JP5133946B2 - Information search apparatus and information search method - Google Patents

Description

The present invention belongs to the field of information search using the Internet, and particularly relates to an information search apparatus and an information search method for providing useful information based on a query specified by a user.

Conventionally, in a general search service using the Internet, when a user makes a search request based on a query (a search condition such as a keyword), information on a web page (for example, a title, URL, Snippets) are provided as search results. At this time, information related to the query is provided together with the search result. For example, in a portal site that provides services (hereinafter referred to as properties) of various genres such as shopping, auctions, weather, news, etc. in addition to a search service, to guide to the properties in the portal site related to the query together with the search results Attempts have been made to provide information.
Patent Document 1 discloses an information search method capable of searching for information suitable for a user's purpose by specifying a query including a type keyword (genre keyword) and an action keyword (target keyword).

JP 2002-183203 A

  By the way, it can be said that the use value of the search service by the Internet is rich in information and novelty (high update frequency). In other words, users often use search services because they often want to obtain seasonal information according to current affairs and trends, and the search intent (what genre information they want to obtain) ) Can be considered to be almost the same. Therefore, if the search intention of the user can be grasped for each query, useful information corresponding to the search intention can be provided to the user who has searched by the same query.

However, in the conventional search service, since the genre to which a certain query belongs cannot be determined uniquely, it is difficult to provide useful information to the user in conjunction with the query.
For example, when a novel titled “ABC” is published and then converted into a TV drama or movie, if the query “ABC” is specified in the search service, at least book sales, TV, and movies are included in this query. Corresponds to the related genre. In such a case, it is difficult for the search service side to determine which genre information is useful for the user (whether it matches the search intention) unless a genre query is specified. In addition, the search intention of the majority of users by the query “ABC” is “book sales” immediately after the novel is published, “TV” immediately after being converted into a TV drama, and “movie” immediately after being converted into a movie. As such, it will change according to current events and trends.

On the other hand, the properties of the portal site include properties that can be searched within the site. In such a case, the excitement of the query can be grasped by analyzing the frequency of use of the query. For example, in the movie property, if the search frequency by the query “ABC” is high, it can be determined that there are many users who want to obtain information on the movie “ABC”.
However, some properties are not searchable within the site, and some properties are not frequently used for site search, so even if you record the query usage status (query log) for each property, simply compare them. It is not possible. In addition, users who perform site searches provided by properties are considered to be some users who have abundant internet knowledge and clearly recognize the information that they want to obtain.
Therefore, even if the query log for each property is analyzed, it cannot be said that the general user's search intention is reflected, so it is difficult to provide useful information to the user in conjunction with the query.

  The present invention provides a query analysis device, a query analysis method and program, and an information search device that can appropriately determine a user's search intention in order to provide useful information in conjunction with a query specified by the user. With the goal.

In order to achieve the above object, the invention described in claim 1 includes a site classification database in which domain information of sites published on the Internet is classified and registered in advance by genre,
A related page extracting means for calculating a selection frequency for each search result based on a click log for a query specified in the search service, and extracting a web page highly relevant to the query from the selection frequency;
Genre determining means for determining the genre of the query based on the domain information of each web page extracted by the related page extracting means and the genre associated with the domain information registered in the site classification database;
Search request accepting means for accepting a search request from a client terminal connected via the Internet;
Search execution means for searching for a site that matches the query included in the search request received by the search request receiving means;
A search result providing means for providing a search result acquired by the search execution means to the client terminal;
The search result providing unit generates information on a search result screen including information related to the genre of the query included in the search request and the search result based on the genre of the query determined by the genre determining unit. It is an information search device characterized by being provided .

According to a second aspect of the present invention, in the information search device according to the first aspect , the related page extracting unit acquires the click log at regular intervals and extracts a related page.

According to a third aspect of the present invention, in the information search device according to the first or second aspect, the related page extracting unit extracts a web page having the selection frequency equal to or higher than a threshold value.

According to a fourth aspect of the present invention, in the information search device according to any one of the first to third aspects, the genre determining unit sets each web page extracted by the related page extracting unit to each genre. The classified genre distribution is calculated, and the genre of the query is determined based on the distribution degree in the genre distribution.

According to a fifth aspect of the present invention, in the information search device according to any one of the first to fourth aspects, information related to the genre of the query is guided to a property corresponding to the genre of the query. It is characterized by being information .

According to a sixth aspect of the present invention, in the information search device according to any one of the first to fifth aspects, the search result providing means is based on the genre of the query determined by the genre determining means. The search results acquired by the search execution means are narrowed down or rearranged .

The invention according to claim 7 is an information search method in an information search apparatus including a site classification database in which domain information of sites published on the Internet is classified and registered in advance by genre.
Calculating a selection frequency for each search result based on a click log for a query specified in the search service;
Extracting a web page highly relevant to the query from the selection frequency;
A genre determination step of determining a genre of the query based on domain information of each extracted web page and a genre associated with domain information registered in the site classification database;
A search request receiving step for receiving a search request from a client terminal connected via the Internet;
A search execution step of searching for a site that matches a query included in the search request received in the search request reception step;
A search result providing step of providing the client terminal with the search result acquired in the search execution step,
In the search result providing step, based on the genre of the query determined in the genre determination step, information on a search result screen including information related to the genre of the query included in the search request and the search result is generated. It is characterized by providing .

  According to the present invention, it is possible to appropriately determine what information the user wants to obtain through a query, and thus useful information can be provided in conjunction with the query.

It is a figure shown about an example of the network structure to which the information retrieval apparatus which concerns on this invention is applied. It is a figure which shows an example of the hardware constitutions which implement | achieve the function as a query analysis apparatus of a search server. It is a figure which shows an example of a data structure of a site classification database. It is a figure which shows an example of a data structure of a query classification | category database. It is a figure which shows an example of the data structure of a search database. It is a figure which shows an example of a data structure of a click log database. It is a figure shown about the example of a display of a general search result screen. It is a flowchart shown about an example of a query analysis process. It is a flowchart shown about an example of a search process. It is a figure shown about the example of a display of the search result screen to which this invention was applied.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
FIG. 1 is a diagram showing an outline of a network configuration to which an information search apparatus according to the present invention is applied. In the present embodiment, the search server 1 constitutes an information search device according to the present invention.
As shown in FIG. 1, a client terminal 2 such as a personal computer 2a or a portable terminal 2b and a web server 3 that discloses various information are connected to the search server 1 via a communication network N, and data is mutually transmitted. Communication is possible. The communication network N is a telephone line network such as the Internet or a telecommunication carrier, a mobile phone communication network, or the like.

In FIG. 1, the search server 1 provides a general search service to the client terminal 2. Specifically, when a search request by a query is made in the client terminal 2, the search server 1 provides the client terminal 2 with a search result (for example, a web page provided by the web server 3) that matches the specified query. To do. When the user selects a desired search result at the client terminal 2, the web page is transitioned to.
The search server 1 has a function as a query analysis apparatus 1A according to the present invention and a function as a search engine 1B that executes a search by a query. By providing the search server 1 with a function as the query analysis device 1A, when the search engine 1B provides a search result to the client terminal 2, information useful for the user can be added.

  Here, the search service by the search server 1 is provided as one service of a portal site, for example. That is, the portal site provides services (properties) of various genres such as shopping, auction, weather, news, etc. in addition to the search service. In the present embodiment, the search server 1 provides a search result that matches the query in response to a search request by the query, and provides information for guiding to a property corresponding to the genre to which the query belongs. It has become.

FIG. 2 is a diagram illustrating an example of a hardware configuration that implements the function of the search server 1 as the query analysis device 1A.
As shown in FIG. 2, the query analysis apparatus 1 </ b> A includes a general control unit 11, a storage unit (auxiliary storage) 12, an input / output unit 13, a communication unit 14, a site classification database 15, a query classification database 16, and the like. Realized by computer. Each block is electrically connected by a bus line 20.

The control unit 11 includes a CPU (Central Processing Unit) 111 as an arithmetic / control device, a RAM (Random Access Memory) 112 and a ROM (Read Only Memory) 113 as a main storage device.
The ROM 113 stores a program called BIOS (Basic Input Output System) and basic setting data. The RAM 112 is loaded with programs such as an OS and various applications read from the storage unit 12.
The CPU 111 executes various processes in accordance with programs developed in the RAM 112 and controls each block.

The storage unit 12 includes, for example, a hard disk and stores an OS, various application programs (for example, a server program and a web application program), various data, and the like.
The input / output unit 13 constitutes a user interface for use by the administrator of the search server 1.
The communication unit 14 performs processing according to a communication protocol such as TCP / IP, and transmits / receives data to / from the client terminal 2 via the communication network N.

The site classification database 15 is referred to when determining the genre to which each web page belongs, here the genre to which the search result corresponding to the query belongs. The search result corresponding to the query is the result of the search executed in response to the search request by the query. This site classification database 15 is constructed in advance by a portal site operator, for example.
FIG. 3 is a diagram illustrating an example of a data configuration of the site classification database. As shown in FIG. 3, a genre, domain information of a website representing the genre, and useful information (recommended site) regarding the genre are registered in association with each other.
Here, the domain information is information for identifying a computer (web server) that publishes a website. Therefore, the URL includes the same domain information as long as the web pages make up the same website.
Further, the genre in the site classification database 15 is provided corresponding to the property provided by the portal site. That is, the genre “movie” shown in FIG. 3 corresponds to a property that provides information on a topical movie released in the theater, and “job search” corresponds to a property that provides information on job hunting, job change, recruitment, etc. “Comedy” corresponds to a property that distributes recommended videos of comedians, etc., and “Book sale” corresponds to the provision of information about published books and magazines and the property that sells books and magazines.

Referring to the genre “movie” in FIG. 3, “eiga.com”, “movies.co.jp”, “cinema.co.jp”,... Are registered as domain information of the representative site. A web page including such domain information in a URL, that is, a web page constituting a representative site is classified into a genre “movie”. In addition, “yah! Movie” (property provided by the portal site) is registered as a recommended site of the genre “movie”. When the client terminal 2 makes a search request based on a query belonging to the genre “movie”, the recommended site information is provided.
The representative site for each genre is, for example, a website that provides information related to the genre and has many page views, or a website provided by a major company (for example, the genre “ A movie distributor's website is selected for “movie”.

In FIG. 2, the query classification database 16 is referred to when determining the genre to which the query belongs. The query classification database 16 is updated at predetermined intervals by a query analysis process described later based on the click log accumulated in the search engine 1B.
FIG. 4 is a diagram illustrating an example of a data configuration of the query classification database 16. As shown in FIG. 4, a genre and its distribution degree are registered for each query. That is, the query classification database 16 indicates what genre information search is used for each query in the search service.

Referring to the query “ABC” in FIG. 4, the distribution is 20% for movies, 5% for comedy, and 70% for book sales. This distribution shows that 20% of users who access movie-related web pages, 5% of users who access comedy-related web pages, and 5% of users who access book sales-related web pages from the search result by the query “ABC”. It shows that it was 70%.
It can be determined that the genre having the highest distribution degree for each query is a genre representing the user's search intention according to the query. For example, for the query “ABC”, “book sale” indicates the user's search intention by the query “ABC”, that is, a search is performed with the query “ABC” in order to purchase (or consider purchasing) a book. This means that there were the most users.

  The function of the search server 1 as the search engine 1B is realized by a configuration substantially similar to the hardware configuration that implements the query analysis device 1A shown in FIG. Instead of the site classification database 15 and the query classification database 16 in FIG. 2, a search database (see FIG. 5) and a click log database (see FIG. 6) are provided. That is, the search engine 1B is realized by a general computer including a control unit 11, a storage unit (auxiliary storage) 12, an input / output unit 13, a communication unit 14, a search database, a click log database, and the like.

FIG. 5 is a diagram illustrating an example of a data structure of the search database.
The search database is referred to when searching for information in response to a search request by a query. In this search database, a plurality of web pages that match this query are registered for each query. For example, a search database is constructed by automatically collecting web pages that are opened by a data collection program that circulates on the Internet and registering the collected web pages.

As shown in FIG. 5, the search database is composed of, for example, an index part IDX and a database part DB.
In the index part IDX, a web ID is registered for each keyword. In FIG. 5A, web IDs “# 01, # 02, # 11,...” Are registered for the keyword “YAH”, and web IDs “# 21, # 03, #” are registered for the keyword “XYZ”. 211, ... "are registered.
For example, a web page title, URL, site description (extract of text, snippet), and the like are registered in the database unit DB for each web ID. Note that the data included in the database unit DB is not limited to this, and contents (for example, image files) provided to the client terminal 2 as search results are registered as appropriate. In FIG. 5B, the contents of the web ID “# 01” are the title “YAH! Japan”, the URL “http://www.yah.co.jp”, and “the largest portal site! The description is registered.

FIG. 6 is a diagram illustrating an example of the data configuration of the click log database.
In the click log database, for each query, the web page selected from the search result corresponding to the query and the click date and time are recorded and stored one by one. In FIG. 6, for example, the “URL1” web page is selected as “2009/06/06/10: 00: 00” from the search result by the query “ABC”, and the “URL2” web page is “2009 / “06/06/12: 30” is selected and the like are recorded. Using this click log database, the click rate (CTR, number of clicks / number of displays) for each web page can be calculated.

The functions of the search server 1 as the query analysis device 1A and the search engine 1B are realized by the above-described configuration. That is, the CPU 111 of the search server 1 as the search engine 1B executes server software to execute a search for a web page in response to a search request from a web browser activated on the client terminal 2, and the search result I will provide a. When a search result is selected in the client terminal 2, this click status is recorded in the click log.
For example, as shown in FIG. 7A, when “ABC” is entered in the search window and a search request by the query “ABC” is made, a matching web page is extracted by referring to the search database, and the search result is displayed. Provided to the client terminal 2. In FIG. 7A, as a search result by the query “ABC”, (1) a web page on which information on the drama “ABC” is disclosed, (2) a web page on which information on the novel “ABC” is disclosed, (3) A book sales web page is provided where the novel “ABC” can be purchased.
When a (3) book sales web page is selected on the search result screen shown in FIG. 7 (a), a transition is made to the book sales web page shown in FIG. 7 (b). At this time, a click log indicating that the book sales web page is clicked from the search result is recorded.

  Furthermore, in the present embodiment, the CPU 111 of the search server 1 as the query analysis device 1A analyzes the genre distribution of the query based on the click log generated by the search engine 1B, and determines the genre of the query. . Thereby, when a search request by a query is made, it is possible to provide useful information related to the query together with the search result.

FIG. 8 is a flowchart illustrating an example of query analysis processing in the search server 1. This query analysis process is nothing but the function as the query analysis apparatus 1A. This query analysis process is executed every predetermined period.
In step S101, the click log accumulated for each query is acquired from the click log database (see FIG. 6). Although omitted in FIG. 6, the click log database also records the cumulative number of search results.
In step S102, a click rate is calculated for each search result, that is, for each web page (URL) for each query. Specifically, the click rate is obtained by dividing the number of clicks for each web page recorded in the click log by the number of times displayed as the search result of the query. The higher the click rate, the more frequently the web page is selected by the user.

In step S103, a web page with a high click rate for each query, for example, a web page with a predetermined number of click rates or a web page with a click rate equal to or higher than a threshold is extracted. That is, since a web page with a high user selection frequency in the search result corresponding to the query is a web page reflecting the user's search intention according to the query , such a web page is extracted.
In step S104, domain information is specified from the URL of the web page extracted in step S103. For example, if the URL of the extracted web page is “http://www.eiga.com/roadshow/featured/ABC...”, The “eiga.com” portion is specified as domain information. For example, the domain information may be specified based on the position of “/” based on the description of the URL, or may be specified using a regular expression.

In step S105, the genre to which the domain specified in step S104 belongs is determined with reference to the site classification database 15. That is, the genre to which the web page extracted in step S103 belongs is determined based on the domain information. For example, when the identified domain information is “eiga.com”, according to the site classification database 15 shown in FIG. 3, the genre to which this domain belongs is “movie”. Therefore, the genre of the web page in which “eiga.com” is included in the URL as domain information is “movie”. If the domain information is the same even if the URL details are different, the URLs are classified into the same genre.
For all the web pages extracted in step S103, the genre to which these belong is determined. By the processing up to step S105, it is determined to which genre the web page (web page with a high click rate) reflecting the user's search intention by the query belongs.

In step S106, a genre distribution is calculated for each query. Specifically, the web pages belonging to each genre are counted, and the distribution degree (count value for each genre / count value for all genres) is calculated based thereon. At this time, the distribution degree may be calculated by weighting the click rate for each web page.
In step S107, the calculated genre distribution is reflected in the query classification database 16 and updated. Based on the genre distribution for each query registered in the query classification database, the genre to which the query belongs, that is, the user's search intention by the query can be appropriately grasped.
For example, when a novel titled “ABC” is published and then converted into a TV drama or movie, if the query “ABC” is specified in the search service, at least book sales, TV, and movies are included in this query. Corresponds to the related genre. Even in such a case, it is possible to appropriately grasp the search intention by the query “ABC”.
For example, immediately after being converted into a movie, since “movie” occupies a large proportion in the genre distribution by the query “ABC”, it is possible to recognize the user's search intention to search for information related to the movie of the query “ABC”. . That is, it is possible to cope with a case where the user's search intention by the query changes according to current affairs / fashion.

  As described above, the search server 1 as a query analysis apparatus according to the embodiment includes a site classification database (15) in which domain information of sites published on the Internet is classified and registered in advance by genre, and designated in the search service. Based on the click log for the query that has been made, the selection frequency (click rate) for each search result is calculated, and related page extraction means for extracting a web page highly relevant to the query from this selection frequency (step S102 in FIG. 8, S103), genre determining means for determining the genre of the query based on the domain information of each web page extracted by the related page extracting means and the genre associated with the domain information registered in the site classification database ( Steps S104 to S106) of FIG.

  Thus, based on the click log indicating the user's search intention (search tendency, which genre information he / she wants to obtain) based on the query, the genre highly relevant to the query, that is, what information the user obtains by the query It is possible to appropriately determine whether the user wants to do so, so that useful information can be provided in conjunction with the query. Moreover, since the genre is specified using the domain information, the construction of the site classification database 15 is easy (there is no need to register the URL for each web page), and the web page selected from the search result is The possibility of belonging to any genre increases. Therefore, an accurate query analysis based on the click log can be performed, and the genre of the query can be appropriately determined.

  Moreover, the related page extraction means with which the search server 1 is provided acquires a click log for every fixed period, and extracts a related page. That is, the query analysis process shown in FIG. 8 is updated at regular intervals. As a result, the genre of the query is determined based on the click log that changes from moment to moment. Therefore, even if the user's search intention by the query changes according to current events and trends, the genre of the query that is currently optimal Can be determined.

  Further, the related page extracting means included in the search server 1 extracts a web page whose selection frequency is equal to or higher than a threshold value (it is considered that the search intention appears). In other words, the web page is extracted and analyzed only when it is considered that the user's search intention by the query appears. Thereby, it is possible to prevent the determined query genre from being inappropriate.

  Furthermore, the genre determining means provided in the search server 1 calculates a genre distribution in which each web page extracted by the related page extracting means is classified into each genre, and based on the distribution degree in the genre distribution, the genre of the query Determine. Accordingly, it is possible to appropriately determine the genre of the query in consideration of the genre distribution. For example, when there are a plurality of genres having a prominent distribution degree, the plurality of genres can be determined as the genre of the query.

FIG. 9 is a flowchart illustrating an example of search processing in the search server 1. This search process is nothing but the function as the search engine 1B.
In step S201, it is determined whether or not there is a search request by a query from the client terminal 2. Specifically, a search service provided by the search server 1 (for example, a top page of a portal site) is accessed by a web browser activated on the client terminal 2, and a query is input to the search window to instruct a search. When an operation (click operation of the search button) is performed, a search request by a query is transmitted to the search server 1. In step S201, the presence / absence of this search request is determined. If it is determined that there is a search request from the client terminal 2, the process proceeds to step S202. If it is determined that there is no search request, the process ends.

In step S202, a query included in the search request is acquired.
In step S203, referring to the search database, a web page that matches the query is extracted as a search result. For example, when the query “YAH” is specified, according to the search database shown in FIG. 5, the web page corresponding to the web ID “# 01, # 02,...”, That is, “YAH! Japan” or “YAH mail”. Are extracted as web pages that match the query “YAH”.
In step S204, the genre of the query is acquired with reference to the query classification database 16. For example, when the query “ABC” is designated, according to the query classification database shown in FIG. 4, the genre “book sales” having the highest distribution is acquired.

In step S205, the provided information including the extracted search result is edited to generate a search result screen. At this time, information related to the genre of the query acquired in step S204 and useful to the user (for example, information for guiding to a property corresponding to the genre of the query) is provided on the search result screen. In the present embodiment, since the genre of the query is determined by the query analysis device 1A, such information can be provided.
For example, if the genre of the query “ABC” is “book sales”, information for guiding to the property of book sales provided by the portal site is read from the site classification database 15 and arranged in the search result screen. .

In step S206, the search result screen file is transmitted to the client terminal 2 that requested the search. On the web browser of the client terminal 2, a search result screen including a search result by the query and information for guiding to a property corresponding to the genre of the query is displayed (see FIG. 10).
In step S207, it is determined whether or not the user has performed an operation of selecting a search result on the search result screen. If it is determined that the search result is selected, the process proceeds to step S208. If it is determined that the search result is not selected, the process ends (or waits until it is selected).
In step S208, the selected search result and the date and time are recorded in the click log corresponding to the designated query.

  For example, as shown in FIG. 10, when a search request by the query “ABC” is made, advertisement information for guiding to a book sales property provided by the portal site as a recommended site is provided in addition to the web search result. . Note that the advertisement information shown in FIG. 10 is a search result of an in-site search that is searched by the query “ABC” in the recommended site specified in the site classification database 15. When this advertisement information is selected, the novel “ABC” is introduced in the book sales property, and a transition is made to a web page that can be easily purchased. This is extremely useful information for a user who is searching for information for purchasing the novel “ABC” by the query “ABC”.

As described above, the search server 1 as an information search device includes a query analysis device (1A) and search request reception means for receiving a search request from a client terminal (2) connected via the Internet (step S201 in FIG. 9). , S202), search execution means for searching for a site matching the query included in the search request received by the search request reception means (step S203), and the search result (web search result) acquired by the search execution means as the client Search result providing means (steps S204 to S206) provided to the terminal.
The search result providing means edits and provides the search result based on the genre of the query determined by the query analysis device. Specifically, information for guiding to the property corresponding to the genre of the query is added to the search result.

  As a result, useful information is effectively provided to the user in accordance with the search by the query, so that a search service excellent in convenience for the user is realized. On the other hand, portal sites that provide search services can guide potential customers by recommending their own sites (property) that are useful to users, which is effective in acquiring new customers and advertising effectiveness. Can also be expected.

As mentioned above, although the invention made by this inventor was concretely demonstrated based on embodiment, this invention is not limited to the said embodiment, It can change in the range which does not deviate from the summary.
For example, in the above embodiment, as an example of editing the search result, a case has been described in which information for guiding to a property corresponding to the query genre is added to the search result. However, the search result is narrowed down based on the query genre. Alternatively, rearrangement may be performed.

  The embodiment disclosed this time should be considered as illustrative in all points and not restrictive. The scope of the present invention is defined by the terms of the claims, rather than the description above, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.

1 Search server (query analysis device, information search device)
2 Client terminal 3 Web server N Communication network 11 Control unit (related page extracting means, genre determining means, search request receiving means, search executing means, search result providing means)
12 storage unit 13 input / output unit 14 communication unit 15 site classification database 16 query classification database 20 bus line

Claims (7)

A site classification database in which domain information of sites published on the Internet is classified and registered in advance by genre,
A related page extracting means for calculating a selection frequency for each search result based on a click log for a query specified in the search service, and extracting a web page highly relevant to the query from the selection frequency;
Genre determining means for determining the genre of the query based on the domain information of each web page extracted by the related page extracting means and the genre associated with the domain information registered in the site classification database;
Search request accepting means for accepting a search request from a client terminal connected via the Internet;
Search execution means for searching for a site that matches the query included in the search request received by the search request receiving means;
A search result providing means for providing a search result acquired by the search execution means to the client terminal;
The search result providing unit generates information on a search result screen including information related to the genre of the query included in the search request and the search result based on the genre of the query determined by the genre determining unit. An information retrieval device characterized by being provided. The information search apparatus according to claim 1, wherein the related page extracting unit acquires the click log at regular intervals and extracts a related page. The information search apparatus according to claim 1, wherein the related page extracting unit extracts a web page having the selection frequency equal to or higher than a threshold value. The genre determining unit calculates a genre distribution in which each web page extracted by the related page extracting unit is classified into each genre, and determines a genre of the query based on a distribution degree in the genre distribution. The information search device according to any one of claims 1 to 3. 5. The information search apparatus according to claim 1, wherein the information related to the genre of the query is information for guiding to a property corresponding to the genre of the query. The search result providing means narrows down or sorts the search results acquired by the search execution means based on the genre of the query determined by the genre determination means. The information search device according to any one of the above. An information search method in an information search apparatus having a site classification database in which domain information of sites published on the Internet is classified and registered in advance by genre,
Calculating a selection frequency for each search result based on a click log for a query specified in the search service;
Extracting a web page highly relevant to the query from the selection frequency;
A genre determination step of determining a genre of the query based on domain information of each extracted web page and a genre associated with domain information registered in the site classification database;
A search request receiving step for receiving a search request from a client terminal connected via the Internet;
A search execution step of searching for a site that matches a query included in the search request received in the search request reception step;
A search result providing step of providing the client terminal with the search result acquired in the search execution step,
In the search result providing step, based on the genre of the query determined in the genre determination step, information on a search result screen including information related to the genre of the query included in the search request and the search result is generated. An information search method characterized by being provided .

Images