search history of the user, that is, a history indicating which search word analysis algorithm 31 the user has used, the search term group acquisition unit 22a is based on the search history. It may be configured to determine which search word analysis algorithm 31 is used and to output the search word to the search word analysis device 30 that stores the determined search word analysis algorithm 31. In this case, the search term group acquisition unit 22a can determine to use the search term analysis algorithm 31 used most recently by the user, and also determines to use the search term analysis algorithm 31 most frequently used by the user. You can also

  In addition, when the search request includes a browsing history of the user, that is, a history indicating which Web page the user has browsed, the search term group acquisition unit 22a selects which search term based on the browsing history. It may be configured to determine whether to use the analysis algorithm 31 and to output the search word to the search word analysis device 30 storing the determined search word analysis algorithm 31. In this case, the search word group acquisition unit 22a determines whether the user is working or private based on the browsing history. For example, when there are a lot of company HPs, dictionaries HP, etc. on the web page browsed within the most recent predetermined number of times in the browsing history, it can be determined that the user is in work when there are many BBS, blogs, etc. . When it is determined that the user is working, it is determined that the search word analysis algorithm 31A having the functional characteristics (1) to (3) described above is used, and when the user is determined to be private Can be determined by using the search word analysis algorithm 31B having the functional characteristics (1) to (4) described above.

  Further, the search word group acquisition unit 22a can generate an additional search word based on the browsing history of the user included in the search request, and output the generated additional search word to the search word analysis unit 30 together with the search word. .

  Here, a case where the search word group acquisition unit 22a generates an additional search request based on the browsing history will be described. The search term group acquisition unit 22a refers to the index database 52 based on the browsing history, and extracts keywords from the Web page browsed by the user. Subsequently, the search term group acquisition unit 22a determines the importance of the keyword based on the browsing time of the Web page included in the search request (the time during which the Web page has been browsed, which Web page has been browsed most recently) and the like. And the keyword having high importance is output to the search word analysis device 30 together with the search word as an additional search word. In this case, the search results (Web page information acquired by the Web page information acquisition unit 22b described later) are narrowed down by increasing the number of search terms, and a better search result is presented to the user. Will be able to.

<Crawling device 40, index generation device 50, and Web page information acquisition unit 22b>
The crawl device 40 includes a crawl unit 41 that periodically crawls the web page 100 in the Internet, and a web page storage unit 42 that stores the web page acquired by the crawl unit 41 as a cache.

  The index generation device 50 refers to the Web page stored in the Web page storage unit 42, generates an index for searching for the Web page, and the Web page of the Web page corresponding to the generated index And an index database 52 that stores information in association with each other. Here, the Web page information is a URL (Uniform Resource Locator) of the Web page including the search index. In the index database 52, as reference information for determining the display order of Web pages, the number of linked Web pages, category registration information, last update date, browsing frequency, type (word-of-mouth site, BBS site, etc.), etc. Such Web page reference information is stored in association with the Web page information. Here, the browsing frequency of the Web page can be acquired from a counter set on the Web page, access analysis, or the like.

  The Web page information acquisition unit 22b refers to the index database 52 based on the search term group output from the search term group acquisition unit 22a and a search request other than the search term, and includes a search index corresponding to the search term group. Web page information of the Web page is acquired, and the acquired Web page information is output to the display order acquisition unit 22c. The Web API of the index database 52 is published in advance, and the web page information acquisition unit 22b can receive the web page information using the web API published in advance. The Web page information acquisition unit 22b outputs a search request other than the acquired Web page and search word to the display order acquisition unit 22c and the trend analysis unit 25 and recommendation information generation unit 26 described later.

<Display Order Determination Device 60 and Arranged Search Result Acquisition Unit 22c>
The display order determination device 60A stores a display order determination algorithm 61A, determines the display order of Web page information using the display order determination algorithm 61A, and arranges the Web page information based on the display order. To generate an arrayed search result, and outputs the generated arrayed search result to the arrayed search result acquisition unit 22c. Here, the display order determination algorithm 61A is an algorithm having the characteristic (11) described above.

  The display order determination device 60B uses the display order determination algorithm 61B different from the display order determination algorithm 61A to determine the display order of the Web page information, and arranges the Web pages based on the display order. A result is generated, and the generated arranged search result is output to the arranged search result acquisition unit 22c. Here, the display order determination algorithm 61B is an algorithm having the characteristic (14) described above.

  The arranged search result acquisition unit 22c refers to the characteristics of the display order determination algorithms 61A and 61B stored in the characteristic storage unit 24, and uses the display based on these characteristics and information included in the search request. The order determination algorithm 61 is selected, and Web page information is output to the display order determination device 60 in which the selected display order determination algorithm 61 is stored. The arranged search result acquisition unit 22c stores which display order determination algorithm 61 is stored in which display order determination device 61, and the display order determination algorithm 61 to be used is stored based on the stored contents. Web page information can be output to the display order determination device 60. When the display order determination algorithm 61 is specified in the search request, the display order acquisition unit 22c does not refer to the characteristic storage unit 24 and the display order in which the specified display order determination algorithm 61 is stored. Web page information can be output to the determination device 60. The Web APIs of the display order determination algorithms 61A and 61B are disclosed in advance, and the display order acquisition unit 22c can transmit and receive Web page information and the arranged search results using the Web API disclosed in advance. The arranged search result acquisition unit 22 c outputs the acquired arranged search result to the search result output unit 23.

  More specifically, the arrayed search result acquisition unit 22c stores the display order determination algorithm 61 when information specifying the display order determination algorithm 61 to be used for the search request is included in the search request. The web page information may be output to the display order determination device 60. In this case, when the user of the client device 10 operates the input unit 11 to input a search term, the display order determination algorithm 61 to be used as auxiliary information can be input. Alternatively, the user of the client device 10 may input the display order determination algorithm 61 to be used as a default by operating the input unit 11 in advance. In these cases, the control unit 13 outputs a search term and information specifying the display order determination algorithm 61 to be used to the search request acquisition unit 21 as a search request.

  In addition, when the search request includes the search history of the user, that is, a history indicating which display order determination algorithm 61 the user has used, the arranged search result acquisition unit 22c is based on the search history. It is also possible to determine which display order determination algorithm 61 is to be used and output the Web page information to the display order determination device 60 that stores the determined display order determination algorithm 61. In this case, the arranged search result acquisition unit 22c can determine that the display order determination algorithm 61 used most recently by the user is used, and if the display order determination algorithm 61 used most frequently by the user is used. It can also be determined.

  In addition, when the search request includes the browsing history of the user, that is, the history indicating which Web page the user has browsed, the arranged search result acquisition unit 22c displays which display based on the browsing history. It may be configured to determine whether to use the order determination algorithm 61 and output the search word to the display order determination device 60 that stores the determined search word analysis algorithm 61. In this case, the arranged search result acquisition unit 22c determines whether the user is working or private based on the browsing history. When it is determined that the user is working, it is determined to use the display order determination algorithm 61A having the above-described functional characteristics of (11), and when it is determined that the user is private, The display order determination algorithm 61B having the functional characteristic (14) can be used.

<Trend analysis unit 25 and recommendation information generation unit 26>
The search device 20 further includes a trend analysis unit 25 and a recommendation information generation unit 26. The trend analysis unit 25 performs a trend analysis on the search word group based on the Web page information, and outputs the result of the trend analysis to the search result output unit 23. The recommendation information generation unit 26 generates recommendation information about a news article or a product related to a search term group based on the Web page information, and outputs the recommendation information to the search result output unit 23. The search result output unit 23 outputs the result of trend analysis and recommendation information to the client device 10 together with the arranged search result.

  More specifically, the trend analysis unit 25 refers to the Web page based on the Web page information stored in the index database 52, and determines the ranking of words (topics) used in the Web page at the end of the Web page. By analyzing together with the update date, the result (ranking) of the trend analysis is generated as a weekly ranking, a monthly ranking, or the like.

  Also, the recommendation information generation unit 26 extracts popular movies, books, products, and the like from the topics that are the analysis results of the trend analysis unit 25, and generates the extracted movies, books, products, etc. as recommendation information. Here, the recommendation information generation unit 26 analyzes the user's preference based on the user's search history included in the search request, and matches the analysis result with the extracted movie, book, product, etc. The structure which produces the recommendation information suitable for preference may be sufficient.

<Operation example>
Next, a search method by the search device 20 according to the embodiment of the present invention will be described. FIG. 2 is a flowchart for explaining a search method by the search device according to the embodiment of the present invention.

  First, the search request acquisition unit 21 acquires a search request from the client device 10 (step S1: search request acquisition step).

  Subsequently, the search term group acquisition unit 22a of the search result generation unit 22 outputs the search terms included in the search request to the search term analysis device 30 in which the search term analysis algorithm 31 is stored (step S2). In step S2, the search term group acquisition unit 22a selects a search term algorithm 31 suitable for analyzing the search term based on the search request and the characteristics of the search term analysis algorithms 31A and 31B stored in the characteristic storage unit 24. The determined search word is output to the search word analysis device 30 that stores the determined search word analysis algorithm 31. Receiving the input of the search word, the search word analyzing device 30 generates a search word group by analyzing the search word using the search word analysis algorithm 31, and sends the generated search word group to the search word group acquiring unit 22a. Output.

  Subsequently, the search word group acquisition unit 22a acquires the search word group output from the search word analysis device (step S3). The combination of steps S2 and S3 corresponds to the search word group acquisition step in the claims.

  Subsequently, the web page information acquisition unit 22b of the search result generation unit 22 refers to the index database 52 based on the search term group, and acquires the web page information including the search term group (step S4: Web page information). Acquisition step).

  Subsequently, the arranged search result acquisition unit 22c of the search result generation unit 22 outputs the Web page information to the display order determination device 60 (step S5). In step S5, the table-sequenced search result acquisition unit 22c is suitable for determining the display order of Web pages based on the search request and the characteristics of the display order determination algorithms 61A and 61B stored in the characteristic storage unit 24. The display order determination algorithm 61 is determined, and the Web page information is output to the display order determination device 60 that stores the determined display order determination algorithm 61. The display order determination device 60 that has received the input of the Web page information determines the display order using the display order determination algorithm 61, generates the arranged search result by arranging the Web pages based on the display order, The generated arranged search result is output to the arranged search result acquisition unit 22c.

  Subsequently, the arranged search result acquisition unit 22c acquires the arranged search result output from the display order determination device 60 (step S6). The combination of steps S5 and S6 corresponds to the arranged search result acquisition step in the claims.

  Subsequently, the search result output unit 23 outputs the arranged search results to the client device 10 (step S7: search result output step).

The search device 20 according to the embodiment of the present invention embodies a search service that appropriately combines a plurality of search functions such as a search word analysis algorithm 31, an index database 52, and a display order determination algorithm 61, and stores search results obtained by such search service as a client. Since it outputs to the apparatus 10, a favorable search result can be shown with respect to a user.
In addition, the search device 20 includes one search word analysis algorithm 31 and one display order determination algorithm 61 suitable for the search among the plurality of search word analysis algorithms 31A and 31B and the plurality of display order determination algorithms 61A and 61B. Since it is selected and used, a suitable search result can be generated and output to the client device 10.
In addition, since the search device 20 uses the index database 52 to generate and output the result of trend analysis and recommendation information, conventionally, the search device 20 displays the result of trend analysis and recommendation information for only a specific Web page for the entire Web. It can be generated as a target, and more suitable information can be output to the client device 10.

  Moreover, since the search device 20 generates an additional search word based on the browsing history of the user included in the search request and uses the generated additional search word for the search, the search device 20 preferably supports the input of the search request by the user. be able to.

  As mentioned above, although embodiment of this invention was described, this invention is not limited to the said embodiment, A design change is possible suitably in the range which does not deviate from the summary of this invention. For example, at least one of the search word analysis devices 30A and 30B, the crawl device 40, the index generation device 50, and the display order determination devices 60A and 60B may be configured integrally with the search device 20. The search device 20 also selects a translation algorithm for translating a search term and a translation result Web page in the same manner as the search term analysis algorithm 31 and the display order determination algorithm 61 described above. It can also be used to display search results.

  In addition, by adopting a configuration in which the user logs in to the search device 20 by operating the input unit 11 of the client device 10, auxiliary information such as a search history and a browsing history is stored in advance in the storage unit of the search device 20 as a user ID. The structure memorize | stored for every may be sufficient. In this case, the first partial search request including the auxiliary information such as the search history and the browsing history and the user ID is output from the control unit 13 of the client device 10 to the search device 20 before the search word. When the control unit 13 of the client device 10 outputs the second partial search request including the search word and the user ID to the search device 20, the search request acquisition unit 21 of the search device 20 By referring to the storage unit using the user ID included in the general search request, auxiliary information such as the search history and browsing history of the user is read, and a search request including the search word and the read auxiliary information is generated. And can be obtained.

  When the display order determination algorithm 61 determines the display order of Web pages based on the user's browsing frequency, the control unit 13 of the client device 10 uses the cookie or the like to check the user's browsing frequency. Can be stored, and a search request can be generated from the stored browsing frequency and the input search word and output to the search device 20. Moreover, the structure which outputs the 1st partial search request containing a user's browsing frequency and user ID to the search device 20 before a search word may be sufficient as the control part 13. FIG. In this case, when the control unit 13 of the client device 10 outputs the second partial search request including the search word and the user ID to the search device 20, the search request acquisition unit 21 of the search device 20 By referring to the storage unit using the user ID included in the partial search request, the browsing frequency of the user is read, and a search request including the search word and the read browsing frequency can be generated and acquired. it can.

  Further, a plurality of search word analysis devices 30 are integrally configured, and the search word group acquisition unit 22a integrally includes information specifying the search word analysis algorithm 31 to be used together with the search words. The search word analysis device 30 that is output to the analysis device 30 and is integrally configured selects a search word analysis algorithm 31 to be used from the plurality of stored search word analysis algorithms 31 and generates a search word group. The configuration used for Similarly, a plurality of display order determination devices 60 are integrally configured, and the arranged search result acquisition unit 22c is integrally configured with information specifying the display order determination algorithm 61 to be used together with the Web page information. The search word analysis device 60 that is output to the display order determination device 60 and configured integrally selects the search word analysis algorithm 61 to be used from the plurality of stored display order determination algorithms 61 and arranges the search results. It may be a configuration used to generate

DESCRIPTION OF SYMBOLS 1 Search system 10 Client apparatus 20 Search apparatus 21 Search request acquisition part 22 Search result generation part 22a Search word group acquisition part 22b Web page information acquisition part 22c Arranged search result acquisition part 23 Search result output part 24 Characteristic storage part 25 Trend generation Unit 26 Recommendation information generation unit 30, 30A, 30B Search term analysis device (search term analysis unit)
31, 31A, 31B Search term analysis algorithm 52 Index database 60, 60A, 60B Display order determining device (display order determining unit)
61, 61A, 61B Display order decision algorithm

Claims (7)

A search request acquisition unit for acquiring a search request from a client device;
A search result generation unit that generates a search result by combining a plurality of search algorithms and an index database based on the acquired search request;
A search result output unit for outputting the generated search result to the client device;
With
The index database is a database in which a search index and web page information related to a web page including the search index are associated with each other.
The plurality of search algorithms are arranged by a search word analysis algorithm for analyzing a search word included in the search request, and determining a display order of the Web page information and arranging the Web page information based on the display order A display order determination algorithm for generating search results,
The search result generation unit
A search for outputting a search word included in the search request to a search word analysis unit in which the search word analysis algorithm is stored, and acquiring a search word group as an analysis result by the search word analysis algorithm from the search word analysis unit A word group acquisition unit;
A web page information acquisition unit that refers to the index database based on the search term group and acquires web page information including the search term group;
Output the acquired Web page information to a display order determination unit in which the display order determination algorithm is stored, and obtain an arranged search result generated by the display order determination algorithm from the display order determination unit A result acquisition unit;
With
The search result output unit outputs the arranged search result to the client device. A characteristic storage unit for storing characteristics of a plurality of search word analysis algorithms and characteristics of a plurality of display order determination algorithms;
The search word group acquisition unit selects one search word analysis algorithm based on the search request and the characteristics of the plurality of search word analysis algorithms, and the search word analysis in which the selected search word analysis algorithm is stored Output the search request to the
The arranged search result acquisition unit selects one display order determination algorithm based on the search request and characteristics of the plurality of display order determination algorithms, and the display order in which the selected display order determination algorithm is stored. The search apparatus according to claim 1, wherein the Web page information is output to a determination unit. A trend analysis unit that performs a trend analysis on the search term based on the Web page information;
The search device according to claim 1, wherein the search result output unit outputs the result of the trend analysis to the client device together with the arranged search results. A recommendation information generation unit that generates recommendation information related to a news article or a product related to the search term based on the Web page information;
The search device according to claim 1, wherein the search result output unit outputs the recommendation information to the client device together with the arranged search results. The search request includes a user browsing history,
The said search term group acquisition part produces | generates an additional search term based on the said browsing history, and outputs the produced | generated additional search term to the said search term analysis part with the said search term. Search device. The search term group acquisition unit transmits and receives data to and from the search term analysis unit using a predefined Web API,
The web page acquisition unit transmits / receives data to / from the index database using a predefined web API,
The search apparatus according to claim 1, wherein the arranged search result acquisition unit transmits and receives data to and from the display order determination unit using a predefined Web API. A search request acquisition unit for acquiring a search request from a client device;
A search result generation unit that generates a search result by combining a plurality of search algorithms and an index database based on the acquired search request;
A search result output unit for outputting the generated search result to the client device;
A search method using a search device comprising:
The index database is a database in which a search index and web page information related to a web page including the search index are associated with each other.
The plurality of search algorithms are arranged by a search word analysis algorithm for analyzing a search word included in the search request, and determining a display order of the Web page information and arranging the Web page information based on the display order A display order determination algorithm for generating search results,
The search request acquisition unit acquires the search request from the client device, and
The search result generation unit outputs a search word included in the search request to a search word analysis unit in which the search word analysis algorithm is stored, and from the search word analysis unit as an analysis result by the search word analysis algorithm. A search term group acquisition step for acquiring a search term group;
The search result generation unit refers to the index database based on the search term group and acquires Web page information including the search term group; and
The search result generation unit outputs the acquired Web page information to the display order determination unit in which the display order determination algorithm is stored, and the arranged search generated by the display order determination algorithm from the display order determination unit An ordered search result acquisition step for acquiring results;
The search result output unit outputs the arrayed search results to the client device; and
The search method characterized by including.

Images