WO2000055741A1 - Signets synergiques d'acces a internet combinant la recherche sur internet et la gestion d'hyperliens - Google Patents

Signets synergiques d'acces a internet combinant la recherche sur internet et la gestion d'hyperliens Download PDF

Info

Publication number
WO2000055741A1
WO2000055741A1 PCT/US2000/004588 US0004588W WO0055741A1 WO 2000055741 A1 WO2000055741 A1 WO 2000055741A1 US 0004588 W US0004588 W US 0004588W WO 0055741 A1 WO0055741 A1 WO 0055741A1
Authority
WO
WIPO (PCT)
Prior art keywords
link
links
folder
users
user
Prior art date
Application number
PCT/US2000/004588
Other languages
English (en)
Inventor
David Siegel
Original Assignee
Blink.Com, Inc.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Blink.Com, Inc. filed Critical Blink.Com, Inc.
Priority to AU35001/00A priority Critical patent/AU3500100A/en
Priority to JP2000605902A priority patent/JP2002539559A/ja
Publication of WO2000055741A1 publication Critical patent/WO2000055741A1/fr

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/955—Retrieval from the web using information identifiers, e.g. uniform resource locators [URL]
    • G06F16/9562—Bookmark management

Definitions

  • the present invention relates generally to access of internet information and more particularly to facilitating automatic management of access to web pages.
  • the Internet World Wide Web or simply "web," is a place where people store and distribute information.
  • the usefulness of the web depends a great deal on people's ability to locate material relevant to their interests.
  • the amount of information available on the web has grown dramatically resulting in increasing difficulty of finding information only relevant to a particular topic of interest.
  • the web is essentially a collection of linked content pages that can be imagined to resemble a spider's web.
  • the usefulness of the web is, to a large extent, determined by how easily information stored in one place can be located by someone in another.
  • Information stored on a Web page is typically accessed through a program called a web browser (e.g., Netscape or Internet Explorer) via a "Universal Resource Locator" or "URL".
  • the URL is commonly referred to as a "web address", a "hyperlink”, or simply a "link”.
  • An example of a URL is http://www.ibm.com/.
  • Web pages can further include embedded hyperlinks which, if selected in some way (e.g., by clicking with a mouse), will transfer the browser to the corresponding page in similar fashion.
  • search engines that catalog all web links and allow a user to perform a search for some specified content.
  • search engines like AltaVista and Infoseek, which automatically and continuously scan the Web for all content and build a database of addresses and keywords. A user can search the AltaVista database for potentially relevant content.
  • These automatic web search engines perform textual matches against a set of keywords that a user provides in their query. Pages that simply contain the text, not necessarily pages that are relevant to the concept sought for, are found.
  • a resulting problem is that, in an effort to generate site hits, many Web page authors attempt to fool search engines into returning their pages as a match for a search when the user performs a common search for something else. Most users of search engines have experienced this problem.
  • a web site for a book store might insert words in their home page like "white house", in hope of getting their link displayed when someone performs a search trying to find the home page of the White House.
  • Sites such as Yahoo and the Mining Company maintain large collections of organized links which are manually compiled and grouped into categories and made available to users.
  • bookmarks This often results in disorganized and hard to use personal bookmark collections when all the bookmarks in left one unorganized grouping. At some later point, they may sort the bookmarks and put them into related folders. For example, when using the bookmark feature in Internet Explorer or Netscape, a user has to search for the right folder to file a new link in the (potentially large) collection of folders that the user has already created. If a relevant folder cannot be found, the user must determine an existing folder that a new subfolder should be created in to store the new link. As the user develops a larger bookmark collection, it will take increasingly more time to organize new entries and determine the best place to store the link.
  • Wide Web links also known as URLs, or universal resource locators
  • the system of the present invention stores web links for multiple users in a database and provides methods for extracting and displaying the web links, methods for finding web links that are related to existing links that a user has stored in the system, and other related features.
  • the system and method of the present invention allows users to organize and manage collections of links and to find related links in other user link collections.
  • the database matching feature can use other information for finding relationships, such as profile information on the user, including, e.g., age, gender, and occupational profession. In essence, all participating users of the system help to organize a web directory.
  • the present invention produces lists of links that are self-organizing into clusters of related sites based upon the collective organizational efforts of all users of the system. Unlike conventional systems, it is neither keyword nor textual searching based. The information implicit in an individual's clustering of related web links is used to create a searching and clustering mechanism that allows users to find content that is related to existing content that they have already found.
  • the present invention offers ease of use and does not require users to make explicit categorization decisions. This approach creates a self-organizing web directory that takes advantage of the experiences of all participating users.
  • the self organizing link list of the present invention allows users of the World Wide Web to organize their own web links to content that they find interesting and to locate interesting and relevant content with current search engines; allows more efficient link management; are more specific, and less prone to false matches than the results of conventional keyword-based search engines; are formed without manual intervention; reflect the actual preferences and usage patterns of users, and are relatively free from the editorial bias or knowledge limitations of the human editors of catalog search sites; and, reflect the collective efforts of other users who have found interesting relevant content and provides a systematic approach to share their findings.
  • Various embodiments of the invention may offer one or more of these advantages.
  • Figure 2 illustrates a flowchart of the main processing loop according to the preferred embodiment of the present invention
  • Figure 3 illustrates an exemplary main screen according to the preferred embodiment of the present invention
  • Figures 4A and 4B illustrate user definition and folder mapping database schema under the preferred embodiment of the present invention
  • Figure 5 illustrates a folder matrix derived for an illustrative number of links under the preferred embodiment of the present invention
  • Figure 6 illustrates a link scoring table demonstrating link scoring under the preferred embodiment of the present invention
  • Figure 7 illustrates a link relation result table derived under the preferred embodiment of the present invention
  • Figure 8 illustrates a flowchart of a preferred implementation of the
  • FIG 9 illustrates a flowchart of a preferred implementation of related link processing of the present invention.
  • a central server computer 10 preferably via a local web browser (e.g., Netscape or Internet Explorer).
  • the central server computer 10 accesses data storage device 20 which includes a 0 database.
  • users will access the present invention by a PC via the public Internet.
  • implementations of the present invention can take place over a private Intranet or the implementation can fully avail itself of alternative technologies for web users access, e.g., cellular 5 telephone access, Web TVTM etc.
  • the system can, in alternative embodiments, be operated on more than one central computer and more than one database.
  • Conventional software solutions are utilized to keep the multiple copies of the database synchronized, and to distribute the load of web Q traffic among more than one server.
  • the "favorites" or “bookmarks” function of the web browser is centralized at the single common site 100. All users of the service store their bookmarks at the common central site 100, rather than using the decentralized (local) bookmark feature of their own browsers.
  • bookmark collections or lists created by individual users and stored in database 20 at the central site are periodically analyzed through comparison of list pairings. When two lists are found that are highly correlated, those two lists are deemed related and a new, centralized list is generated as a ranked, complete or partial union of the two original lists. Other users of the system, as well as the owners of the two original lists, will now access and use this new centralized list. It gives a more comprehensive list of related sites than either of the original lists alone.
  • This process allows users of the Internet to search for and group URL bookmarks (links) into a hierarchical set of folders that are stored on a central server and that can be accessed from any web browser on the Internet. Based on the collective results of individual user folder groupings, relationships between links are automatically found, helping users to locate relevant information on the Internet.
  • the process will propose links to the user that are similar to the links that the user has previously collected in a folder.
  • Other links will be identified and rank ordered links on the basis of a measure of similarity based on a number of criteria including the number of other users that grouped them in a similar way, a distance metric that relates links by intersecting the link content of user folders, user provided ratings, and by profile information on the user's background.
  • Web users access the present invention by connecting to a designated web site.
  • An illustrative example of the main system flow path according to the preferred embodiment is shown in Figure 2.
  • User login 200 initiates the flow where a username and password is either entered upon query into the system directly from the web based application, or passed automatically via "cookies" or other browser provided features.
  • the system validates 202-204 this information against the user account database. Where the user is not a registered user, a registration procedure is undertaken 203, the user is registered and a user account is created 213 and the flow continues to 206. In the event of an invalid username/password combination, an error message is posted 205 and the user is asked to start again at user login 200.
  • the system starts 206 a new web session for the user at the user's home page, and generates (or displays) 208 a custom home page directory that lists the top-level links and directories set associated with the user.
  • the displayed page (and all subsequent pages) in this embodiment contains links 30 (see Figure 3 infra) or buttons 32 (see Figure 3 infra) that allow the user to perform various link-related functions.
  • the user can click on a link (i.e., a bookmark) 210 to go to the corresponding web page 211 and update a hit count 212 corresponding to that site; or the user can click on a folder 214 to display the corresponding folder content 215, which may include more links and folders; or, the user can alternatively click on a given set of commands 216 to, e.g., perform 217 the corresponding operation such as: navigating to a subdirectory (if one exists), and moving back to a previous directory in the directory hierarchy; adding a new link to the current directory; and, deleting a link from the current directory.
  • a link i.e., a bookmark
  • This organization of links is similar to a standard computer file system, organized into a hierarchical structure of files and folders. While similar to the approach used in many web-based directories, such as Yahoo, in a site like Yahoo, there is one web directory which is the same for everyone. It is the only directory that a Yahoo user can navigate. To the contrary, when a user clicks on a link in the system of the present invention, the system records in an entry for the selected link in the user's database the date and time, and a counter is incremented to keep track of the total number of times the user has clicked on the link. That same information is then added to an aggregated database with the pooled information of all users for that link. Thus, the system keeps track of the total number of visits to a link for each individual and also for all users. Next the system will transfer web control to the selected page and the web browser will display the selected page (possibly in a new window).
  • FIG. 3 An example of a screen layout according to this embodiment of the present invention is illustrated in Figure 3.
  • the main screen will provide various links 30 and buttons 32 implementing the various features discussed with respect to Figure 2 which links and buttons can be clicked to, e.g., go to a link or add new links and folders.
  • the "add new link” screen accessed by clicking the Add button 32 has fields in which to enter the name of the link, a description, and a rating.
  • the "add new folder” field allows the user to specify the name of a folder, a description, and to indicate if the folder is "public". Public folders can be referenced and viewed by other users, and provide a method for sharing links. The system will track the number of users that have referenced each public folder to determine, among other things, the popularity of the folder.
  • an optional browser “add on” allows automatic insertion of the currently displayed page as well.
  • the system checks if any other users have previously added the link. If not, the system creates a new entry in the "database table" for the link and assigns to the link a unique "link id". The system then inserts an entry for this link into the current user link directory that is being displayed. Internally, the link will be referred to by the "link id", rather than with the text URL. This allows the system to easily identify all users that have referenced any link, along with other features to be later described. Once the link has been added, the hit count and the last access dates and times are initialized to 0 and "never" respectively.
  • Figure 4 A illustrates an example of a user account database table for tracking users. Users of the system are given unique accounts that contain descriptive information about them. The information is stored in a database. While several categories of user information are shown in Figure 4 A, the User Definition Table, other information can be dynamically added to this table to help classify the user as desired.
  • a database table as illustrated in Figure 4B Folder Mapping Table maps folders to users. This user and folder information are used in the search and matching processes described infra.
  • a search facility that finds links in the folders that the user has created.
  • the search capability makes use of the histories of data, including the last date the link was referenced, the number of times the link has been accessed by the user and by all users of the system, the user-provided rating, and the descriptive textual information.
  • the search capability also uses information obtained from other users who have added the same links to their collections. For example, a search can be performed to identify links from the user's collection that have the highest overall hit count across all users of the system. This would identify the most heavily used links in the user's collection, system wide, which provides an indication that the links are in general interesting.
  • One of the important features of the present invention is an automatic categorization technique that groups together similar links. Links are collected in folders with arbitrary names that are created by each user. (A folder can, in turn, contain subfolders.) This is accomplished by taking advantage of the information that is collected implicitly when a user groups a set of links together into a single folder.
  • the automatic categorization technique facilitates locating information on the Internet that might otherwise be difficult or impossible to find, in essence, by taking advantage of the collective body of users categorizing decisions.
  • the categorization process relies only on the placement of links into a particular folder. Users do not assign categorizations to the links themselves.
  • categorization information implicit in the folder groupings overcomes a number of major shortcomings of conventional approaches based upon users manually assigning categories to a link.
  • One such problem is that users will select different names for the same category.
  • One person might assign a link for an online bookstore to the "shopping" category, another to "bookstores", and yet another to "bookstore”.
  • Non-English speaking users would likely select category names in their primary language. All of this would make it hard to relate categories together.
  • getting a user to assign a category to a link requires a time consuming process that many users may choose not to perform.
  • the implicit categorization approach based on the folder the user has placed the link into overcomes this problem.
  • a folder matrix (as illustrated in Figure 5) represents the collection of links stored in a particular folder.
  • the folders are represented by the columns of the table, and are labeled with letters A-F.
  • the links in the system are represented by the rows, and are labeled with numbers 1-7.
  • An "X" in a cell of the matrix indicates that the folder in the checked column contains the link in the checked row.
  • folder "A” contains links "1" and "2", as noted by the check marks that appear in cells Al and A2.
  • the link number in this table refers to a unique link ID present in the link database, in which additional information on that link is stored.
  • the folders in the Figure 5 "Folder Matrix" refer to folders created and maintained by one or more users of the system.
  • folders "A” through “F” do not necessarily belong to one particular user.
  • the ownership information for these folders is stored in a database table.
  • Al is referred to as link 1 in folder A.
  • each user of the system (identified by a unique user account id) has a collection of such folders and links.
  • the overall approach to using these collections globally to categorize the web links stored in the system is based on a rating of the similarity of folders. Similar folders are defined to be those that have similar link collections. For example, if many users are collecting links on vegetable gardens, it is to be expected that there will be overlap among the vegetable garden links that such users have collected. Not all collections of vegetable garden links will be the same. Some users will have links that most, if not all, of the other users will not have in their own folders.
  • the system searches the link folders and, when presented with an exemplar folder, will find other links that are likely to be related to the links in the exemplar folder.
  • the folders in which different users of the system have stored links provides information as to the relationships between the links. For example, link 1 is related to link 2 because link 1 appears together with link 2 in both folders A and B.
  • the users of the system who created folder A and B found it desirable to organize links 1 and 2 together, in the same folder. Since link 1 and 2 both appear together in folder A and B, the system considers them to have a direct relationship between each other. Links can also be related to each other via folders indirectly.
  • link 1 and link 4 In this case, there is no particular folder that contains both link 1 and link 4 (as there is no column in the table in Figure 5 that has an "X" for both those links). However, link 1 and 4 have an indirect relationship via link 3 in folder D, which contains link 4. Folder D also contains link 3. Link 3 is directly related to link 1 via folder B. Thus, link 1 and 4, via link 3, have a relationship between them. The creator of folder B had reason to organize links 1 and 3 together, to capture some relationship between them. The creator of folder D had some reason to related links 4 and 3. Thus, there is evidence of an indirect relationship between links 1, 3, and 4.
  • a user can have a folder with links on to sites with content related to vegetable gardens. There are two links in the folder, one for a web page on carrots, and another on tomatoes. A second user has a collection of links in a folder on carrots and peas. A third user has a collection of links on peas and peppers. In this collection of folders, no individual folders can relate the pepper link to the carrot link, since there is no a single folder containing both of these links. On the other hand, a path can be traced between carrots and peppers by the connection between carrots and peas in one of the folders. This indirect connection is not as strong as a direct relationship in which the links are all in one folder. Thus, the scoring of the relationship between the links in different folders will be smaller than the score of links in the same folder.
  • One important use is to enable users to find additional links that are related to links that they have already collected and stored in a folder.
  • the user from the previous example may have created a folder that has a set of links that are about "gardening” or the folder may have more specific links, such as "vegetable gardens”.
  • the information is gathered by searching all link folders in the system, finding ones that appear to be similar to the user's "gardening" folder (the "exemplar folder"), and the best matches are presented.
  • the same searching approach is used to find the best matching "public folders" to a user's exemplar folder.
  • the resulting matches are ordered by the popularity of the public folder, based on how many other users have connected it into their folder tree. If the user finds a particular public folder link collection useful, they click on the appropriate button that adds it, as a read-only 0 folder, to their link collection. Any changes that the folder owner makes to this folder are immediately reflected in all read-only copies.
  • the user may enable a feature that sends an email message whenever the owner of the public folder link collection makes additions or changes.
  • sort button is be pressed, and the system automatically proposes the best folders in his or her personal link collection to place each unsorted link.
  • the present invention searches the global database, looking to see if other users have put that link in a folder. If matches are found, the folder into which the link was placed is compared with the entire folder collection in the user's personal directory,
  • the best matching folder is then proposed as the best location for storing the link. If the link to be sorted is found in more than one folder in the global directory, each folder is matched against the user folder collection, and a consensus proposed file location is
  • a scoring function scores the strength of the relationship between links.
  • a scoring table is illustrated in Figure 6 which represents the scoring process for the relationship between link 1 in folder A and other links (2-7).
  • ⁇ factors are used, including the degree of directness of the relationship (the number of folders that are required to connect one link to another), the number of folders that capture the relationship, and the frequency of use of links in the relationship path.
  • the same scoring function is used to assess the degree of indirect relationships, such as the one between links 1 and 4, via link 3 in folders B and D.
  • the degree of directness is a factor in the scoring function.
  • the scoring function lowers the score as function of the length of the path.
  • Different well known approaches can be used, including using a linear discount or an exponential decay. The exponential approach accelerates the reduction in score as the path linearly increases in length.
  • Another factor used in the scoring function is to discount paths between links that users are not frequently accessing or that have not recently been accessed and to overweight paths with links that have been frequently and or recently used. For example, consider the connection between links 1 and 4 is via link 3 in folder B and D (see Figure 5 "Folder Matrix"). The score of this connection is reduced if the owners of folders B and D have not recently or frequently used link 3.
  • a profile correlation function provides a measure of the degree of the relation between two user profiles. For example, two users with profile information indicating that they are physicians aged 42 and 49 respectively living in the United States are statistically more likely to have link collections that are related than a 42 year old American doctor and a 20 year old farmer from China. Thus, overlapping links found in the folders of more related users are scored higher than the same links in folder of users that are not as related. A consequence of this approach is that matches are proposed which are customized to the profile of the particular user.
  • the system allows a user to substitute alternative profile information to explore how other people group links together. A young physician with a collection of links on coping with arthritis wishes to the arthritis link collections that older people have collected. The system allows the physician to substitute his or her profile with that of the target user, and to find the best matching links given that target profile.
  • Folder Matrix "Folder Matrix"). There, link 3 appears in Folder B, which also contains all the links in Folder A. Thus, Folder B contains all the links in Folder A plus link 3.
  • link 3 is related to the links in folder
  • a with a lower score of 0.84 The reason that the score is lower is that the connection between link 7 and folder A is indirect, and requires a path through a number of common folders (rather than by a the direct connection between link 3 and folder A). None the less, since there is some connection, the score is greater than 0.
  • the order of presentation is based on the score, and larger scores are presented first. There is an arbitrary score cutoff, so that only links where score is sufficiently large are presented. Alternatively, links are presented in decreasing order of score, and the user decides when to stop viewing lower scored links.
  • the present invention determines how closely a particular link is related to the links in a given folder.
  • a variation of this approach is used to determine how much a particular folder (which contains a collection of links) is related to another folder.
  • the system finds closely matching folders in the database.
  • the scoring process is applied to determine the score of each link to the exemplar folder.
  • the overall scores for the candidate folder are combined together into a single score for the ° entire folder.
  • the average score of the candidate folder is calculated. When this has been done for all folders, the highest overall average scores represent the folders that match most closely to the exemplar folder.
  • a weighted average score is calculated which assigns higher values to folders with large numbers of high scored links and lower values to folders with smaller numbers of related links.
  • FIG. 7 Illustrated in Figure 7, is an example of determined inter-folder relationship.
  • the example table indicates the shortest path via common, shared, links between folders.
  • folder A is connected to folder F by a 3- link path.
  • the connection between folder A and folder F is indirect. 0
  • FIG. 8 illustrates a flowchart of the matching function of the 5 present invention for matching related bookmarks.
  • step 800 all appropriate variables are initialized (e.g., match count, ⁇ ).
  • match count e.g., ⁇
  • ⁇ e.g., ⁇
  • Related bookmarks are added to / ' from the related table to result set.
  • Step 804 The match count for related bookmarks is incremented.
  • Step 806 The total number of matches are recorded.
  • Step 808 A 0 check is made to see if there are any remaining bookmarks in the folder. (Step 810). If yes, the process loops back to step 804 to process the next bookmarks.
  • Step 812) the results are displayed (Step 814).
  • 5 Shown in Figure 9 is an illustrative flowchart for related link processing under the present invention.
  • a first bookmark folder is fetched (902). If the end of file is reached (904), then terminate (912). All pairs ( i,j) of bookmarks in the instant folder are looped through (906), o incrementing the relationship count between the pair (908). This is performed for all the bookmarks in the particular folder. (910). When all bookmarks in one folder have been processed, the program goes to process the next folder (902) until finished.
  • the system of the present invention thus helps to overcome deficiencies of conventional automatic web search engines discussed above, as matching is done by a consensus of users associating links in folders. Even if one user put the book store link along with a collection of links on the ⁇ United States Government, the weighting function used rejects that association
  • the automatic folder-based categorization and private user link directories are used to help in the creation of a comprehensive web directory.
  • users file links in their own personal folder hierarchy.
  • the matching and categorization process uses this information to find related links and folders.
  • This basic mechanism is extended to form a mechanism for creating a 0 global web link directory. The process works as follows. Effectively, all of the private user folders are intersected to form a master folder and link directory.
  • the folder matching process is used to determine which master folder is the best place in which it should be stored. This is accomplished by searching all private user folders that contain the link. As users add the link to their own private collection of links, the system "learns" which other links are most closely related on the basis of the other links in the same folder. The more private user folders that contain this link, the more data there is available for finding related links. With this information, the process searches the master link directory to find the folder that best matches the private user folders that contain the link. The link is added to the one or more master folders that contain the closest related links.
  • the basic matching process also is used to suggest the folder hierarchy ordering of the master link directory. That is, it determines the folders into which a subfolder is place. The result is a tree like structure for the storage of links.
  • a folder called “gardening” contains subfolders called “vegetable” and “flowers”
  • the problem becomes, how can the master link directory be organized this way, based on the private organization structures of individual users?
  • the match is used to determine the user link folders that best match a master link folder.
  • the matching is performed across all users, and then the user private link folders are examined. Across many user link collections are found that, e.g., match the flower link folder and in most case the vegetable link folder, that are contained in a common parent folder. Thus, the system scans users private link folders to learn this taxonomy.
  • a payment incentive scheme can be implemented at the central server in an alternative embodiment.
  • the system tracks the number of users that reference a public link folder. With this reference count mechanism, along with access statistics (how frequently people are clicking on links in a public folder), formulas have been developed to rank the popularity of a public folder.
  • a simple approach is to rank the folders based on the number of read-only references. The more read-only references from other users, the higher the ranking. The highest ranking folder creators could receive payment for their contributions made to the overall link collection.
  • This payment scheme would also be useful for promoting the service, as people would spread the word about their own personal link collection, to increase the amount of payment that they would receive for their collection.
  • a modification to the simple reference counting scheme for determining payment is to have a blending function that included both total number of references and the number of clicks on a collection over a period of time.
  • a linear combination of the two parameters is used, with the scale factors selected based on the total payment funds available.
  • a user creates a folder of links on "gardening".
  • the system finds other users that have such a link collection, and invites them to join a dynamically created discussion group of all users with similar gardening folders.
  • new people in the overall community of users create folders that match the gardening folder, they too are invited to join the group.
  • This process for inviting people to join the community described overcomes privacy concerns. Users are asked to join the community.
  • discussion groups are optionally supported on any public folder. If a public folder owner chooses to do so, he or she may enable the discussion feature. This allows anyone accessing the public folder to post a message that will be displayed along with the public folder links. Since it is likely that the group of people accessing the public folder share the interests reflected by the content referenced by the links in the public folder, this facilitates the creation of a community that evolves around that specific interest.
  • Alternative uses in varying embodiments include relating and scoring links for targeting web advertising. To achieve this, a potential advertiser creates a collection of links that is believed to be representative of the product or service that is to be marketed. For example, if an advertiser were selling gardening products, a collection of gardening links is gathered and placed in an exemplar link folder.
  • a link to the advertiser's web site is then automatically inserted into the user's link collection, and appears whenever the user enters that folder.
  • Multiple advertiser links may be inserted into the same folder, permitting all garden advertisers to insert a reference to the web site's URL in all gardening-related web folders in the system.
  • the same approach is used to display a banner ad at the top of a page whenever a user enters a folder that matches an exemplar from an advertiser.
  • a user can insert a dynamic hot list on any web page thereby creating embedded link lists.
  • An exemplary Java implementation of the embedded link list feature is attached as Appendix III.
  • An access module that runs on a remote computer makes a database connection to a central server. This server returns the contents of a link folder to the remote application. The folder is then displayed on the web page as a hot list. This approach makes the hot list dynamic. As additions or changes are made to the folder, the next time the web page is accessed, the new additions or changes are displayed.
  • the hot list is supplemented by other similar links, and includes not only the links in the particular folder, but closely related ones as well.
  • Folder f new Folder (p, c) ;
  • Folder f new Folder () ; f .addLmks (s) ,- return addFolder (f) ;
  • Integer link (Integer) elems nextElement ( ) , if ('exemplar contains (link) ) results [link mtValue ( ) ] ++ ,
  • switch parser ttype
  • case StreamTokenizer ⁇ switch (parser ttype) ⁇ case StreamTokenizer .
  • TT_EO break; case StreamTokenizer .
  • TT_NUMBER
  • public boolean contains ( Integer link) ⁇ return _folder . contamsKey (link) ; ⁇ .
  • Enumeration elems _folder . keys () ; while (elems . hasMoreElements () ) ⁇
  • Hashtable _ht public RelatedUrlGenerate () ⁇
  • Item item ( Item) elems . nextElement () ;
  • EmbedLinks e new EmbedLinks (_session) ;

Landscapes

  • Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

Dans diverses réalisations, l'invention se rapporte à un système et à un procédé de gestion et de classification de liens du réseau Internet (également appelés URL, ou localisateur de ressources universel) qui permettent de créer des répertoires publics et personnels de ces liens dans le but d'améliorer la navigabilité sur le réseau Internet. Dans diverses réalisations, le système de la présente invention stocke des liens du Web pour de multiples utilisateurs dans une base de données et fournit des procédés d'extraction et d'affichage de ces liens du Web, des procédés de recherche des liens du Web qui sont reliés à des liens existants qu'un utilisateur a stocké dans le système, ainsi que d'autres caractéristiques associées (804). Les utilisateurs peuvent organiser et gérer des collections de liens et rechercher des liens connexes dans des collections de liens d'autres utilisateurs (808). La caractéristique de mise en correspondance de bases de données peut utiliser d'autres informations pour rechercher des relations, telles que des informations de profils relatives à l'utilisateur et comprenant, par exemple, l'âge, le sexe et le type d'activité professionnelle.
PCT/US2000/004588 1999-03-18 2000-02-23 Signets synergiques d'acces a internet combinant la recherche sur internet et la gestion d'hyperliens WO2000055741A1 (fr)

Priority Applications (2)

Application Number Priority Date Filing Date Title
AU35001/00A AU3500100A (en) 1999-03-18 2000-02-23 Synergistic internet bookmarks combining internet searching and hot linking
JP2000605902A JP2002539559A (ja) 1999-03-18 2000-02-23 インターネット検索とホットリンクを結びつける、相乗作用を生むインターネットブックマーク

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US12504899P 1999-03-18 1999-03-18
US60/125,048 1999-03-18

Publications (1)

Publication Number Publication Date
WO2000055741A1 true WO2000055741A1 (fr) 2000-09-21

Family

ID=22417978

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2000/004588 WO2000055741A1 (fr) 1999-03-18 2000-02-23 Signets synergiques d'acces a internet combinant la recherche sur internet et la gestion d'hyperliens

Country Status (3)

Country Link
JP (1) JP2002539559A (fr)
AU (1) AU3500100A (fr)
WO (1) WO2000055741A1 (fr)

Cited By (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2367661A (en) * 2000-03-09 2002-04-10 Ibm Managing objects
FR2815437A1 (fr) * 2000-10-18 2002-04-19 Guilhem Giraud Dispositif informatique pour gerer de maniere centralisee les pages de liens hypertextes d'entreprises possedant un site sur le world wide web
WO2002035386A1 (fr) * 2000-10-27 2002-05-02 Catherine Brexel Annuaire personnel interactif
JP2002203170A (ja) * 2000-12-28 2002-07-19 Wildcard Kk 通信回線を介して広告情報を閲覧可能とする方法、サーバー、コンピュータプログラム、及び、コンピュータ読み取り可能な記録媒体
JP2002203169A (ja) * 2000-12-28 2002-07-19 Wildcard Kk 通信回線を介して広告情報を閲覧可能とする方法、サーバー、コンピュータプログラム、及び、コンピュータ読み取り可能な記録媒体
JP2002203171A (ja) * 2000-12-28 2002-07-19 Wildcard Kk 通信回線を介して広告情報を閲覧可能とする方法、サーバー、コンピュータプログラム、及び、コンピュータ読み取り可能な記録媒体
DE10110924A1 (de) * 2001-03-07 2003-04-30 Deutsche Telekom Ag Verfahren zum Bereitstellen wenigstens einer personalisierten Web-Benutzerseite
GB2385159A (en) * 2002-02-07 2003-08-13 3G Lab Ltd Classification of access objects for content provision to a mobile
US7260600B1 (en) * 2000-11-17 2007-08-21 International Business Machines Corporation User specified parallel data fetching for optimized web access
US7363035B2 (en) 2002-02-07 2008-04-22 Qualcomm Incorporated Method and apparatus for providing content to a mobile terminal
US8364718B2 (en) 2008-10-31 2013-01-29 International Business Machines Corporation Collaborative bookmarking
US9003272B2 (en) 2010-09-16 2015-04-07 International Business Machines Corporation Method and system of a user associating a first webpage link and second webpage link and viewing of the contents of the webpage links by the selection of the first webpage link
WO2017001945A1 (fr) * 2015-06-30 2017-01-05 Yandex Europe Ag Procédé et système de détection d'activité de pourriels dans un système en nuage

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2004272329A (ja) 2003-03-05 2004-09-30 Nec Corp 携帯端末装置
JP4718291B2 (ja) * 2005-10-03 2011-07-06 株式会社オービック 情報共有システム

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5895461A (en) * 1996-07-30 1999-04-20 Telaric, Inc. Method and system for automated data storage and retrieval with uniform addressing scheme
US5956027A (en) * 1995-12-12 1999-09-21 At&T Corp Method and apparatus for sharing a web page
US6023701A (en) * 1997-09-25 2000-02-08 International Business Machines Corporation Skeleton page retrieval mode for web navigation
US6052717A (en) * 1996-10-23 2000-04-18 Family Systems, Ltd. Interactive web book system
US6052687A (en) * 1997-06-13 2000-04-18 Fujitsu Limited Relational database search system and method, and intermediate link table generating method and storage medium
US6064979A (en) * 1996-10-25 2000-05-16 Ipf, Inc. Method of and system for finding and serving consumer product related information over the internet using manufacturer identification numbers

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5956027A (en) * 1995-12-12 1999-09-21 At&T Corp Method and apparatus for sharing a web page
US5895461A (en) * 1996-07-30 1999-04-20 Telaric, Inc. Method and system for automated data storage and retrieval with uniform addressing scheme
US6052717A (en) * 1996-10-23 2000-04-18 Family Systems, Ltd. Interactive web book system
US6064979A (en) * 1996-10-25 2000-05-16 Ipf, Inc. Method of and system for finding and serving consumer product related information over the internet using manufacturer identification numbers
US6052687A (en) * 1997-06-13 2000-04-18 Fujitsu Limited Relational database search system and method, and intermediate link table generating method and storage medium
US6023701A (en) * 1997-09-25 2000-02-08 International Business Machines Corporation Skeleton page retrieval mode for web navigation

Cited By (29)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2367661B (en) * 2000-03-09 2004-11-24 Ibm A method and system for managing objects
GB2367661A (en) * 2000-03-09 2002-04-10 Ibm Managing objects
FR2815437A1 (fr) * 2000-10-18 2002-04-19 Guilhem Giraud Dispositif informatique pour gerer de maniere centralisee les pages de liens hypertextes d'entreprises possedant un site sur le world wide web
US7020646B2 (en) 2000-10-27 2006-03-28 Catherine Brexel Interactive personal directory
FR2816081A1 (fr) * 2000-10-27 2002-05-03 Catherine Brexel Annuaire personnel interactif
US7810023B2 (en) 2000-10-27 2010-10-05 Catherine Brexel Process for generating personalized digital directory pages
WO2002035386A1 (fr) * 2000-10-27 2002-05-02 Catherine Brexel Annuaire personnel interactif
USRE45730E1 (en) 2000-10-27 2015-10-06 Catherine Brexel Process for generating personalized digital directory pages
US7260600B1 (en) * 2000-11-17 2007-08-21 International Business Machines Corporation User specified parallel data fetching for optimized web access
JP4603682B2 (ja) * 2000-12-28 2010-12-22 ワイルドカード株式会社 通信回線を介して広告情報を閲覧可能とする方法、サーバー、及び、コンピュータ読み取り可能な記録媒体
JP2002203170A (ja) * 2000-12-28 2002-07-19 Wildcard Kk 通信回線を介して広告情報を閲覧可能とする方法、サーバー、コンピュータプログラム、及び、コンピュータ読み取り可能な記録媒体
JP4603683B2 (ja) * 2000-12-28 2010-12-22 ワイルドカード株式会社 通信回線を介して広告情報を閲覧可能とする方法、サーバー、及び、コンピュータ読み取り可能な記録媒体
JP2002203171A (ja) * 2000-12-28 2002-07-19 Wildcard Kk 通信回線を介して広告情報を閲覧可能とする方法、サーバー、コンピュータプログラム、及び、コンピュータ読み取り可能な記録媒体
JP2002203169A (ja) * 2000-12-28 2002-07-19 Wildcard Kk 通信回線を介して広告情報を閲覧可能とする方法、サーバー、コンピュータプログラム、及び、コンピュータ読み取り可能な記録媒体
JP4603681B2 (ja) * 2000-12-28 2010-12-22 ワイルドカード株式会社 通信回線を介して広告情報を閲覧可能とする方法、サーバー、及び、コンピュータ読み取り可能な記録媒体
DE10110924B4 (de) * 2001-03-07 2004-01-15 Deutsche Telekom Ag Verfahren und Computer-Lesbarer Datenträger zum Bereitstellen wenigstens einer personalisierten Web-Benutzerseite
DE10110924A1 (de) * 2001-03-07 2003-04-30 Deutsche Telekom Ag Verfahren zum Bereitstellen wenigstens einer personalisierten Web-Benutzerseite
US7363035B2 (en) 2002-02-07 2008-04-22 Qualcomm Incorporated Method and apparatus for providing content to a mobile terminal
GB2385159B (en) * 2002-02-07 2004-03-24 3G Lab Ltd Providing content to a mobile terminal
US8428564B2 (en) 2002-02-07 2013-04-23 Qualcomm Incorporated Method and apparatus for providing updated content data to a mobile terminal
GB2385159A (en) * 2002-02-07 2003-08-13 3G Lab Ltd Classification of access objects for content provision to a mobile
US8364718B2 (en) 2008-10-31 2013-01-29 International Business Machines Corporation Collaborative bookmarking
US9003272B2 (en) 2010-09-16 2015-04-07 International Business Machines Corporation Method and system of a user associating a first webpage link and second webpage link and viewing of the contents of the webpage links by the selection of the first webpage link
US9727540B2 (en) 2010-09-16 2017-08-08 International Business Machines Corporation Method and system of a user associating a first webpage web link and second webpage link and viewing of the contents of the webpage links by the selection of the first webpage link
US10095667B2 (en) 2010-09-16 2018-10-09 International Business Machines Corporation Method and system of a user associating a first webpage web link and second webpage link and viewing of the contents of the webpage links by the selection of the first webpage link
US10503815B2 (en) 2010-09-16 2019-12-10 International Business Machines Corporation Method and system of a user associating a first webpage web link and second webpage link and viewing of the contents of the webpage links by the selection of the first webpage link
US11194955B2 (en) 2010-09-16 2021-12-07 International Business Machines Corporation Method and system of a user associating a first webpage web link and second webpage link and viewing of the contents of the webpage links by the selection of the first webpage link
WO2017001945A1 (fr) * 2015-06-30 2017-01-05 Yandex Europe Ag Procédé et système de détection d'activité de pourriels dans un système en nuage
US9787763B2 (en) 2015-06-30 2017-10-10 Yandex Europe Ag Method of and system for detecting spam activity in a cloud system

Also Published As

Publication number Publication date
JP2002539559A (ja) 2002-11-19
AU3500100A (en) 2000-10-04

Similar Documents

Publication Publication Date Title
US12001490B2 (en) Systems for and methods of finding relevant documents by analyzing tags
US6493702B1 (en) System and method for searching and recommending documents in a collection using share bookmarks
US6430558B1 (en) Apparatus and methods for collaboratively searching knowledge databases
US7421441B1 (en) Systems and methods for presenting information based on publisher-selected labels
US8612417B2 (en) Inverse search systems and methods
US6732088B1 (en) Collaborative searching by query induction
US8005835B2 (en) Search systems and methods with integration of aggregate user annotations
US7685209B1 (en) Apparatus and method for normalizing user-selected keywords in a folksonomy
US8862580B1 (en) Category-based search
EP1782380A2 (fr) Systemes et procedes de recherche utilisant des recherches contextuelles en ligne
US20030217056A1 (en) Method and computer program for collecting, rating, and making available electronic information
WO2000055741A1 (fr) Signets synergiques d'acces a internet combinant la recherche sur internet et la gestion d'hyperliens
Paepcke et al. Beyond document similarity: understanding value-based search and browsing technologies
WO2000048057A2 (fr) Moteur de recherche de signets
Li et al. Web Mining to Identify People of Similar Background
Mane et al. Oncosifter: A Customized Approach to Cancer Information

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A1

Designated state(s): AE AL AM AT AU AZ BA BB BG BR BY CA CH CN CR CU CZ DE DK DM EE ES FI GB GD GE GH GM HR HU ID IL IN IS JP KE KG KP KR KZ LC LK LR LS LT LU LV MA MD MG MK MN MW MX NO NZ PL PT RO RU SD SE SG SI SK SL TJ TM TR TT TZ UA UG US UZ VN YU ZA ZW

AL Designated countries for regional patents

Kind code of ref document: A1

Designated state(s): GH GM KE LS MW SD SL SZ TZ UG ZW AM AZ BY KG KZ MD RU TJ TM AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE BF BJ CF CG CI CM GA GN GW ML MR NE SN TD TG

121 Ep: the epo has been informed by wipo that ep was designated in this application
DFPE Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101)
ENP Entry into the national phase

Ref country code: JP

Ref document number: 2000 605902

Kind code of ref document: A

Format of ref document f/p: F

REG Reference to national code

Ref country code: DE

Ref legal event code: 8642

122 Ep: pct application non-entry in european phase