JP2008181186A - How to find the relevance between keywords and sites using query logs - Google Patents

Abstract

To provide a new method for obtaining a relationship between a search keyword and a site using a query log.
When a server located between a user terminal and a search server performs a search using a keyword from the user terminal, a unique search result page is generated and displayed on the user terminal. The server includes query log storage means for storing a query log in a user search result, the session ID when the user performs a search, the URL of the Web page clicked by the user from the search result page, the Web page The date and time when the link is clicked, the step of accumulating the keyword in the query log storage means, the step of counting the number of URL clicks for each keyword using the query log storage means, and the query log storage means in advance Extracting a query log including keywords in the inputted predetermined keyword list.
[Selection] Figure 1

Description

  The present invention relates to a method for obtaining a degree of association between a keyword and a site using a query log. More specifically, the present invention relates to a method, a server, and a program for obtaining a degree of association between a keyword and a site using a query log including a click log in a keyword search result.

  Today, it is now possible to search for information that anyone wants at any time from a vast amount of information using the Internet. In many cases, a user who seeks certain information inputs a keyword representing the feature of the information at various search sites on the Internet and performs a query. By performing such a query, the search engine of the search site searches the Web site on the Internet, and as a result, the search result page is displayed to the user. On this search result page, links to Web sites describing various information are displayed in accordance with the priority order determined by the search engine of the search site.

  However, the information on the linked Web site obtained from the search result page does not necessarily include the information requested by the user, and there are some that are not related to the input keyword. In the case where there is no information to be requested for the link destination clicked first on the search result page, in many cases, the user sequentially follows the link destination page of the next display order. For this reason, it is desirable that Web sites that are more closely related to keywords are displayed on the search result page. On the other hand, there are many sites that contain information that is inappropriate or “harmful” for certain users such as adults, violence, grotesques, discriminatory words, and so on. It is also necessary to prevent it from being displayed on the user search results page.

Therefore, there are many methods for filtering such “harmful” sites (inspecting the data content to determine whether it can be passed). For example, in Patent Document 1, a keyword file registered in advance is read, a search based on the keyword is automatically performed periodically, and a “bad URL (Universal Resource Allocator)” extracted from the search result information is stored in the bad URL database. A management server that registers and determines whether or not to send transmission request information from a client device to a Web server is disclosed.
JP 2004-46739 A

  However, it is not specifically described how the “bad URL” is extracted based on the keyword in the management server of Patent Document 1. Today there are a huge number of what are called “poisonous” sites, but there are also many “healthy” sites that contain “inappropriate” keywords (for example, current news, reviews, commentary, etc.) site). For this reason, there is a limit to the automatic extraction of “harmful” sites, and in the current situation, manual extraction methods such as specialists are widely used. On the other hand, there is a method of automatically filtering by a machine-learning method, but there are problems such as a large amount of calculation and troublesome training data creation for learning. Also, this method cannot be applied when there are not enough words at the entrance of the “harmful” site and the page is composed solely of images.

  In view of the above problems, the present invention provides a new method for empirically obtaining the degree of relevance between a keyword used for a search and a Web site using a log of search queries actually performed by a large number of users. The purpose is to do.

  The present invention provides the following solutions.

(1) A method in which a server that can be connected to a plurality of user terminals via a communication network obtains the degree of association between a keyword and a website,
The server generates a search result page when the user of the terminal performs a search by a keyword and displays the search result page on the user's terminal, and a click log from the search result page of the user. A query log storage means for storing,
In the server,
The query log includes the session ID when the user performs the search, the URL of the Web page clicked by the user from the search result page, the access date and time when the link to the Web page is clicked, and the keyword. Accumulating in storage means;
Using the query log storage means to count the number of clicks for the URL for each keyword;
Extracting a query log including each keyword in a predetermined keyword list inputted in advance from the query log storage means;
Including methods.

  According to such a configuration, this server generates a unique search result page (from the search result obtained by the search engine) and displays it on the user's terminal when the user performs a search by keyword ( Specifically, the URL of the search result page is transmitted. When the user clicks the link destination displayed on the search result page, the session ID when the search is performed, the URL of the clicked link destination, the access date and time when the click is performed, and the keyword (keyword used for the search) Are stored in a query log storage means (for example, a query log database described later) connected to the server. Then, using this query log storage means, the number of URL clicks actually clicked by the user from the search result page is counted for each keyword at the time of search. Then, a query log including each keyword (also referred to as a seed query) in a predetermined keyword list (for example, a keyword list for extracting harmful sites) set in advance is extracted from the query log storage means. .

  As a result, it is possible to accumulate a large number of URLs actually clicked by a user in a storage means such as a query log database from a search result page which is a result of a user searching using various keywords. URLs that match a predetermined keyword list can be extracted from the clicked URL set, and this extracted keyword and URL set can be used to collect and verify sites.

(2) The method according to (1), further including a step of displaying the URLs on a site verifier terminal connected to the server in the descending order of the number of clicks of the URLs included in the extracted query log. Method.

  According to such a configuration, a site verifier (a person who collects specific types of Web sites and verifies whether or not those sites are actually in that type) can be used for parental control, for example. When trying to collect a list of harmful sites, etc., the site verifier's verification efficiency can be improved by preferentially displaying the sites where the number of clicks by the user is high (that is, the impact of the harmfulness is actually high). Can be increased.

(3) From the access date and time for each URL included in the extracted query log, the stay time in the Web page clicked by the user is calculated, and for the click URLs with the same frequency, the stay time in the descending order of the stay time. The method according to (2), further comprising displaying a URL on a site verifier terminal connected to the server.

  According to such a configuration, even if there is a URL with the same number of clicks by acquiring the stay time of the Web page of the URL from the access date and time when the user clicked on the specific URL by some method, A long Web page is considered to have more information requested by the user, that is, the search keyword entered by the user and the site having the URL have high relevance, and the URL is preferentially displayed for verification. Efficiency can be increased.

(4) From the access date and time for each URL included in the extracted query log, the stay time on the Web page clicked by the user is calculated, and the keyword is obtained by multiplying the weight obtained from the stay time and the number of clicks. The method according to (1), further comprising a step of calculating a degree of association of a site and displaying the URL on a terminal of a site verifier connected to the server in descending order of the degree of association.

  According to such a configuration, the number of clicks is multiplied by the weight (weight) obtained from the stay time, and the value is considered as the degree of relevance of the site. At this time, the staying time is not obtained only for URLs having the same frequency as in (3), but the staying time is obtained for all clicked URLs, and is used as a weight for the number of clicks. As a result, the degree of relevance between the keyword and the site can be obtained in many ways.

(5) The staying time is the time when the user clicked a link to a certain web page and the time when the user clicked a link to another web page on the web page posted on the search result page. The method according to (3) or (4), which is determined by a difference.

  According to such a configuration, the user staying time (viewing time) in the URL clicked first by the difference between the time when the link to a certain Web page is clicked and the time when the link to another Web page is clicked next is calculated. Can be sought. That is, the staying time can be easily obtained from the access date and time (time) stored in the query log. In this method, the URL clicked at the end of the session does not have the next URL, and therefore, for convenience, a sufficiently long value, for example, 30 minutes is given as the stay time.

(6) The link to the Web page posted on the search result page is a redirector URL, and when the user clicks the link, the redirector redirects the original link destination to the original link destination. Method.

  According to such a configuration, the link to the Web page posted on the search result page is not the URL of the Web page directly, but the URL of the redirector page set in advance. As a result, when the user clicks a link to a certain web page from the search result page, it jumps to the redirector page, counts the time and number of clicks when the redirector clicks, and saves it in the query log. Redirect to the original web page. By doing so, the user's query log can be easily accumulated without making the user aware of it.

ただし、ｔ（ｓ）：セッションｓにおけるＵＲＬｕの滞在時間
ｑ：シードクエリー
Ｓ：セッション集合
Ｑ：同種のサイトを検索するためのシードクエリーの集合
ＵＲＬ＿Ｓｃｏｒｅ（ｑ，ｕ）：シードクエリーｑに対するＵＲＬｕのスコア (7) The method according to (4) or (5), wherein URL_Score (u), which is a score related to a stay time for a specific URL for the site verifier, is obtained by the following formula.

Where t (s): URLu stay time in session s
q: Seed query
S: Session set
Q: A set of seed queries for searching similar sites
URL_Score (q, u): URLu score for seed query q

  Here, the seed query means a keyword used by the site verifier to collect sites. According to such a configuration, even when there are a plurality of seed queries for the same type of site as a keyword list, the sum of the staying times is obtained as a score for the seed query for a specific URL, and the score is obtained as a seed query. The total score for the set of seed queries with a URL is obtained by adding all the sets. If the URL having a high overall score is preferentially displayed, the efficiency of verification can be increased. A set of seed queries is previously defined by a site verifier using a keyword list or the like.

(8) The method according to any one of (1) to (7), wherein the counting step is periodically performed at predetermined time intervals.

  According to such a configuration, the server of the present invention automatically searches the site by a given keyword list periodically (for example, every 24 hours, every week, every month, etc.) Even if a new site appears, it can be immediately added to the site verification target.

(9) The method according to any one of (1) to (8), wherein the predetermined keyword list includes, as keywords, predetermined language and discriminatory words for parental control for identifying harmful sites.

  The method of the present invention can be used to identify harmful sites in parental control. Predetermined obscene words and discriminatory words are entered by the site verifier through a keyword list.

(10) The method according to any one of (1) to (8), wherein the predetermined keyword list includes a predetermined trade prohibited article name in an online auction as a keyword.

  In the method of the present invention, a predetermined trade prohibited article name (for example, an article name classified as “medicine”, “narcotics”, “weapons”, “transaction prohibited animals and plants”, “grocery”, etc.) in the Internet auction is used. Can be used to search for listing pages. The predetermined trade prohibited product name is input by the site verifier by a keyword list based on the rules of the net auction.

(11) A server that can be connected to a plurality of users' terminals via a communication network and obtains the degree of association between a keyword and a website,
The server includes a search result page generation unit that generates a search result page when a user of the terminal performs a search using a keyword,
A query log database for storing a query log in the search results of the user;
Query that stores the session ID when the user performs the search, the URL of the Web page that the user clicked from the search result page, the access date and time when the link to the Web page is clicked, and the keyword A log storage unit;
Using the query log database, a click number counting unit for counting the number of clicks for the URL for each keyword;
A query log extraction unit for extracting a query log including keywords in a predetermined keyword list inputted in advance from the query log database;
A server comprising

  According to such a configuration, it is possible to provide a server device having the same function and effect as the method described in (1).

(12) The server according to (11), further comprising means for displaying the URL on a terminal of a site verifier connected to the server in descending order of the number of clicks of the URL included in the extracted query log. .

  According to such a configuration, it is possible to provide a server device having the same operational effects as the method described in (2).

(13) The server further includes a stay time calculation unit that calculates a stay time in the Web page clicked by the user, and for the URLs having the same number of clicks, the URLs are listed in descending order of the stay time. The server according to (12), further comprising means for displaying on a terminal of a site verifier connected to the server.

  According to such a structure, the server apparatus which has the same effect as the method as described in (3) can be provided.

(14) It further includes a stay time calculation unit that calculates a stay time on the Web page clicked by the user from an access date and time for each URL included in the extracted query log, and the weighting degree obtained from the stay time and the click The server according to (11), further including means for calculating a degree of association between a keyword and a site by multiplying the number, and displaying the URL on a terminal of a site verifier connected to the server in descending order of the degree of association. .

  According to such a structure, the server apparatus which has the same effect as the method as described in (4) can be provided.

(15) A computer program for obtaining a degree of association between a keyword and a website in a server connectable to a plurality of user terminals via a communication network,
The server includes a means for generating a search result page when the user of the terminal performs a search using a keyword and displaying the search result page on the user terminal, and a query for storing a click log in the search result of the user Log storage means,
To the server,
The query log includes the session ID when the user performs the search, the URL of the Web page clicked by the user from the search result page, the access date and time when the link to the Web page is clicked, and the keyword. Accumulating in storage means;
Using the query log storage means to count the number of clicks for the URL for each keyword;
Extracting a query log including keywords in a predetermined keyword list input in advance from the query log storage means;
A computer program that runs

  According to such a configuration, the present invention can be provided in the form of a computer program as means for realizing a server device having the same operational effects as the method described in (1).

  According to the present invention, by using a click log in keyword search and a query log including access date and time, a site verifier can, for example, a page of a harmful site, a page of an access-restricted site, an exhibition page of prohibited items in an auction, for example. It is possible to efficiently and continuously collect a page of a topical site for a specific subject.

  Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings.

[Configuration of Query Log Total Server]
FIG. 1 is a diagram showing an overall configuration of a system according to an example of a preferred embodiment of the present invention and functional blocks of a query log tabulation server as a core thereof. As an overall configuration of this system, a query log totaling server 10 is connected to a plurality of user terminals 20 via the Internet 21 and further connected to a search server 30 via a network 31. The network 31 may be an arbitrary communication network, and may be a local area network (LAN), a wide area network (WAN), or the Internet. A user searches for a Web site containing information to be searched on the Internet using a search keyword from a user terminal 20 such as a PC (Personal Computer), a PDA (Personal Digital Assistant), or a mobile phone. A search performed by a general user at a search site or the like using a keyword is referred to as a query in this specification. A keyword used by a site verifier to collect sites is called a seed query. Usually, the query request is transmitted to the search server 30, the keyword is received by the query receiving unit 32 of the search server 30, and search processing is performed using various known search engines 33. The search server 30 can communicate with the user terminal 20 directly or indirectly via the Internet. In the present invention, the query log totaling server 10 is connected between the user terminal 20 and the search server 30. . That is, the query log totaling server 10 plays a role of mediating between the user terminal 20 and the search server 30.

  When the query log totaling server 10 receives the query request from the user terminal 20, it transfers it to the search server 30. At this time, the session ID of the query is stored. When the search result for the user query including the corresponding session ID is received from the search server 30, the search result generation unit 11 generates a unique search result page for the user terminal 20. Normally, the search result page includes a link to a Web page that is matched (determined as a keyword), but the server 10 can be posted on this unique search result page by a redirector method described later. Whenever the clicked link destination is clicked, the query log storage unit 12 stores the user's query log. In addition to the session ID, the query log includes the URL of the Web page that the user clicked from the search result page, the access date and time when the link to the Web page is clicked, and the keyword used for the search.

  The query log data is stored in a query log database (query log DB 17; hereinafter, the database may be abbreviated as DB). The query logs of a large number of users stored in the query log DB 17 are subjected to timely tabulation work by a tabulation unit (not shown) of the query log tabulation server 10. The totaling unit is constituted by, for example, the click number totaling unit 13 and the stay time calculating unit 14, and the keyword used in the query for the data in the query log and the clicked URL are set at a certain time interval (for example, every 24 hours, (Every week, every month, etc.) The obtained various total information may be stored in the query log DB 17 or may be stored in another database. Such various databases may be configured as the storage unit 10b of the query log totaling server 10, but may be configured to be connected to the outside of the server.

  The query log totaling server 10 is connected to the site verifier terminal 40 via the network 31 or directly via the display / operation unit 16 as shown in the figure. The site verifier is a person who collects specific types of Web sites and verifies whether or not those sites actually meet the type “classification”. The site verifier uses the data accumulated in the query log DB 17 to collect a desired type of site (for example, a “harmful” site) and verifies its contents. Specifically, the site verifier creates a keyword list for collecting sites and inputs it to the query log totaling server 10. When there are many keyword lists, they may be registered and stored in the keyword list DB 18.

  The query log extraction unit 15 uses a method as described later to determine the degree of relevance for keywords included in this predetermined keyword list (which may be a single keyword or a set of keywords). Extract high site URLs. The extracted results are used by the site verifier by some method (for example, displaying the rankings).

  In addition, each function part 11-15 of such a query log totalization server 10 comprises the control part 10a (typically the function performed by CPU of a computer). In the above embodiment, the query log totaling server 10 and the search server 30 have been described as being separate servers for easy understanding. However, the search server 30 is not necessarily limited to this configuration. The function of the query log totaling server 10 may be included.

  FIG. 2 is a diagram summarizing the above processing flow in the system. Details will be omitted because they overlap with the above description. Although the query log totaling process of the query log totaling server 10 is not shown in this figure, the query log totaling process is performed independently of the process here (described later).

  FIG. 3 is a diagram showing the concept of the redirector 42 as one method for aggregating query logs. When the search result page generation unit 11 of the query log totaling server 10 receives a search result for the user query from the search server 30, the search result page generation unit 11 replaces the link destination to each Web page posted on the search result page 41 with the URL of the redirector 42. . Then, the search result page 41 including the redirector URL is returned to the user terminal 20. When the user clicks a link (for example, page A link) to a desired Web page from the search result page 41 received, the user actually jumps to the redirector 42.

  When the user clicks on the link of the Web page posted on the search result page, the redirector 42, the various data for the query, specifically, the IP address of the user terminal 20 (not necessarily a fixed IP address), Query log, such as the access date and time when a link to a Web page is clicked, the session ID of the query session, one or more keywords used by the user for the query, and the original URL of the Web page that the user clicked from the search results page The data 44 is stored in the query log DB 17. Thereafter, the user is caused to jump to the original linked web page (linked page 43). Thus, by using the redirector 42, the query log totaling server 10 can easily collect query logs for each user identified by the IP address and for each session identified by the session ID.

  FIG. 4 is a diagram illustrating an example of the query log data 44 stored in the query log DB 17. As shown in the figure, this table shows the IP address for identifying the user terminal 20, the access date and time clicked from the search result page, the query session ID, the keyword used in the query, and the URL actually clicked in a table format. Stored. In this example, the search result obtained as a result of searching from the user terminal 20 (IP address 110.149.145.1) using “uncorrected” and “image” as query keywords in the query session ID: Bp4ed6917. From the page, five URLs as shown on the right end of the table are actually accessed when the user clicks on the link to the Web page sequentially from 21:45:53 on October 6, 2006 ( Time) is recorded.

  Although not specifically shown, the display order (rank) of each URL on the search result page may also be recorded. This is because the higher the probability that the user clicks on the link at the top of the search result page, it is possible to take it into account. For example, it is known that the link displayed on the first page of the search result page (from Yahoo! search, the first to tenth pages are displayed on one page) has a high click rate. ing. Therefore, when a link displayed on the second page or later is clicked, a higher degree of weight than the link on the first page may be given.

  FIG. 5 is a diagram illustrating an example of a totaling process procedure of the query log totaling server 10. First, in step S1, a set of query logs is acquired. That is, the query logs of all users for a certain counting period are collected for each session ID.

  Next, in step S2, the set is sorted by access date. Further, in step S3, a difference between the access date / time for one URL in a session and the access date / time for the next clicked URL is obtained, and this is set as the stay time in the first URL. This process is repeated for all URLs in the session. This is shown in detail in FIG.

  FIG. 6 is a diagram showing an outline of a method for obtaining the stay time of a Web page. This figure is Yahoo! In the search, an example is shown in which a search result page 50 is displayed as a result of a search performed by the user using “AAA” and “BBB” as search keywords. In practice, the term “AAA” is “uncorrected” and “BBB” is “image”. In this search result, about 6 million or more sites have been hit, but it is very difficult for the site verifier to verify all the sites.

  For example, the user clicks the “AAABBBB forefront” page of the first ranking of the search result page 50 at time t1s, browses the Web page 51 that is the linked page of “AAABBB forefront”, and then returns to the browser “return”. ”Button at time t 1 e to return to the search result page 50. Similarly, it is assumed that the Web page 52 that is the linked page of “AAABBB treasure” in the second rank is clicked at time t2s and returned to time t2e. At this time, the stay time of the user on the page 51 is t1e-t1s, and the stay time of the page 52 is t2e-t2s. However, since the time from t1e to the next t2s is generally short and can be ignored, the method of the present invention assumes that t1e is approximately equal to t2s. That is, the staying time of the page 51 can be approximately calculated by t2s-t1s. The stay time is similarly obtained for the next page 52 and thereafter.

  However, if the page 52 is the last viewed page in this query session, the page clicked next does not exist, so the above method cannot be used. However, in this case (when the next URL does not exist), a sufficiently long time, for example, 30 minutes is set as the staying time. Or you may adjust so that the weight degree of the last URL may be made larger than a normal weight. For example, the average staying time other than the last URL may be doubled as the last URL staying time. This is because the page clicked last has a high probability that the information requested by the user exists. That is, it can be inferred that the relationship between the keyword and the site is high. Thus, in the embodiment of the present invention, the calculation of the stay time is approximately obtained. However, the redirector 42 can detect the pressing of the “return” button or the like of the browser by some method, and the stay time is more accurately determined. Of course, you may ask for it.

  When the stay time calculation step is completed, the process returns to step S4 in FIG. Next, in step S5, the URLs with the same number of clicks are sorted in the order of staying time, and the tabulation process ends. In step S5, the stay time is obtained for all URLs, not only for URLs with the same frequency, and the weight (weight) is obtained therefrom, and the value multiplied by the click frequency is obtained. You may make it sort in order. For example, when the staying time is within 30 seconds, the weighting factor is set to 1, and thereafter, every time the staying time increases by 30 seconds, the weighting factor is added by one. By doing this, a Web page with a short stay time even if the click frequency is high may be due to reasons such as low relevance to the keyword or whether the title and content of the Web page do not match. Since the rank of the page is relatively low, it helps to improve the site verification efficiency.

  Further, as described above, the rank (rank) displayed on the search result page may be considered as a separate weight. For example, since it is natural that many Web pages ranked higher in the search result page are clicked, a URL having a higher number of clicks than a predetermined number or a URL having a high click rate although the rank is low. On the other hand, a method such as doubling the weight is conceivable.

Also, URL_Score (u), which is a score related to the staying time for a specific URL for the site verifier, is obtained by the following formula.

Where t (s): URLu stay time in session s
q: Seed query
S: Session set
Q: A set of seed queries for searching similar sites
URL_Score (q, u): URLu score for seed query q

  In this way, even when there are a plurality of seed queries for obtaining the same type of site for the site verifier, the sum of the staying times is obtained as a score for a specific URL, and the score is obtained for all sets of seed queries. The score for each URL is obtained by adding at.

  FIG. 7 is a diagram illustrating an example of query log tabulation result data stored in the query log DB 17. Here, the results of aggregation for each URL with respect to the search keyword (keyword used in the query) “AAA” are shown. For example, URL http: // xxx. aaa. When bbb was counted at 12:00 on December 1, 2006, the number of clicks 143 and the weight (weight) calculated from the stay time was 43, but the next 24 hours in December 2006 At 12:00 on the 2nd, the number of clicks is 189 and the weight is 89. In this example, data for every 24 hours is collected as the total period, but these may be further collected to generate total data for every week, for example, every month. In this example, only a single keyword “AAA” is shown, but a plurality of words, for example, “AAA” and “BBB”, “AAA” or “BBB” are also counted as one keyword.

[Example]
FIG. 8 is a diagram showing a utilization method in parental control as an embodiment of the present invention. In this example, it is assumed that a tabulation result data table indicated by reference numeral 72 is stored in the query log DB 17. The tabulation result data 72 is basically the same as the table described with reference to FIG. Further, in this example, it is assumed that the verifier of the harmful site has input the keyword list 75 into the query log totaling server 10 by storing the keyword list 75 in the NG keyword DB 74. In the keyword list 75, keywords classified into obscene, violent, discrimination, grotesque, and the like are input.

  The query log totaling server 10 compares this keyword list 75 with the totaling result data table 72, and information including NG keywords (here, “ddd”, “bbb”, “fff”) from the totaling result data table 72 ( (A total data record) is extracted (processing indicated by reference numeral 73). Then, the degree of association is calculated from the total data record by multiplying the number of clicks and the weight. In this example, the degree of association between the NG keyword “ddd” and the URL http: // xxx is 16821. Similarly, the degree of association between the NG keyword “bbb” and the URL http: // yyy is 13400, and the degree of association between “fff” and the URL http: // zzz is 4224. The query log totaling server 10 creates a determination list DB 76 storing the relevance data 77, sorts the URLs in descending order of relevance, and displays them on the site verifier's terminal. The site verifier browses the contents of the Web pages in order from the top of the sorted URL, performs access restriction determination (processing indicated by reference numeral 78), and registers the URL of the corresponding Web page in the blacklist DB 79. To do. In this way, by using the query log tabulation server 10 of the present invention for collecting and verifying harmful sites, the site verifier's site verification efficiency can be increased.

  In the above embodiment, the method of the present invention was used to create a blacklist in parental control. However, as another application example, a search for an exhibition page including trade prohibited articles in a net auction, or a topic related to a specific subject matter. This site can be used for an information service characterized in that the site is efficiently fixed and information is quickly provided.

[Hardware configuration of query log summary server]
FIG. 9 is a diagram illustrating an example of a hardware configuration of the query log totaling server 10 according to an example of the preferred embodiment of the invention. The query log totaling server 10 includes a central processing unit (CPU) 1010 (a plurality of CPUs such as the CPU 1012 may be added in a multiprocessor configuration), a bus line 1005, a communication I / F 1040, and a main memory 1050. , A BIOS (Basic Input Output System) 1060, a USB port 1090, an I / O controller 1070, an input means such as a keyboard and mouse 1100, and a display device 1022.

  Storage means such as a tape drive 1072, a hard disk 1074, an optical disk drive 1076, and a semiconductor memory 1078 can be connected to the I / O controller 1070.

  The BIOS 1060 stores a boot program executed by the CPU 1010 when the query log totaling server 10 is started, a program depending on the hardware of the query log totaling server 10, and the like.

  The hard disk 1074 constituting the storage unit 10b stores a program for the query log totaling server 10 to execute the functions of the present invention, and various databases can be configured as necessary.

  As the optical disc drive 1076, for example, a DVD-ROM drive, a CD-ROM drive, a DVD-RAM drive, or a CD-RAM drive can be used. In this case, the optical disk 1077 corresponding to each drive is used. A program or data can be read from the optical disk 1077 by the optical disk drive 1076 and provided to the main memory 1050 or the hard disk 1074 via the I / O controller 1070. Similarly, the tape medium 1071 corresponding to the tape drive 1072 can be used mainly for backup.

  The program provided to the query log totaling server 10 is provided by being stored in a recording medium such as the hard disk 1074, the optical disk 1077, or a memory card. This program may be read from the recording medium via the I / O controller 1070 or downloaded via the communication I / F 1040 to be installed and executed in the query log totaling server 10.

  The aforementioned program may be stored in an internal or external storage medium. Here, in addition to the hard disk 1074, the optical disk 1077, or the memory card, a magneto-optical recording medium such as MD and a tape medium can be used as the storage medium constituting the storage unit 10b. Further, a storage device such as a hard disk 1074 or an optical disk library provided in a server system connected to a dedicated communication line or the Internet may be used as a recording medium, and the program may be provided to the query log totaling server 10 via the communication line. .

  Here, the display device 1022 displays a screen for accepting data input to the user or displays a calculation processing result screen by the query log totaling server 10, and includes a cathode ray tube display device (CRT), a liquid crystal display, and the like. Includes a display device such as a device (LCD).

  Here, the input means accepts input by the user, and may be configured by a keyboard, a mouse 1100, and the like.

  The communication I / F 1040 is a network adapter that enables the query log totaling server 10 to be connected to a terminal via a dedicated network or a public network. The communication I / F 1040 may include a modem, a cable modem, and an Ethernet (registered trademark) adapter.

  In the above example, the query log totaling server 10 has been mainly described. However, the functions described above can also be realized by installing a program in a computer and operating the computer as a server device. Therefore, the functions realized by the server described as an embodiment in the present invention can be realized by executing the above-described method by the computer, or by introducing the above-mentioned program into the computer and executing it. It is.

  As mentioned above, although embodiment of this invention was described, this invention is not restricted to embodiment mentioned above. The effects described in the embodiments of the present invention are only the most preferable effects resulting from the present invention, and the effects of the present invention are limited to those described in the embodiments of the present invention. is not.

1 is a diagram illustrating an overall configuration of a system according to an example of a preferred embodiment of the present invention and functional blocks of a query log tabulation server as a core thereof. It is the figure which showed the flow of the process in the system which concerns on an example of suitable embodiment of this invention. It is a figure which shows the concept of the redirector 42 as one method for totaling a query log. It is a figure which shows an example of the query log data 44 stored in query log DB17. It is a figure which shows an example of the total process procedure of the query log total server. It is a figure which shows the outline of the method of calculating | requiring the stay time of a web page. It is a figure which shows an example of the data of the query log total result stored in query log DB17. It is the figure which showed the utilization method in parental control as one Example of this invention. It is a figure which shows an example of the hardware constitutions of the query log totalization server 10 which concerns on an example of suitable embodiment of invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Query log total server 10a Control part 10b Memory | storage part 11 Search result page production | generation part 12 Query log preservation | save part 13 Click count total part 14 Stay time calculation part 15 Query log extraction part 16 Display part / operation part 17 Query log DB
18 Keyword List DB
20 User terminal 21 Internet 30 Search server 31 Network 32 Query accepting unit 33 Search engine 40 Site verifier terminal 41 Search result page 42 Redirector 43 Link page 44 Query log data 50 Search result page 51 Link page 52 Link page 72 Total Result data 73 NG keyword match information extraction process 74 NG keyword DB
75 NG Keyword Table 76 Judgment List DB
77 Relevance data 78 Access restriction judgment processing 79 Black list DB

Claims (15)

A server that can be connected to a plurality of user terminals via a communication network is a method for determining the degree of association between a keyword and a website,
The server generates a search result page when the user of the terminal performs a search by a keyword and displays the search result page on the user's terminal, and a click log from the search result page of the user. A query log storage means for storing,
In the server,
The query log includes the session ID when the user performs the search, the URL of the Web page clicked by the user from the search result page, the access date and time when the link to the Web page is clicked, and the keyword. Accumulating in storage means;
Using the query log storage means to count the number of clicks for the URL for each keyword;
Extracting a query log including keywords in a predetermined keyword list input in advance from the query log storage means;
Including methods.   The method according to claim 1, further comprising: displaying the URLs on a site verifier terminal connected to the server in descending order of the number of clicks of the URLs included in the extracted query log.   The stay time on the Web page clicked by the user is calculated from the access date and time for each URL included in the extracted query log. For the URLs with the same frequency of click, the URLs are listed in the order of long stay time. The method of claim 2, further comprising displaying on a site verifier terminal connected to the server.   Based on the access date and time for each URL included in the extracted query log, the stay time on the Web page clicked by the user is calculated, and the relationship between the keyword and the site is calculated by multiplying the weight obtained from the stay time and the number of clicks. The method according to claim 1, further comprising calculating a degree and displaying the URL on a site verifier terminal connected to the server in descending order of the relevance.   The staying time is obtained by a difference between a time when the user clicks a link to a certain web page and a time when a user clicks a link to another web page on the web page posted on the search result page. The method according to claim 3 or 4.   The method according to claim 1, wherein a link to a Web page to be posted on the search result page is a redirector URL, and when the user clicks the link, the redirector redirects to an original link destination. . The method according to claim 4 or 5, wherein URL_Score (u), which is a score related to a staying time for a specific URL for the site verifier, is obtained by the following formula.

Where t (s): URLu stay time in session s
q: Seed query
S: Session set
Q: A set of seed queries for searching similar sites
URL_Score (q, u): URLu score for seed query q   The method according to any one of claims 1 to 7, wherein the counting step is periodically performed at predetermined time intervals.   9. The method according to claim 1, wherein the predetermined keyword list includes, as keywords, predetermined words and discriminatory words for parental control for identifying harmful sites.   The method according to any one of claims 1 to 8, wherein the predetermined keyword list includes a predetermined trade prohibited article name in an online auction as a keyword. A server that can be connected to a plurality of users' terminals via a communication network and obtains the degree of association between a keyword and a website,
The server includes a search result page generation unit that generates a search result page when a user of the terminal performs a search using a keyword,
A query log database for storing a query log in the search results of the user;
Query that stores the session ID when the user performs the search, the URL of the Web page that the user clicked from the search result page, the access date and time when the link to the Web page is clicked, and the keyword A log storage unit;
Using the query log database, a click number counting unit for counting the number of clicks for the URL for each keyword;
A query log extraction unit for extracting a query log including keywords in a predetermined keyword list inputted in advance from the query log database;
A server comprising   The server according to claim 11, further comprising means for displaying the URLs on a site verifier terminal connected to the server in descending order of the number of clicks of the URLs included in the extracted query log.   The server further includes a stay time calculation unit that calculates a stay time on the Web page clicked by the user, and the URLs are connected to the server in the descending order of the stay time for click URLs having the same frequency. The server according to claim 12, further comprising means for displaying on a terminal of a verified site verifier.   A stay time calculation unit for calculating a stay time on the Web page clicked by the user from an access date and time for each URL included in the extracted query log, and further comprising: a weight degree obtained from the stay time and the number of clicks The server according to claim 11, further comprising means for multiplying a keyword and a site to calculate a degree of association, and displaying the URL on a site verifier terminal connected to the server in descending order of the degree of association. A computer program for determining the degree of association between a keyword and a website in a server connectable to a plurality of user terminals via a communication network,
The server includes a means for generating a search result page when the user of the terminal performs a search using a keyword and displaying the search result page on the user terminal, and a query for storing a click log in the search result of the user Log storage means,
To the server,
The query log includes the session ID when the user performs the search, the URL of the Web page clicked by the user from the search result page, the access date and time when the link to the Web page is clicked, and the keyword. Accumulating in storage means;
Using the query log storage means to count the number of clicks for the URL for each keyword;
Extracting a query log including keywords in a predetermined keyword list input in advance from the query log storage means;
A computer program that runs

Images