JP2010117925A - Device and method for searching document data - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To make a page searched within a site as an object of search from the outside of the site. <P>SOLUTION: In the site internal search device 30, a query receiving part 31 receives a plurality of search words, a site search part 32 searches a site internal search page by inputting one search word of the plurality of search words and a special word to a search engine, and calculates a word score of the search word by using also a search result based on only the search word, and a site determination part 36 calculates a site score including the site internal search page based on a site search score given to the site internal search page by the search engine and the word score. A site internal search part 38 performs site internal search with a site with higher site score as an object by inputting the remaining search word to the site internal search page, and a search result integration part 40 creates display information by integrating the result of the site internal search. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an apparatus and method for retrieving document data. In particular, the present invention relates to an apparatus and method for searching document data based on a plurality of search terms.

In recent years, many websites (hereinafter simply referred to as “sites”) have provided web pages (hereinafter simply referred to as “pages”) having a search form. This can be done using a page (site search page).
Here, as a related art regarding site search, there is one that displays an identifier of a site search server together with a search result (see, for example, Patent Document 1). In this patent document 1, an identifier of a site server corresponding to an inputted search expression is searched, an identifier of an in-site search server attached to the site server is generated by a conversion program associated with this identifier, and this identifier is A search result corresponding to the input search expression is generated.

JP 2006-99279 A

  By the way, a search engine is generally used to search information on the Internet. This search engine includes a robot type search engine and a directory type search engine. A robotic search engine is a crawler that automatically accesses content on the Internet while following hyperlinks (links) and accumulates index information for the content. When a client requests a search, the search results are based on the index information. Is a search engine that returns A directory-type search engine is a search engine in which an administrator registers contents in an appropriate directory and returns a search result based on the registered information when a search request is received from a client.

Among these, the robot type search engine can collect information on a page that can be accessed by following a link, and can therefore be a target of a search according to a search request. However, since the pages searched by the above-described site search can be directly searched from the site search page, there are cases where links are not made from other sites, and such pages cannot be targeted for search. There was a problem.
Here, the invention of Patent Document 1 only displays the identifier of the in-site search server, and does not target the page searched by the in-site search from outside the site.

  An object of the present invention is to make document data (for example, a page) searched in a document data group (for example, a site) an object of search from outside the document data group.

  For this purpose, the present invention is an apparatus for searching document data based on a plurality of search terms, and a plurality of document data groups related to a first search term among the plurality of search terms. Each document data group determining a plurality of document data groups including search document data for searching in each document data group, and each document data group of the plurality of document data groups determined by the determination unit A search unit for searching a plurality of document data related to a second search word different from the first search word among the plurality of search words from each document data group using the search document data included in And an output unit that outputs information on a plurality of document data searched by the search unit.

Here, the determination unit inputs the first search word and a specific search word predetermined as a word for searching the search document data into the search engine, and adds the search result to the search result returned from the search engine. Based on this, a plurality of document data groups may be determined. The determination unit may determine a plurality of document data groups based on a predetermined property level of each document data group included in the search result. The determining unit inputs the first search term into the search engine, and determines a plurality of document data groups based on the presence or absence of a specific character string in the search result returned from the search engine. Good.
The output unit associates each document data group of the plurality of document data groups determined by the determination unit as information on the plurality of document data and a plurality of document data searched from each document data group by the search unit. Output display information for display. In that case, the retrieval unit retrieves a plurality of document data from each document data group by inputting the second search word to each search engine for retrieval in each document data group of the plurality of document data groups. The output unit may determine the layout of the display information based on each search result returned from each search engine and output the display information of the layout. The output unit may determine the layout based on the number of the plurality of document data included in each search result, or the degree of a predetermined property of each of the plurality of document data included in each search result. The layout may be determined based on the total result.

  The present invention is also an apparatus for searching document data based on a plurality of search terms, which is related to a first search term among the plurality of search terms and is a first for searching in a document data group. A first document data group including a plurality of search document data and a second search word that is related to a second search word different from the first search word among the plurality of search words, and is used for searching in the document data group. A second search unit that uses the first search document data from the determination unit that determines the second document data group including the search document data and the first document data group that is determined by the determination unit The first document data related to the word is searched, and the second document related to the first search word is searched using the second search document data from the second document data group determined by the determination unit. A search unit for searching for data, and the first document data and the second document data searched by the search unit; And an output unit for outputting information about the data, apparatus is also provided.

Here, the determination unit inputs the first search word and a specific search word predetermined as a word for searching the search document data for searching in the document data group to the search engine, The first document data group is determined based on the first search result returned from the search engine, the second search word and the specific search word are input to the search engine, and the first search data returned from the search engine is returned. The second document data group may be determined based on the second search result. The determination unit is configured to determine a predetermined property level of each document data group included in the first search result and a predetermined numerical value related to document data related to a specific search word among the document data related to the first search word. The first index obtained from the above, the degree of the predetermined property of each document data group included in the second search result, and the specific search word of the document data related to the second search word The first document data group and the second document data group may be determined based on a second index obtained from a predetermined numerical value regarding the document data to be processed. The determining unit inputs the first search term into the search engine, determines the first document data group based on the presence or absence of the specific character string in the first search result returned from the search engine, The second search term may be input to the search engine, and the second document data group may be determined based on the presence or absence of a specific character string in the second search result returned from the search engine.
The output unit associates the first document data group determined by the determination unit and the first document data searched by the search unit as information related to the first document data and the second document data. First display information for display and a second display for displaying the second document data group determined by the determination unit and the second document data searched by the search unit in association with each other It may be one that outputs information.

  Further, the present invention is an apparatus for searching document data based on a plurality of search terms, and by inputting a first search term and a second search term among the plurality of search terms to a search engine. , A plurality of document data, a first search word string including a first search word and a predetermined specific search word, and a second search word including a second search word and a specific search word The first search unit for searching a plurality of search document data for search in the document data group, the first search word string, and the second search Each of the word strings is input to the search engine, and the score given by the search engine to the plurality of search document data and the first search word are selected from the plurality of search document data searched by the first search unit. Search the search engine and search by the first search unit A first index obtained from the ratio of the plurality of search document data searched by the first search unit by inputting a first search word string of the plurality of document data into the search engine, and a plurality of The search engine gives the score given by the search engine and the second search word to the search engine and searches the second search word string among the plurality of document data searched by the first search unit. A first search for searching in the first document data group based on a second index obtained from the ratio of a plurality of search document data input to the engine and searched by the first search unit A first search document data determined by the determination unit determining the second search document data for searching in the second document data group and the second document data group, and the first search document data determined by the determination unit, Related to the second search term from one document data group The first document data is searched, and the second document data related to the first search word is searched from the second document data group using the second search document data determined by the determination unit. First search information for associating and displaying the second search unit, the first search document data determined by the determination unit, and the first document data searched by the second search unit, and And an output unit for outputting second display information for displaying the second search document data determined by the determination unit and the second document data searched by the second search unit in association with each other. An apparatus is also provided.

  Furthermore, the present invention is a method for searching for document data based on a plurality of search terms, wherein a plurality of document data groups related to a first search term among the plurality of search terms, each document data A step of determining a plurality of document data groups including search document data for search within each document data group, and search document data included in each document data group of the determined plurality of document data groups And a step of searching a plurality of document data related to a second search word different from the first search word among the plurality of search words from each document data group, and the plurality of searched document data A method comprising: outputting information.

  Furthermore, the present invention is a program for causing a computer to function as an apparatus for searching for document data based on a plurality of search terms, wherein the computer includes a plurality of search terms related to a first search term among the plurality of search terms. A document data group, each document data group determining a plurality of document data groups including search document data for searching in each document data group, and a plurality of document data determined by the determination unit A plurality of document data related to a second search word different from the first search word out of the plurality of search words from each document data group using the search document data included in each document data group of the group There is also provided a program that functions as a search unit that searches for information and an output unit that outputs information on a plurality of document data searched by the search unit.

  According to the present invention, document data searched in a document data group can be a search target from outside the document data group.

The best mode for carrying out the present invention (hereinafter referred to as “embodiment”) will be described in detail below with reference to the accompanying drawings.
First, a computer system to which this embodiment is applied will be described.
FIG. 1 is a diagram showing an example of the overall configuration of such a computer system.
As shown in the figure, in this computer system, clients 10a and 10b, a search engine 20, and an in-site search device 30 are connected via a network 80.

  The clients 10a and 10b are terminal devices such as PCs used by users. More specifically, when a web browser is installed and each user inputs a search query (hereinafter simply referred to as “query”) to the search engine 20 from the web browser, a web server (not shown) connected to the network 80 is illustrated. )) Is displayed as a search result. In the figure, the clients 10a and 10b are shown. However, when there is no need to distinguish them, they may be referred to as clients 10. Although only two clients 10 are shown in the figure, three or more clients 10 may be provided.

  The search engine 20 is a robot type that automatically accesses the contents on the Internet while the crawler follows the link, accumulates index information of the contents, and returns a search result based on the index information when a search request is received from the client. It is a search engine.

When a query is input to the search engine 20, the site search device 30 receives the query from the search engine 20, performs a site search, and returns the result of the site search to the search engine 20.
Although the search engine 20 and the site search device 30 are shown as separate devices here, they may be provided in one device.
The network 80 is a communication means used for transmitting / receiving information. Examples of the network 80 include the Internet and a LAN (Local Area Network).

The general operation of the computer system having such a configuration will be described.
FIG. 2 is a diagram showing such a schematic operation. In FIG. 1, the search engine 20 receives a query from the client 10 and returns the result of the site search by the site search device 30 to the client 10. Here, the site search device 30 directly receives the client 10 The query is received from the client, and the result of the site search is directly returned to the client 10.

  In the figure, it is assumed that the user has a desire to search for pages related to search words 1, 2, and 3. Therefore, first, the user inputs the search words 1, 2, and 3 into the site search device 30 using the client 10, and the site search device 30 determines a site for site search (1). That is, the in-site search device 30 first obtains the sites A, B, and C related to the search word 1 as the search result 300 by inserting the search word 1 into the search engine 20. Next, a site to be searched in the site is selected based on a certain standard from the sites included in the search result 300 acquired here. Here, it is assumed that the sites B and C among the sites A, B, and C included in the search result 300 are determined as sites to be searched in the site.

Next, the site search device 30 performs a site search (2) by inserting the search words 2 and 3 into the site search page at the site determined in (1). Here, since the sites B and C are determined as sites to be searched in the site, the search result 350b is obtained from the site B, and the search result 350c is obtained from the site C.
Thereafter, the in-site search device 30 integrates the search results 350b and 350c, executes re-ranking, and outputs the result to the client 10 to present it to the user (3).
Although only three search words are shown in the figure, three or more search words may be used.

Next, a functional configuration of the in-site search device 30 that performs such a general operation will be described.
FIG. 3 is a block diagram illustrating a functional configuration example of the site search device 30.
As shown in the figure, the in-site search device 30 includes a query reception unit 31, a site search unit 32, a query storage unit 33, a word score storage unit 34, and a site search result storage unit 35. In addition, a site determination unit 36, a site score storage unit 37, a site search unit 38, a site search result storage unit 39, and a search result integration unit 40 are provided.

The query receiving unit 31 receives a query including a plurality of search words from the search engine 20.
The site search unit 32 stores the query received by the query reception unit 31 in the query storage unit 33, and searches the page by inputting one search word included in the query into the search engine 20, and includes 1 included in the query. Two search words and special words (described later) are input to the search engine 20 to search the search page within the site. Based on the result of the search, the score of the search word (hereinafter referred to as “word score”) is stored in the word score storage unit 34, and information on the site search result (site search result) is stored in the site search result storage unit. 35. In this embodiment, a page is used as an example of document data, and an in-site search page is used as an example of search document data. Further, a site search unit 32 is provided as an example of a first search unit that searches for document data and search document data.

The query storage unit 33 stores the query received by the query receiving unit 31.
The word score storage unit 34 stores information (word score information) related to the word score calculated by the site search unit 32.
The site search result storage unit 35 stores the site search result acquired by the site search unit 32.

  The site determination unit 36 determines a site to be searched in the site based on the word score information stored in the word score storage unit 34 and the site search result stored in the site search result storage unit 35. The site score (hereinafter referred to as “site score”) is stored in the site score storage unit 37. In the present embodiment, a site is used as an example of a document data group. A site determination unit 36 is provided as an example of a determination unit that determines a document data group and a determination unit that determines search document data.

The site search unit 38 performs a site search for a site in which information related to the site score (site score information) is stored in the site score storage unit 37. At this time, among the search words included in the query stored in the query storage unit 33, a search word that is not used by the site search unit 32 is used to search for a site to be searched in the site. Then, information related to the result of the site search (site search result) is stored in the site search result storage unit 39. In the present embodiment, an in-site search unit 38 is provided as an example of a search unit that searches document data from a document data group and a second search unit.
The search result integration unit 40 integrates the search results based on the in-site search results stored in the in-site search result storage unit 39. In the present embodiment, a search result integration unit 40 is provided as an example of an output unit that outputs information about document data.

  Each of these functional units is realized by cooperation of software and hardware resources. Specifically, the CPU 90a (see FIG. 13) of the in-site search device 30 performs the function units of the query receiving unit 31, the site search unit 32, the site determination unit 36, the in-site search unit 38, and the search result integration unit 40. A program to be realized is read from the magnetic disk device 90g (see FIG. 13) into the main memory 90c (see FIG. 13) and processed. Further, the query storage unit 33, the word score storage unit 34, the site search result storage unit 35, the site score storage unit 37, and the in-site search result storage unit 39 are realized by, for example, a magnetic disk device 90g (see FIG. 13). .

Next, the operation of the present embodiment will be described in detail.
First, the user inputs a query in which a plurality of search words are combined under the “and” condition to the search engine 20. In this case, a check box for instructing to perform a site search may be provided on a page for inputting a query provided by the search engine 20. That is, if the check box is not checked, the search engine 20 returns a normal search result based on this query to the client 10, and if the check box is checked, passes the query to the in-site search device 30. Then, the result of the in-site search based on this query may be returned to the client 10.

Hereinafter, a process for obtaining the result of such in-site search will be described.
In the in-site search device 30, first, the query receiving unit 31 receives a query from the search engine 20 and passes it to the site searching unit 32. Then, the operation of the site search unit 32 starts.

FIG. 4 is a flowchart showing the operation of the site search unit 32 at this time.
First, the site search unit 32 acquires a query from the query reception unit 31 and stores it in the query storage unit 33 (step 301).
And the site search part 32 cuts out one search word from the some search words contained in the query memorize | stored in the query memory | storage part 33 (step 302).
Next, the site search unit 32 inputs the extracted search word into the search engine 20 (step 303). Then, the number of pages obtained as search results (number of search results) is acquired, and the number of search results is stored in the word score storage unit 34 (step 304).

  Further, the site search unit 32 inputs a query in which a special word is added to the search word cut out in step 302 to the search engine 20 and performs a search again (step 305). Here, the special word added to the search word is a keyword for searching for a site including a search page within the site. For example, “search”, “search form”, “search within site”, “search”, etc. It is done. However, the pages searched in this way include pages that include search words and special words but are not site search pages. Therefore, the site search unit 32 narrows down the pages obtained by the search in step 305 to pages including special words between the <form> tag and the </ form> tag (step 306). As the special word here, for example, “search”, “search form”, “search in site”, “search”, and the like can be considered. Then, the site search unit 32 acquires the number of site search pages (the number of site search pages) and stores the number of site search pages in the word score storage unit 34 (step 307).

Further, the site search unit 32 calculates a word score by the following formula and stores it in the word score storage unit 34 (step 308).
Word score = number of search pages in site / number of search results Note that this calculation method is to search the entire web with an abstract search word, and then perform a search within the site with a specific search word. Based on the assumption that the target page of
Alternatively, the word score may be the total number or the ratio of the total number of pages related to the special word, the score of the page related to the special word (PageRank (registered trademark), the degree of association with the search word, etc.), or the ratio. That is, any numerical value may be adopted as long as it is a predetermined numerical value relating to the page.
Then, the site search unit 32, for the search word cut out in step 302, the URL of the search page in the site and the score given to the search page in the site by the search engine 20 (hereinafter referred to as “site search score”). Are stored in the site search result storage unit 35 (step 309). Here, as the site search score, a score indicating the degree of association between the content of the page and the search word, a score indicating the importance of the page independent of the search word, such as PageRank (registered trademark), can be considered. Any score may be adopted as long as the score indicates the degree of the predetermined property of the page.

Thereafter, the site search unit 32 determines whether the search word still remains in the query after the search word is cut out in step 302 (step 310). If it is determined that the search word remains, steps 302 to 309 are repeated. If it is determined that no search word remains, the process is terminated.
In this operation example, in step 305, a query in which a special word is added to the search word is input to the search engine 20, and a search is performed. In step 306, the page obtained by this search is retrieved from the <form> tag. Narrow down to pages that contain special words up to the </ form> tag. However, in step 305, only the search word is input to the search engine 20 to perform a search, and the page obtained by this search includes a special word from the <form> tag to the </ form> tag in step 306. You may make it narrow down to a page.
Also, in this example of operation, in step 306, the pages were narrowed down to pages containing special words from the <form> tag to the </ form> tag, but the page was changed based on the presence or absence of other character strings such as the presence or absence of <form>. You may make it narrow down.

Here, the word score storage unit 34 in which information is stored in steps 304, 307, and 308 will be described.
FIG. 5 is a diagram illustrating an example of information stored in the word score storage unit 34.
As shown in the figure, the word score storage unit 34 stores, for each search word, the number of search results, the number of search pages in the site, and the word score (unit:%) obtained from the number of search pages in the site / number of search results. Is remembered.
For example, by inputting the search word “blog” to the search engine 20, 183,000 pages are hit, and by inputting a query in which the special word “search” is added to the search word “blog” to the search engine 20, It is indicated that 50000 pages are determined as search pages within the site, and as a result, the word score “27” is given to the search word “blog”.

The site search result storage unit 35 in which information is stored in step 309 will be described.
FIG. 6 is a diagram illustrating an example of information stored in the site search result storage unit 35.
As shown in the figure, the site search result storage unit 35 stores the URL of the in-site search page and the site search score in association with each search word.
The URL of the search page in the site is the URL obtained by performing a search by adding a special word to the search word in step 305 in FIG.
As described above, the site search score is a score given to the page by the search engine 20 by the search in step 305 in FIG.

In (a), the URL of the in-site search page obtained using the search word “blog” and the site search score are stored in the storage area for the search word “blog”. Here, it is assumed that the sites A, B, C, D, E,... Are obtained in order from the top using the search word “blog”.
In (b), the URL of the in-site search page obtained using the search word “travel” and the site search score are stored in the storage area for the search word “travel”. Here, it is assumed that the sites F, G,... Are obtained in order from the top using the search word “travel”.
In (c), the URL of the in-site search page obtained using the search word “Greece” and the site search score are stored in the storage area for the search word “Greece”. Here, it is assumed that the site H,... Is obtained using the search word “Greece”.
In this specification, each site is referred to as sites A, B, C, D, E, F, G, and H. In the drawings, each site is indicated by a specific URL.

  In this way, in the figure, the URL of the in-site search page obtained using each search word and the site search score are stored in the storage area corresponding to each search word. However, the form of storing the URL of the in-site search page obtained using each search word and the site search score is not limited to this. For example, the URL of the in-site search page obtained using each search word and the site search score may be stored in the same storage area. In that case, in addition to the URL of the site search page and the site search score, a search word is added as an item, and the search word used to obtain the URL of the site search page and the site search score is clearly indicated. Just keep it.

As described above, when information is stored in the word score storage unit 34 and the site search result storage unit 35, the site determination unit 36 determines a target site to be searched in the site.
FIG. 7 is a flowchart showing an operation example of the site determination unit 36 at this time.
First, the site determination unit 36 reads a line of records from the search result for a certain search word among the site search results stored in the site search result storage unit 35 (step 321). Further, the word score of the search word is acquired from the word score storage unit 34 (step 322). Then, the site score is calculated by multiplying the site search score included in the search result read in step 321 and the word score acquired in step 322 (step 323). That is, since a search result (site) based on each search word is given a site search score by the search engine 20, this is reused in the present embodiment, and in step 322, as shown in the following equation: The site score is recalculated by multiplying the word scores of each search word obtained.
Site score = word score of a search word used for searching the site × site search score assigned to the site by the search engine 20

  Thereby, the site determination part 36 memorize | stores the URL of a search page in a site, a site score, and a search word in the state arranged in the score order in the site score memory | storage part 37 (step 324). That is, when a record composed of such information is to be stored in the site score storage unit 37, if there is already a record, the record to be stored will include a site score larger than the site score included in the record. It is stored as a record following the record and before the record including the site score smaller than the site score included in the record. At that time, the records after the newly stored record are moved down one by one.

Thereafter, the site determination unit 36 determines whether or not N rows of records have been read from the site search result corresponding to the search word currently focused on (step 325). Here, N is a natural number designated by the user as the number of sites to be searched in the site.
As a result, if it is determined that N rows of records have not yet been read, steps 321 to 324 are repeated. On the other hand, if it is determined that N rows of records have been read, it is determined whether there are other unprocessed search words (step 326). If it is determined that there are other unprocessed search words, steps 321 to 325 are repeated. If it is determined that there are no other unprocessed search words, the process is terminated.

Here, the site score storage unit 37 in which information is stored in step 324 will be described.
FIG. 8 is a diagram illustrating an example of information stored in the site score storage unit 37.
As shown in the figure, the site score storage unit 37 stores the URL of the in-site search page, the site score, and the search word in association with each other.
Here, the site score information in the sites A, B, F, G, C, H, D, and E is illustrated in order from the top. That is, this is an example when N designated by the user as the number of sites to be searched in the site is “8”.
For example, in the case of the site A whose URL of the search page in the site is “http://aaablog.jp/”, the word score of the search word “blog” used for searching this site is as shown in FIG. The site search score given by the search engine 20 when searching this site is “20” as shown in FIG. Therefore, the site score “540” is stored by multiplying these scores.

As described above, when information is stored in the site score storage unit 37, the in-site search unit 38 performs in-site search for the site stored here.
FIG. 9 is a flowchart showing an operation example of the site search unit 38 at this time.
First, the site search unit 38 reads a line of records from the site score information stored in the site score storage unit 37 (step 341). Further, a search word other than the search word included in the read record is acquired from the query storage unit 33 (step 342). The site search is performed by inputting the search word acquired in step 342 to the site search page included in the record read in step 341 (step 343).

  At this time, the search word used for searching the site is not used in the search in the site. The search word for searching the search page in the site is a search word for the search in the site for a specific search. This is because it is considered unsuitable.

In addition, the site search is realized as follows, for example.
1. From the search page in the site, find the following <form> tag.
<form method = "get" name = "sample" action = "http://www.aaa.com/">
<input type = "radio" name = "site" value = "site"checked> Site search
<input type = "radio" name = "site" value = "web"> Web search
<input type = "text" name = "search_word">
<input type = "submit" value = "search">
</ form>
2. A search word used in the site search is substituted into an input corresponding to <type = "text"...>.
In this case, if necessary to improve accuracy, processing is performed in a knowledge base.
Also, for example, when selecting either in-site search or web search with a radio button, the check is switched to the side including the character string “site”.
3. The following Java script ("Java" is a registered trademark) is executed. This is an example in the case of performing a site search using the search word “travel Greece”.
document.sample.search_word.value = "Travel Greece";
document.sample.submit ();

  Thereby, the site search unit 38 acquires the result of the site search, and stores it in the site search result storage unit 39 together with the site score and the search word included in the record read in step 341 (step 344). Here, the results of the site search include the URL of the page including the URL of the page found in the site search (URL of the site search result) and the number of pages found in the site search (number of site search results). ) The number of search results in the site is acquired by a method depending on the site. For this purpose, knowledge-based text analysis or the like may be used. For example, since many sites include a character string “XX out of cases”, the character string can be obtained as a key.

Thereafter, the site search unit 38 determines whether or not N rows of records have been read from the site score information (step 345). In this case, N is a natural number designated by the user as the number of sites to be searched in the site.
As a result, if it is determined that N rows of records have not yet been read, steps 341 to 344 are repeated. On the other hand, if it is determined that N rows of records have been read, an in-site search score is calculated and stored in the in-site search result storage unit 39 (step 346). Here, the site search score is obtained by the following equation.
Site search score = number of site search results on the target site / number of site search results on all sites Note that the denominator “all sites” is the search word used to search the target site. It means all the sites that have been determined to be used for site search.
Alternatively, the in-site search score may be a total sum of relevance with the search word in each search result or a sum of scores independent of the search word like PageRank (registered trademark) in each search result. In other words, any score may be adopted as long as it is a score indicating a total result of a predetermined level of the page.

  Further, the site search unit 38 calculates a final score for determining the order of presentation of the search results to the final user, stores it in the site search result storage unit 39, and stores information in descending order of the final score. Rearrange (step 347). Here, the final score may be obtained by adding the above-mentioned in-site search score to the site score obtained in step 324 of FIG. As a simple example, the final score may be calculated by multiplying the site score and the in-site search score.

Here, the in-site search result storage unit 39 in which information is stored in steps 344, 346, and 347 will be described.
FIG. 10 is a diagram illustrating an example of information stored in the site search result storage unit 39.
As shown in the figure, the site search result storage unit 39 associates the URL of the site search result, the site score, the search word, the number of site search results, the site search score, and the final score. Is remembered.
Here, in the order from the top, information on the search results in the sites B, A, F, G, H, C, D, E is illustrated.

  For example, the following information is shown in the first line. That is, the URL of the search result page obtained by inputting “travel Greece” to the search page in the site B and performing the search in the site is shown as the URL of the search result in the site. Further, the site score “486” of the site B is shown as the site score, the search word “blog” used to search the site B is shown as the search word, and the number of search results in the site The number “50” of pages in which the URL is described in the search result page is shown. Furthermore, as the site search score, the ratio of the number of search results in the site B to the number of search results in the site in the site searched by the search word “blog” is shown. In this example, it is “0.50 (= 50 / (50 + 40 + 5 + 3 + 2))”. Then, “243 (= 486 × 0.50)” is shown as the final score.

  In the second line, the following information is shown. That is, the URL of the search result page obtained by inputting “travel Greece” to the search page of the site A and performing the search in the site is shown as the URL of the search result of the site. Further, the site score “540” of the site A is shown as the site score, the search word “blog” used to search the site B is shown as the search word, and the number of search results in the site As shown, the number “40” of pages in which the URL is described in the search result page is shown. Further, as the site search score, the ratio of the number of search results in the site A to the number of search results in the site in the site searched by the search word “blog” is shown. In this example, “0.40 (= 40 / (50 + 40 + 5 + 3 + 2))”. Then, “216 (= 540 × 0.40)” is shown as the final score.

  As shown in FIG. 8, among the sites obtained using the search word “blog”, the site score of the site A was “540”, and the site score of the site B was “486”. Based only on this result, a score should be given so that site A is displayed above. However, when the site A and the site B are searched by the site search query “travel Greece”, 40 search results from the site A and 50 search results from the site B are obtained. Therefore, the final score is higher for the site B so that the site B is displayed higher than the site A in response to the search result in the site.

As described above, when information is stored in the in-site search result storage unit 39, the search result integration unit 40 generates information on the in-site search result that the search engine 20 includes in the search result.
FIG. 11 is a flowchart showing an operation example of the search result integration unit 40 at this time.
First, the search result integration unit 40 reads a one-line record from the site search results stored in the site search result storage unit 39 (step 361). Then, a frame in which the URL of the search result in the site included in the read record is set is generated (step 362). For example, the frame may be generated by setting the URL of the search result in the site in the src attribute of the <frame> tag. Further, the frame height corresponding to the number of search results in the site included in the read record is calculated by a predetermined calculation formula, and the calculated frame height is also set as the accompanying information of the frame (step 363). Here, as the predetermined calculation formula, for example, a calculation formula for multiplying the number of search results in the site by a predetermined height as an appropriate height for displaying one search result can be considered. Further, the frame height obtained in this way may be set in the rows attribute of the <frameset> tag, for example.

Thereafter, the search result integration unit 40 determines whether or not N rows of records have been read from the site search results (step 364). In this case, N is a natural number designated by the user as the number of sites to be searched in the site.
As a result, if it is determined that N rows of records have not yet been read, steps 361 to 363 are repeated. On the other hand, if it is determined that N rows of records have been read, the search result integration unit 40 passes the set of frames (frame set) in which information is set in steps 362 to 363 to the search engine 20 (step 365). .

As a result, the search engine 20 generates a page in which the frame set is inserted into the page displaying the search result, and transmits the page to the client 10.
Then, the search result is displayed on the client 10.
FIG. 12 is an example of the search result displayed on the client 10 at this time.
In this search result, a frame 401 including an in-site search result in “aaa blog” as site A and a frame 402 including an in-site search result in “bbb blog” as site B are displayed.
Here, the search result in the site is displayed in the frame corresponding to the site. However, this is merely an example, and other display forms may be adopted. For example, it is possible to simply display the in-site search result without clearly indicating which site the in-site search result is.

The present embodiment has been described above.
By the way, in this embodiment, the site does not have to be a group of pages under a specific domain. More generally, it may be regarded as a document data group composed of a plurality of document data regardless of the domain or the like.

Further, in the present embodiment, the site score shown in FIG. 8 takes into account the word score of the search word used for searching the site, but is not necessarily limited to such a form. . For example, the site search score assigned to each site by the search engine 20 may be directly adopted as the site score without adding the word score. In that case, information on which search word is a high-level search word such as “blog” among a plurality of search words specified by the user is given separately, and the target of the search in the site is performed. When searching for a site, it is preferable to use a search word with a high level of abstraction.
Furthermore, in the present embodiment, the site search is performed using a search word different from the search word used to determine the site to be searched in the site. The search word to be used may include the search word used to determine the site to be searched in the site.

As described above, in this embodiment, when a plurality of search words are input to the search engine 20, the in-site search device 30 determines a site for performing the in-site search, and the site is automatically targeted for that site. Search inside. As a result, it is possible to search the site where the search form is installed, and to search data that cannot be searched by the conventional robot type search (data obtained only by searching from a specific server). Became.
In addition, the search results are comprehensively displayed in consideration of the site score and the search score in the site. As a result, the convenience in determining which site the information to be searched exists in is improved.
Furthermore, the site score is calculated in consideration of the word score calculated using the web-based statistical information. As a result, the user does not need to be aware of which search word should be used to search the target site to be searched in the site among the plurality of search words specified by the user.

The present embodiment can be applied to the following cases.
(Case 1)
When it is desired to check the product information, the manufacturer's product information and the like must be searched once again via the manufacturer's site. However, according to the present embodiment, a search word can be automatically input to a search form installed on a site in the manufacturer and displayed on the first search list.
(Case 2)
When you want to know the characteristics of the information managed by the site If the information is operated on various sites or mailing lists, the site search results are displayed in the initial search list, allowing you to quickly You can know the site that manages the features and good information.

  Finally, a hardware configuration of a computer suitable for applying this embodiment will be described. FIG. 13 is a diagram showing an example of the hardware configuration of such a computer. As shown in the figure, the computer includes a CPU (Central Processing Unit) 90a which is a calculation means, a main memory 90c connected to the CPU 90a via an M / B (motherboard) chip set 90b, and an M / B chip set 90b. And a display mechanism 90d connected to the CPU 90a. Further, a network interface 90f, a magnetic disk device (HDD) 90g, an audio mechanism 90h, a keyboard / mouse 90i, and a flexible disk drive 90j are connected to the M / B chip set 90b via a bridge circuit 90e. Has been.

  In FIG. 13, each component is connected via a bus. For example, the CPU 90a and the M / B chip set 90b, and the M / B chip set 90b and the main memory 90c are connected via a CPU bus. Further, the M / B chipset 90b and the display mechanism 90d may be connected via an AGP (Accelerated Graphics Port), but if the display mechanism 90d includes a PCI Express compatible video card, the M / B The chip set 90b and this video card are connected via a PCI Express (PCIe) bus. When connecting to the bridge circuit 90e, for example, PCI Express can be used for the network interface 90f. For the magnetic disk device 90g, for example, serial ATA (AT Attachment), parallel transfer ATA, or PCI (Peripheral Components Interconnect) can be used. Furthermore, USB (Universal Serial Bus) can be used for the keyboard / mouse 90i and the flexible disk drive 90j.

  Here, the present invention may be realized entirely by hardware or entirely by software. It can also be realized by both hardware and software. The present invention can be realized as a computer, a data processing system, and a computer program. This computer program may be stored and provided on a computer readable medium. Here, the medium may be an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system (apparatus or equipment), or a propagation medium. Examples of computer-readable media include semiconductors, solid state storage devices, magnetic tape, removable computer diskettes, random access memory (RAM), read-only memory (ROM), rigid magnetic disks, and optical disks. The Current examples of optical disks include compact disk-read only memory (CD-ROM), compact disk-read / write (CD-R / W) and DVD.

  As mentioned above, although this invention was demonstrated using embodiment, the technical scope of this invention is not limited to the said embodiment. It will be apparent to those skilled in the art that various modifications and alternative embodiments can be made without departing from the spirit and scope of the invention.

It is the figure which showed the whole structure of the computer system in embodiment of this invention. It is a figure for demonstrating schematic operation | movement of embodiment of this invention. It is the figure which showed the function structural example of the site search apparatus in embodiment of this invention. It is the flowchart which showed the operation example of the site search part in the site search device of embodiment of this invention. It is the figure which showed an example of the memory content of the word score memory | storage part in the site search apparatus of embodiment of this invention. It is the figure which showed an example of the memory content of the site search result memory | storage part in the site search apparatus of embodiment of this invention. It is the flowchart which showed the operation example of the site determination part in the site search device of embodiment of this invention. It is the figure which showed an example of the memory content of the site score memory | storage part in the site search apparatus of embodiment of this invention. It is the flowchart which showed the operation example of the site search part in the site search apparatus of embodiment of this invention. It is the figure which showed an example of the memory content of the site search result memory | storage part in the site search apparatus of embodiment of this invention. It is the flowchart which showed the operation example of the search result integration part in the site search device of embodiment of this invention. It is the figure which showed the example of a display of the search result in embodiment of this invention. It is the figure which showed the hardware constitutions of the computer which can apply embodiment of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Client, 20 ... Search engine, 30 ... Site search device, 31 ... Query reception part, 32 ... Site search part, 33 ... Query storage part, 34 ... Word score storage part, 35 ... Site search result storage part, 36 ... Site determination unit, 37 ... Site score storage unit, 38 ... Site search unit, 39 ... Site search result storage unit, 40 ... Search result integration unit

Claims (16)

An apparatus for searching document data based on a plurality of search terms,
A plurality of document data groups related to a first search word among the plurality of search words, wherein each document data group includes search document data for search in each document data group. A determination unit for determining a data group;
Using the search document data included in each document data group of the plurality of document data groups determined by the determination unit, the first search of the plurality of search terms from each document data group A search unit for searching for a plurality of document data related to a second search word different from the word;
And an output unit that outputs information on the plurality of document data searched by the search unit.   The determination unit inputs the first search word and a specific search word predetermined as a word for searching the search document data into a search engine, and the search result returned from the search engine The apparatus according to claim 1, wherein the plurality of document data groups are determined based on the document data.   The apparatus according to claim 2, wherein the determination unit determines the plurality of document data groups based on a predetermined property level of each document data group included in the search result.   The determination unit inputs the first search term into a search engine, and determines the plurality of document data groups based on presence or absence of a specific character string in a search result returned from the search engine. 1 device.   The output unit includes, as information on the plurality of document data, each document data group of the plurality of document data groups determined by the determination unit, and the plurality of document data groups searched from the document data group by the search unit. The apparatus according to claim 1, wherein the display information for displaying in association with document data is output. The search unit inserts the second search word into each search engine for searching in each document data group of the plurality of document data groups, so that the plurality of document data are extracted from each document data group. Search for
The apparatus according to claim 5, wherein the output unit determines a layout of the display information based on each search result returned from each search engine, and outputs the display information of the layout.   The apparatus according to claim 6, wherein the output unit determines the layout based on the number of the plurality of document data included in each search result.   The apparatus according to claim 6, wherein the output unit determines the layout based on a total result of a degree of a predetermined property of each of the plurality of document data included in each search result. An apparatus for searching document data based on a plurality of search terms,
A first document data group that is related to a first search word among the plurality of search words and includes first search document data for searching in the document data group; and among the plurality of search words A determination unit that determines a second document data group that includes a second search document data for searching in the document data group, related to a second search word different from the first search word ,
First document data related to the second search word is searched from the first document data group determined by the determination unit using the first search document data, and the determination unit A search unit for searching for second document data related to the first search word from the determined second document data group using the second search document data;
An apparatus comprising: an output unit that outputs information related to the first document data and the second document data searched by the search unit.   The determination unit inputs the first search word and a specific search word predetermined as a word for searching the search document data for searching in the document data group to the search engine, Based on the first search result returned from the search engine, the first document data group is determined, the second search word and the specific search word are input to the search engine, and the search engine The apparatus according to claim 9, wherein the second document data group is determined based on the second search result returned from.   The determination unit includes a degree of a predetermined property of each document data group included in the first search result, and document data related to the specific search word among the document data related to the first search word. A first index obtained from a predetermined numerical value regarding the degree of a predetermined property of each document data group included in the second search result, and document data related to the second search word 11. The apparatus according to claim 10, wherein the first document data group and the second document data group are determined based on a second index obtained from a predetermined numerical value related to document data related to the specific search term. .   The determination unit inputs the first search term into a search engine and determines the first document data group based on the presence or absence of a specific character string in the first search result returned from the search engine. The second search term is input to the search engine, and the second document data group is determined based on the presence or absence of the specific character string in the second search result returned from the search engine. The apparatus of claim 9.   The output unit includes the first document data group determined by the determination unit as the information related to the first document data and the second document data, and the first document data searched by the search unit. First display information for displaying the document data in association with each other, the second document data group determined by the determination unit, and the second document data searched by the search unit 10. The apparatus according to claim 9, wherein second display information for display is output. An apparatus for searching document data based on a plurality of search terms,
A plurality of document data is searched by inputting a first search word and a second search word of the plurality of search words to a search engine, respectively, and the first search word and a specific predetermined A first search word string including a search word and a second search word string including the second search word and the specific search word are respectively input to the search engine, so that the document data group includes A first search unit for searching a plurality of search document data for searching
The first search word string and the second search word string are respectively input to the search engine, and the plurality of search document data are searched from the plurality of search document data searched by the first search unit. The score given by the search engine to the document data, and the first search word string of the plurality of document data searched by the first search unit by inputting the first search word into the search engine. Is input to the search engine, and the search engine gives the first index obtained from the ratio of the plurality of search document data searched by the first search unit, and the plurality of search document data And the second search word string among a plurality of document data searched by the first search unit by inputting the second search word and the second search word into the search engine. The first search Based on the second index obtained from the ratio of the plurality of search document data searched by the first search document data for search in the first document data group, and the second document A determination unit for determining second search document data for search in the data group;
Using the first search document data determined by the determination unit, first document data related to the second search word is searched from the first document data group, and the determination unit A second search unit for searching the second document data group related to the first search word from the second document data group using the determined second search document data;
First display information for displaying the first search document data determined by the determination unit and the first document data searched by the second search unit in association with each other; and An output unit that outputs second display information for displaying the second search document data determined by the determination unit in association with the second document data searched by the second search unit And a device. A method for searching document data based on a plurality of search terms,
A plurality of document data groups related to a first search word among the plurality of search words, wherein each document data group includes search document data for search in each document data group. Determining a data set;
Using the search document data included in each document data group of the determined plurality of document data groups, the first search word out of the plurality of search words is different from each document data group. Retrieving a plurality of document data related to the second search term;
Outputting information relating to the retrieved plurality of document data. A program that causes a computer to function as a device that retrieves document data based on a plurality of search terms,
The computer,
A plurality of document data groups related to a first search word among the plurality of search words, wherein each document data group includes search document data for search in each document data group. A determination unit for determining a data group;
Using the search document data included in each document data group of the plurality of document data groups determined by the determination unit, the first search of the plurality of search terms from each document data group A search unit for searching for a plurality of document data related to a second search word different from the word;
A program that functions as an output unit that outputs information on the plurality of document data searched by the search unit.

Images