KR20010082966A - Method and system for providing related web sites for the current visitting of client - Google Patents

Abstract

PURPOSE: A method and system for providing a related web site is provided to offer a related list to a user in real time by extracting web sites closely related to a web site the user visits in current while the user uses the Internet. CONSTITUTION: At least one standard URL is created by normalizing a URL of each web site from an Internet bookmark DB(100). At least one related URL is created by extracting a relation between web sites from the standard URL(200). At least one related URL headword is created by visiting each URL through a server agent(300). Web sites closely related to a web site the user visits to in current are searched from the created related URL and the created related URL headword and are provided to a user in real time.

Description

Method and system for providing related web sites for the current visitting of client

The present invention relates to a method for providing a related web site and a system for performing the same, and more particularly, to a method for extracting a related web site using an internet book DB and providing the same to a user side, and a system for performing the same.

Until the early 1980s, the Internet, which was used primarily by small groups of specialists for research institutes, has been exploding in use as the world wide web (WWW) has expanded its use to general commercial use. As of the end of November 1999, the US Internet users were 110 million, accounting for 43% of the world's 259 million Internet users, and Korea's 10,580,000 Internet users were ranked 10th in the world. : Almanac, USA. Considering the number of people in Korea, it is expected that in 2002 more than half of the population will use the Internet.

Individual PC users who wish to access the Internet typically use application software known as web browsers. The web browser establishes a connection to another computer known as a web server via the Internet, and receives information displayed on the user's PC from the web server. The information sent from the web server to the web browser is formatted using a special language called a general Hyper Text Markup Language (HTML), and is organized into papers, commonly known as web pages.

On the other hand, Internet users' interest is to quickly find a site that is suitable for them and use the service among a large number of Internet sites provided in news, media, entertainment, finance, shopping, technology, and culture. A representative way to satisfy the needs of users is to use services of Internet search companies such as Yahoo and Lycos.

These search companies use manual or automated search robots to find and classify the data on the Internet site and make it into a database. When the user enters a word to search, the searched DB is searched to provide the site information that matches the entered word. Or, by presenting the classification tree constructed by the search company to the user, the user himself searches down the classification tree and finds the site he / she wants.

In addition to direct Internet search companies, it is also possible to access desired Internet sites by using a hub site method that has recently been actively used. These hub sites are sites that direct related web sites from one web site.

The user can easily visit the site later by using a search service or storing his favorite Internet site found in another path in the form of a bookmark in his local disk.

Accordingly, the present invention has been made in view of the above-mentioned problems, and an object of the present invention is to extract a web site highly relevant to the web site that the user is currently visiting while the user is navigating the Internet, and to provide the user with a related list in real time. It is to provide a method for providing a related website for providing.

Another object of the present invention is to provide a system for performing the above-described related web site providing method.

1 is a view for explaining a related web site extraction system according to an embodiment of the present invention.

2 is a flowchart illustrating a related web site providing method according to an embodiment of the present invention.

3 is a flowchart illustrating a URL standardization process according to an embodiment of the present invention.

4 is a flowchart illustrating a process of extracting a related URL according to an embodiment of the present invention.

5 is a flowchart illustrating a process of extracting a URL header according to an embodiment of the present invention.

6 is a flowchart illustrating a process of determining the same URL according to an embodiment of the present invention.

<Description of the symbols for the main parts of the drawings>

100: book mark DB 200: URL standardization unit

210: determination target DB 220: the same URL determination unit

230: duplicate domain DB 240: URL normalization unit

250: standard URL DB 300: related URL extraction unit

310: URL related extraction unit 320: related URL DB

400: URL header construction unit 410: URL header extraction unit

420: URL header DB 500: service provider

510: web server

Related website providing method according to one feature for realizing the above object of the present invention,

(a) normalizing the URL of each web site from an internet bookmark DB to generate one or more standard URLs;

(b) generating one or more related URLs by extracting associations between respective web sites from the standard URLs;

(c) visiting each URL through a server agent to generate one or more related URL headings; And

(d) search for sites that are highly relevant to the Internet web site the user is currently visiting, from one or more related URLs generated in step (b) and one or more related URL headings generated in step (c). Providing in real time.

In addition, the related website extraction system according to one feature for realizing another object of the present invention described above,

An Internet book DB that stores URLs of one or more web sites;

A URL normalization unit for constructing a standard URL DB by normalizing URLs stored in the Internet bookmark DB;

A related URL extracting unit for constructing a related URL DB by extracting a relation between respective web sites from the standard URL DB;

A URL heading construction unit for extracting headings of related URLs by visiting each URL through a server agent, and constructing a related URL heading DB; And

And a service providing unit for providing the user with at least one site highly related to the Internet web site the user is currently visiting based on the information stored in the related URL DB and the URL heading DB.

According to such a method for providing a related web site and a system for performing the same, it is possible to extract a web site highly relevant to a web site on the Internet which the user is currently visiting and provide a related list to the user in real time.

Then, embodiments will be described so that those skilled in the art can easily implement the present invention.

1 is a view for explaining a related web site extraction system according to an embodiment of the present invention.

As shown in FIG. 1, the related web site extraction system according to an embodiment of the present invention includes a book mark DB 100, a URL standardization unit 200, an associated URL extraction unit 300, and a URL heading construction unit 400. And a service provider 500.

The book mark DB 100 is bookmark information of Internet users, and subscribers use their bookmarks, preferably a Uniform Resource Locator (URL), to access their bookmarks online on the Internet. Save the address.

In this case, the URL can directly receive the representation and data of any file or service on the Internet with the network service necessary for the web server to search and interpret any specific information, file or resource existing on the Internet.

For example, when expressed as 'http://www.trumpet.com.ar/', 'http' means protocol and 'www.trumpet.com.ar' means where you want to connect. The ':' in the URL indicates the method to access, and after the colon (:) the location of the data or the address of the server providing the service. And the remaining part shows the file name to access the port number to be connected.

The URL standardization unit 200 includes a determination target DB 210, an identical URL determination unit 220, a duplicate domain DB 230, a URL normalization unit 240, and a standard URL DB 250. URLs stored in 100 are normalized and stored in the standard URL DB 250.

In more detail, the same URL determination unit 220 extracts a domain name that needs to be determined from the domain name included in each URL in the bookmark DB 100 and the same object, and stores the same in the determination target DB 210. After that, by reading the determination target domain from the determination target DB 210 and searching through the Internet whether it represents the same site, the information stored in the duplicate domain DB 230 is updated.

The URL normalization unit 240 converts the URLs expressed in various forms into a standard format with reference to the bookmark DB 100 and the duplicate domain DB 230, and the standard URL DB 250 is provided to the URL normalization unit 240. Stores the URL converted to the standard format and the folder unique number of the URL.

The related URL extractor 300 includes a URL related extractor 310 and an associated URL DB 320. The URL extractor 300 reads URL and folder information from standard URLs stored in the standard URL DB 250. Search for high URLs and store <URL, related information> information in the relevant URL DB 320 for the top N highly relevant.

The URL heading construction unit 400 includes a URL heading extraction unit 410 and a URL heading DB 420 to extract URL headings by visiting each URL on the Internet through a server agent and to store the URL headings in the URL heading DB 420. do.

In more detail, the URL heading extractor 410 may be configured on the Internet when URL headings are not written for URLs in the relevant URL DB 320, or when a URL heading point needs to be checked again after a certain period of time. The URL is read and the heading is stored in the URL heading DB 420.

If the relevant URL DB 320 and the URL heading DB 420 are constructed, the service provider 500 may, via the web server 510, based on the related URL information and the URL heading information, and the Internet website that the user is currently visiting. Provide users with one or more sites that are highly relevant to Of course, the service may not be provided when a site that is related to the website that the user is visiting is not searched.

The client unit 600 operating in the user's PC inquires the web server 510 of the Internet website URL information that the user is currently visiting, and the web server 510 has a related URL corresponding to the URL and the degree of the relatedness. , The relevant URL DB 320 and the URL heading DB 420 are searched to transmit the retrieved related URL, heading, and relevance data to the client unit 600 so that the client unit 600 displays the information to the user in real time. To do it.

Among the parts of which information is automatically modified by the configuration modules in the entire operation process described above, the determination target DB 210, the duplicate domain DB 230, and the URL heading DB 420 may have insufficient information by automation. As there is possibility, it is possible to modify the DB manually by administrator.

2 is a flowchart illustrating a related web site providing method according to an embodiment of the present invention.

Referring to FIGS. 1 and 2, first, one or more standard URLs are generated by normalizing the URL of each web site from the Internet bookmark DB 100 and stored in the standard URL DB 250 (step S100).

Subsequently, the relations between the web sites are extracted from the standard URL information stored in the standard URL DB 250, and one or more related URLs are generated and stored in the related URL DB 320 (step S200).

Subsequently, each URL is visited through the server agent to generate one or more related URL headings and stored in the URL heading DB 420 (step S300).

Subsequently, sites most relevant to the Internet web site the user is currently visiting are searched for from one or more related URLs generated in step S200 and one or more related URL headings generated in step S300 and provided to the user in real time (step S400). ).

3 is a flowchart illustrating a URL standardization process according to an embodiment of the present invention.

As shown in Fig. 3, the URL standardization process first checks whether the URL service is HTTP or HTTPS (step S210), and if the URL service is not HTTP or HTTPS, excludes it from standardization (step S220). To exit. Of course, depending on the application, you can also include other services, for example, FTP.

If the URL service is HTTP or HTTPS in step S210, the service name is removed (step S230). For example, if it is displayed as 'http://www.microsoft.com', remove the 'http: //' part.

Next, it is checked whether there is a representative domain corresponding to the domain name of the URL with reference to the duplicate domain DB (step S240), and if there is a representative domain in the duplicate domain DB, the domain name in the URL is replaced with the representative domain (step S240). S245).

If there is no representative domain in the duplicate domain DB in step S240 or after step S245, it is checked whether the final file name of the URL is a basic web file (step S250).

If the final file name of the URL is the basic web file in step S250, the final file name in the URL is removed (step S255). For example, if the default web file is 'default.asp' or 'index.html', remove it.

If the final file name of the URL is not the basic web file in step S250 or after step S255, the directory display portion is removed to generate a standard URL (step S260). That is, if necessary, various methods can be used, such as shortening to a host name of a URL or restricting to only the upper directory of the final pass step.

For example, you can abbreviate the parent directory level 'www.microsoft.com' by removing the slash (/) from 'www.microsoft.com/'.

By using the standardized URL, the above process can be used to retrieve the relevant URL in real time from the web service or to store it in the standard URL DB 250.

The related URL generation step described with reference to FIG. 2 is performed by obtaining the relevant URLs for each URL by using the constructed standard URL DB 250. The detailed process follows the algorithm described in FIG. 4.

4 is a flowchart illustrating a process of extracting a related URL according to an embodiment of the present invention.

1 and 4, since different URLs in the standard URL DB 250 are related to the construction of the related URL DB, different URLs in the related URL DB (let the first URL group u1 for convenience of description of the present invention). For each of these, proceed as follows.

First, all folders to which the first URL group u1 belongs are obtained from the standard URL DB 250, and this is called F1 (step S310). This is because some users can classify the first URL group u1 as folder A, while others can classify the same URL as folder B. That is, all the folders of F1 thus obtained include the first URL group u1.

Subsequently, for each folder of F1 obtained in step S310, all URLs belonging to this folder are found and put into the second URL group u2 (step S320). That is, all URLs belonging to the second URL group u2 are classified into the same folder as the first URL group u1 by one or more users.

Subsequently, different URLs are found from the second URL group u2 obtained in step S320 and put into the third URL group u3 (step S330).

Next, a frequency appearing in the second URL group u2 is obtained for all the URLs belonging to the third URL group u3 (step S340), and the frequency is sorted so that the top N URLs and the corresponding frequencies are related to the URL DB 320. (Step S350).

As described above, the basic assumption of extracting related URLs is that the URL and folder data contained in the bookmark DB classify URLs of the same category into the same folder. In general, this assumption may be valid because the user can see that URLs of the same category are classified in the same folder when building their bookmarks.

Therefore, if the frequency of appearing in a specific first URL and another specific second URL moving folder is high throughout the DB, the first URL and the second URL can be regarded as highly correlated. The second URL is a meaningful site.

The URL heading DB construction described in step S300 of FIG. 2 is to find the heading of the URL and store it in the DB to explain what the URL means with the URL itself when presenting the URL to the user. That is, the process of finding the headword of each URL with respect to the URLs stored in the relevant URL DB proceeds to the step as shown in FIG. 5.

5 is a flowchart illustrating a process of extracting a URL header according to an embodiment of the present invention.

1 and 5, it is first checked whether the heading of the target URL is contained in the URL heading DB (step S410).

If the headword is contained in the existing DB, it is checked whether or not the defined date of the final confirmation date has passed (step S420), and if it has passed, the URL heading DB construction is finished.

If the date in which the final confirmation date is defined in step S420 has not elapsed, the URL heading DB does not contain the heading of the URL, or the last heading date of the URL heading has not passed the defined date and the URL heading is automatically updated. Is checked (step S430).

In step S430, if the URL heading DB contains the heading of the URL, the process feeds back to step S410, and also feeds back to step S410 even if the last modification date of the URL heading has passed the defined date or the URL heading is not automatically updated. do

In addition, when the URL heading DB does not contain the heading of the URL in step S430, or when the date of the final modification of the URL heading is not defined, and the URL heading is automatically generated, the date of the final confirmation date is defined. With respect to the URL exceeding the visit to the web server to read the HTML (step S440), and receives the URL data to extract the headword (step S450).

The headword extracted in step S450 is upgraded to the URL heading DB (step S460).

During URL normalization, standardizing URLs with different domain names representing the same site into one can improve the accuracy of relevance calculations. If the processing for this is the same URL discrimination process and the domain name is different but represents the same site, a duplicate domain DB is constructed with <domain name, representative domain name>. The process is as shown in FIG.

6 is a flowchart illustrating a process of determining the same URL according to an embodiment of the present invention.

1 and 6, first, a determination target domain is manually or automatically selected (step S510). At this time, if the target domain is automatically selected, if a specific word is the same in the domain name, the target domain is selected. For example, 'kr.yahoo.com' and 'yahoo.co.kr' are included in the object of discrimination because '.com', the word before '.com', and 'yahoo', before the word '.co.kr' are common. In addition, when the target domain is manually selected, the administrator directly inputs the target DB as necessary. At this time, the contents change over time, such as news sites, depending on the time, so after processing the target sites d1 and d2, if the other target site d3, such as d1, exists in the target DB, do.

Then, it is checked whether or not it is the end of the data (step S520). If it is the end of the data, execution is completed, and if it is not the end of the data, HTML content is read from sites d1 and d2 (step S530).

Then, whether the target domains d1 and d2 are the same (step S540), and if the target domains d1 and d2 are different in step S540, the process returns to step S410, and if the target domains d1 and d2 are the same, the duplicate domains are updated. (Step S550). Comparing whether the contents of the two URLs in step S540 are the same is treated as the same if the accuracy of the contents of the two files is greater than or equal to a predetermined accuracy.

In the above description, it has been described to provide related site information by extracting a schedule relation with an Internet web site that the user visits. However, if the relevant relation is not searched, the service may not be provided to the user.

As mentioned above, in the URL standardization process, standardizing URLs having different domain names representing the same site into one can improve the accuracy of the relevance calculation.

Although the above has been described with reference to a preferred embodiment of the present invention, those skilled in the art will be able to variously modify and change the present invention without departing from the spirit and scope of the invention as set forth in the claims below. It will be appreciated.

As described above, according to the present invention, the relevance ranking for the Internet sites is prepared by using the Internet book mark DB information directly classified by the user, and the sites highly relevant to the Internet sites currently visited by the users are real-time. I can recommend it.

Claims (10)

(a) normalizing the URL of each web site from an internet bookmark DB to generate one or more standard URLs; (b) generating one or more related URLs by extracting associations between respective web sites from the standard URLs; (c) visiting each URL through a server agent to generate one or more related URL headings; And (d) search for the sites that are highly relevant to the Internet web site the user is currently visiting from the one or more related URLs generated in step (b) and the one or more related URL headings generated in step (c) Deliver in real time Related website providing method comprising a. The method of claim 1, wherein step (a) (a-1) checking whether the URL service is HTTP / HTTPS and excluding it from the standardization target if the URL service is not HTTP / HTTPS; (a-2) if the URL service is HTTP / HTTPS in step (a-1), checking whether a representative domain exists in the duplicate domain DB by removing the service name; (a-3) if the representative domain exists in step (a-2), replacing the domain name in the URL with the representative domain; (a-4) checking if the representative domain does not exist in step (a-2) and after step (a-3) whether the final file name of the URL is a basic web file; (a-5) if the final file name of the URL is a basic web file in step (a-4), removing the final file name in the URL; And (a-6) in the step (a-4), if the final file name of the URL is not a basic web file and after removing the last file name, removing the directory display to generate a standard URL. How to Provide Related Websites. The method of claim 1, wherein step (b) comprises: (b-1) extracting all folders to which the first URL which is a standard URL belong; (b-2) extracting second URLs belonging to all folders obtained in the step (b-1); (b-3) extracting a different third URL from the second URL obtained in step (b-2); (b-4) extracting frequencies appearing in the second URL for all the URLs belonging to the third URL; And and (b-5) storing the top N URLs and the frequencies of all URLs obtained in the step (b-4) by sorting by frequency. The method of claim 1, wherein step (c) comprises: (c-1) reading a first URL from the one or more related URLs and checking whether it is the end of data and ending if it is the end of data; (c-2) checking whether the first URL is new when it is not the end of the data in step (c-1), and feeding back to the step (c-1) when it is non-new; (c-3) reading HTML from an Internet web site if the first URL is new in step (c-2); (c-4) extracting the headword from the HTML read in step (c-3); And (c-5) updating the extracted headwords. An Internet book DB that stores URLs of one or more web sites; A URL normalization unit for constructing a standard URL DB by normalizing URLs stored in the Internet bookmark DB; A related URL extracting unit for constructing a related URL DB by extracting a relation between respective web sites from the standard URL DB; A URL heading construction unit for extracting headings of related URLs by visiting each URL through a server agent, and constructing a related URL heading DB; And A service providing unit that provides the user with at least one site that is highly related to the Internet web site that the user is currently visiting based on the information stored in the related URL DB and the URL heading DB. Related website providing system comprising a. The method of claim 5, wherein the service provider, Related website providing system, characterized in that to provide the at least one site in real time. The method of claim 5, wherein the URL standardization unit, Determination target DB; Duplicate domain DBs that store representative domain names for these domains if they are different domain names but actually specify the same web site; From the domain name contained in each URL in the bookmark DB, it is necessary to extract the domain name that needs to be identified, store it in the target DB, and read the target domain from the target DB to determine the same site. A same URL discrimination unit which searches whether the information is displayed through the Internet and updates the information stored in the duplicate domain DB; A URL normalization unit extracting a URL having a standard format by referring to the bookmark DB and the duplicate domain DB; And And a standard URL DB for storing a URL converted into a standard format by the URL standardization unit and a folder unique number of the URL. The method of claim 5, wherein the related URL extraction unit, A URL-related extracting unit that reads the URL and folder information generated by the URL standardizing unit and arranges them in a relevance ranking for each URL, and outputs <URL, degree of relevance> information for the top N arranged; And And a related URL DB for storing information output from the URL related extracting unit. The method of claim 5, wherein the URL header construction unit, If the first URL headword is not created in the related URL DB built by the related URL extractor or if the first URL heading is to be reconfirmed after a certain period of time has elapsed, the second URL heading is read by reading the URL on the Internet. A URL header extractor; And And a URL heading DB for storing the second URL heading. The method of claim 5, wherein the service provider, And a web server which receives a current URL from a user and provides URL information related thereto to the user.