WO2001042983A1

WO2001042983A1 - On-line searching system and method

Info

Publication number: WO2001042983A1
Application number: PCT/US2000/021675
Authority: WO
Inventors: Ofer Digli; Alon Schwarz; Ori Levy; Nir Melamoud; Roy Eitan
Original assignee: Browseup, Inc.
Priority date: 1999-12-08
Filing date: 2000-08-04
Publication date: 2001-06-14
Also published as: AU6531700A

Abstract

An on-line searching system and method for performing personalized search for content objects (100, 101) using OLINKs (110, 120). The system locates content objects (100, 101) that are contextually related to the base content object.

Description

ON-LINE SEARCHING SYSTEM AND METHOD RELATED APPLICATIONS This application is a continuation-in-part of U.S. Application Serial No.

09/457,619, filed December 8, 1999 and a continuation-in-part of U.S. Npplication Serial No. 09/598,097, filed June 21, 2000, which are incorporated herein in their entirety.

FIELD OF INVENTION

The present invention relates generally to a database management system, and more particularly to a computer network searching system.

DESCRIPTION OF THE RELATED ART The World Wide Web (i.e., the Internet) is presently a two-dimensional information resource composed of pages that are connected to each other by links set by the originators of the pages. Since content is packed and presented in only page format, the resolution of information available to the users is low. The web users browsing or searching the Internet must read or sift through a page that may contain more irrelevant than relevant information.

The low level of resolution in content presentation is also reflected in the currently available search and rank engines. These products supply the user with search results in the form of pages, but not as content objects within the pages. They use variant techniques to analyze the main context within the pages in attempt to build an efficient indexed page database. Some of the products also include a rating system to establish relevancy of each page. Although a page is composed of many different pieces of information (or content objects), none of the presently available products separately rates the content objects within a page. That is, a user searching for a particular content on the

Internet is presented with a page rank rather than a ranking of the desired content within the page. The currently available search engines or systems utilize the index mechanism to categorize its collection of manually or automatically edited pages, but generally do not incorporate user's input in the search process. Some search engines, such as Direct Hit, collects information about the user by tracking user's browsing behavior, collecting votes or counting popularity of a particular web site, web page or document, but such information is generally used for ranking or rating a particular document. In other words, the search engines do not use such user information to search or associatively provide the user with related documents. Accordingly, it is desirable to provide a search engine that utilizes the user information, such as the current browsing location or currently displayed or viewed content, in searching for the desired and/or related content. Additionally, the number of pages that can be gathered and offered to the users by the presently available searching systems is limited. It is believed that only about one percent (1%) of the web content is indexed. This number may be significantly improved

if the web users are allowed to add, edit and control the indexing of the web content and determining the relationship between such web content. However, there exists no system that currently allows the web users to actively shape the content database by adding new content or linking two or more content objects. Therefore, it is desirable to provide a searching system that indexes the web content based on user's behavior or action to provide dynamically varying and associatively linked database. Furthermore, the current searching systems focus on a single content, site or page, evaluating or measuring its relevancy by matching keywords or popularity in usage, but such systems scarcely consider surrounding information or content. That is, the current searching systems generally ignore the relevancy or contextual connection between content objects in searching for a particular content object. For example, the currently available searching systems, such as google.com, scores or rates a particular page based on the number connections or links to that page, but does not enable users to search based on such connections or links. Therefore, it is desirable to provide a searching system that considers the contextually related content objects in locating a specific content object. The present invention proceeds upon the desirability of utilizing an associatively linked database to enable a user to search for content objects that are contextually related to a currently viewed or displayed content object. The present invention provides an online searching system that utilizes user's current browsing location, i.e., currently viewed web page or a specific content within the web page, to search for contextually similar

content objects.

OBJECT AND SUMMARY OF THE INVENTION

Therefore, it is an object of the present invention to provide a web-like or on-line database management and searching system and technique for performing a personalized search for content objects by using olink links by a user (or hereinafter referred to as

OLLNKs, as described in U.S. Patent Npplication Serial No. 09/457,619, filed December 8, 1999, entitled "OBJECT LINKING DATA MANAGEMENT METHOD AND SYSTEM"). Another object of the present invention is to provide an on-line searching system and technique as aforesaid, which searches for content objects that are contextually related to a currently displayed or viewed content object, referred to hereinafter as a base object. N further object of the present invention is to provide an on-line searching system and technique as aforesaid, which additionally limits the search to an area surrounding or adjacent to the base object, referred to hereinafter as a base object area.

N still further object of the present invention is to provide an on-line searching system and technique as aforesaid, which selects candidate direct and indirect OLLΝKs from the base object area and sorts the OLINKs in accordance with its relevancy to the base object.

In accordance with the present invention, the on-line searching system and method selects potential OLINKs from the base object area that are contextually related or connected to the base object. Preferably, the on-line searching system and method can limit the scope of the search based on one or more search criteria, such as time factor, maximum absolute distance, minimum link rate, average link rate, textual connection, and other limitations. The distance information represents the number of links between a content object and the base object. For example, layer 1 represents all content objects that are directly linked to the base object and layer 2 represents all content objects that are indirectly linked to the base object via another content object, i.e., two OLINKs must be traversed to reach the base object from a layer 2 content object. The link rate of an OLINK specifies or represents the relevancy of the content objects connected by the OLINK, i.e., the quality of such OLINK. In accordance with an embodiment of the present invention, the on-line searching system and method selects potential OLLNKs in accordance with one or more user defined filtering criteria, such as, but not limited to, content object type, OLINK's origination date, OLINK's originator identity, location information, OLINK's category. It is appreciated that the filtering criteria limits the scope of search executed and provides the user with more relevant results by filtering out the undesired OLLNKs.

In accordance with another embodiment of the present invention, the on-line searching system and method sorts the selected OLLNKs in accordance with its relevance to the base object. For example, the selected OLINKs may be ordered from most relevant (i.e., highest link rate) to the least relevant (i.e., lowest link rate). The relevancy or degree of relevancy can be measured using sorting parameters, such as the link rate, the distance information and the like, and/or sorting algorithms, e.g., measuring the degree of textual similarity between a key word and the words contained in the embedded data of the OLINK, measuring the degree of textual similarity between a key word and the content object associated with the OLINK or the like. Preferably, the results of the various sorting parameters are summed to sort or determine the order of the selected OLINKs. The user can preferably weigh the sorting parameters to give preference to one parameter over another parameter.

BRIEF DESCRIPTION OF THE DRAWINGS

The following detailed description, given by way of example, and not intended to limit the present invention solely thereto, will best be understood in conjunction with the accompanying drawings in which: Fig. 1 shows the linking between two content objects;

Fig. 2 shows the content objects and OLINKs arranged in a neural network structure;

Fig. 3 is a flowchart of the searching process in accordance with an embodiment of the present invention;

Fig. 4 is a block diagram of the locating and selecting process in accordance with an embodiment of the present invention;

Fig. 5 illustrates the searching process of the present invention based on the minimum link rate; Figs. 6A-6C illustrate the searching process of the present invention based on the minimum average link rate algorithm;

Fig. 7 illustrates the sorting process of the present invention based on the OLINK's accessibility; and

Fig. 8 is a block diagram of illustrating an embodiment of the database management system of the present invention.

DETAILED DESCRIPTION OF THE EMBODIMENTS

The present invention comprises of a novel on-line searching method and system

within a novel database management system. The present invention is ready application in virtually all computer networks, including, but not limited to, the World Wide Web (Internet), intranet, local area network (LAN), wide area network (WAN) and wireless network. The following description is presented to enable any person skilled in the art to make and use the invention. Descriptions of specific applications are provided only as examples. Various modifications to the preferred embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other embodiments and applications without departing from the spirit and scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown, but is to be accorded the widest scope consistent with the principles and features disclosed herein.

In accordance with an embodiment of the present invention, the web-like or online (network) database management system of the present invention can be located on a network computer or server, such as an Internet system server and the user can search for relevant content objects through a client application residing on user's personal computer or remote access device. The system server and the client application communicate over the well-known TCP/IP network protocol. The client application is plugged into a web browser, such as Netscape Navigator or Internet Explorer or other comparable browser. Preferably, the client application contains a software program that uses browser specific plugin module to hook into the browser, allowing the client to control and record the user browsing events and searching queries.

For example, Fig. 8 illustrates a block diagram of a computer network, specifically an Internet, which incorporates the on-line searching system of the present invention. A user client or client application 900 plugs in through a plugin module 904 of the Internet browser 903. It is appreciated that the client application 900 plugs into the browser 903 using a browser specific plugin module 904, which is a specific to each browser. The client application 900 includes a software program that uses browser specific plugin module 904 to hook into the browser 903, thereby allowing the client application 900 to control the browsing experience, e.g. the user can browse the Internet using the OLINKs. The client application 900 can control and record browsing events such as entering a URL or selecting a content object within a page, and other activities. The client application 900 communicates with the system server 901 via a protocol layer 905. The protocol layer 905 is a protocol designed to enable the database management system to operate over the well-known TCP/IP network protocol. The system server 901 contains global OLINKs database 906, which is an Internet OLINK database. The OLINKs stored in the Global OLINK database 906 is shared worldwide and is accessible as long as the network access exists. It is appreciated that the client application 900 and the system server 901 can access other Internet servers 902, including HTTP or FTP servers, in order to obtain content existing outside of the data management system. It is appreciated that accessing and utilizing other Internet servers 902 reduces the communication load on the system server 901 and storage space required to store content objects and OLINKs. The database management system can support multiple system servers through a distributed networking topology.

The database management system of the present invention includes content objects database and OLINKs database. The content objects database includes web objects that were originally inputted by users either from a web site or web page on the Internet or personal objects from a remote computer. In order to identify each of the content objects within the database, the database management system uses a location identifier and a content identifier as described in U.S. Patent Application Serial No. 09/457,619. If a new content object is added from a remote computer, the database management system assigns a unique URL to such personal object and saves it as an HTML web page. The database management system uses the unique URL to identify the personal object. It is appreciated that a web content object is identifiable by its original URL and/or other identification keys, such as text, graphic, or other algorithm that encodes the web content into a unique key. Since the database management system stores the URLs, the content object can be located in its original hosting server and retrieved using the location identifier and the content identifier. It is appreciated that for personal objects the original hosting server is the system server.

The searching system of the present invention enables a user browsing a web page to search for contextually related content objects. Any content objects in the database management system appearing on the web page are marked and displayed to the user. For example, when the user selects a content object from the page, a client window pops up on the screen, displaying the selected content object as a base object and a list of OLINKs that are contextually connected to the base object. The user may further invoke the searching system to search for other content objects within the base object area by inputting keywords, filter criteria, or search criteria or parameters.

Turning now to Fig. 1, an OLINK, as described in the U.S. Patent Application Serial No. 09/457,619, is an association between two content objects having embedded data about the attributes of the association. The embedded data includes, but not limited to, an originator's identity, origination date, the content object address, content identifier, thumbnail of the content objects, description of the OLINK, statistical information about the OLINKs that varies dynamically over time, e.g., based on user's actions or viewing behavior, or system criteria, such as time, capacity, etc., and ranking of the OLINKs based on usage, relevancy, or any other collected data. A content object can be a URL or a specific content within a URL (e.g., tags within the URL recognized by the user's web browser) or one or more files. The specific content includes, but not limited to, a photo in the form of JEPG, GIF, or other formats, a movie in MPEG, MPEG-2, or other formats, animation, text, sound in WAVE, MP-3, or other formats, and other content types. Additionally, the content object can contain or combine one or more of specific content of different data types and formats.

OLLNKs are automatically generated when a user associates or connects one content object with another content object, such as, but not limited to, dragging a content object and dropping it onto another content object or copying and pasting a content object onto another content object. As shown in Fig. 1, two different and separate OLINKs, one OLINK for each direction, can be automatically generated when a user associates one content object with another content object. In other word, the first OLINK 110 connects the first content object 100 to the second content object 101, and the second OLINK 120 connects backwards the second content object 101 to the first content object 100. For example, two OLINKs can be automatically generated when the user drags and drops the first content object 100 onto the second content object 101.

In one aspect of the present invention, the on-line searching system comprises a server, a client application, and a web-like database management system for storing OLINKs and content objects in the web-like or on-line database. The content objects are interconnected by the OLINKs in an associative and intuitive manner. Preferably, the

OLLNKs in the database are arranged in a neural network structure, wherein contextually related content objects are connected to each other. Fig. 2 illustrates an example of such neural network structure, where OLINK A connects a content object 200 to a content object 203 and OLINK D connects the content object 203 to the content object 200. Also, the content object 200 is indirectly connected to content objects 220 and 222, via OLINKs A-I-K, A-J-N, respectively. But, the content objects 220 and 222 are directly connected to the content object 200 by OLINKs M and U, respectively. The web-like database can be conceptually viewed as a multi-layered data structure, wherein each content object relates to unlimited layers of content objects through OLINKs. All content objects that are directly associated or connected to the base object are referred to as a first layer objects or directly linked objects. All content objects that are indirectly associated or connected to the base object through other content objects are refereed to as indirectly linked objects or n^th layer, where n is an integer greater than 1 and representing the number of OLINKs between the content object and the base object. The base object area includes all OLINKs within any layer of the base object. Referring to Fig. 2, the base object 200 is directly connected to the content objects 201, 202 and 203 via OLINKs B, C and A, respectively. Accordingly, the OLINKs A, B and C are referred to as the first layer links or direct links.

All content objects that are directly linked to the first layer objects are referred to as second layer objects. For example, in Fig. 2, the content objects 211, 212, 213, 214, 215 and 216 connecting directly to the first layer objects and indirectly to the base object 200 via OLLNKs B-E, B-F, C-G, C-H, A-I and A-J, respectively, are second layer objects.

All content objects that are directly linked to the second layer objects are referred to as third layer objects, and so on. For example, in Fig. 2, the content objects 220, 221 and 222 connecting directly to the second layer objects and indirectly to the first layer objects and a base object via OLINKs A-I-K, A-I-L and A-J-N, respectively, are third layer objects.

In accordance with an embodiment of the present invention, the data management system includes rating information indicating the quality and relevancy rate of each OLINK, as described in U.S. Patent Application Serial No. 09/457,619. Preferably, the link rate represents the relevancy and the quality of the association or relationship between the base content object and the associated content object, e.g., between the content objects 100 and 101 in Fig. 1. The link rate can be calculated from the collected statistical information, such as originator's identity of the OLINK or the viewing behavior of the users. The statistical information can include such information as an

OLINK display count, an OLINK hit count, a hit rate, a quality ratio, a view time factor, a recent hit rate, a directional ratio, and other comparable and relevant statistics.

In accordance with another embodiment of the present invention, the data management system includes a filtering mechanism for filtering out one or more irrelevant OLINKs according to user specified filtering criteria. The filtering criteria includes, but not limited to, OLINK's origination date of OLINK, identity of OLINK's originator, content object quality, time factors, OLINK type, OLINK category, content object type, and other filtering parameters. When a user provides a base content object to the database management system, the system server locates all OLINKs that are contextually related to the base object and additionally filters or eliminates OLINKs that

do not meet or satisfy the user defined filter criteria.

The searching system of the present invention searches the database for content objects that are contextually related to a content object that is currently being viewed by the user (or simply referred to as a currently viewed content object or base content object) within the base object area. Since the base objects area can comprise unlimited number of layers, the searching system searches the base object area following the OLINK chains from the first layer to the n^th layer. For example, in Fig. 2, the searching system may search the OLINK chains 200-203, 200- 201- 211, 200-202-213-201, 200-202-214, 200- 203-215-220-200 and 200-203-216-222-231.

Preferably, the searching process includes locating and selecting potential OLINKs within the base object area, and sorting the selected OLINKs into an ordered list. The searching process is now described in conjunction with a flow chart in Fig. 3. A user views a base content object from the client application 301(i.e., views a particular page using the web browser) at step 302. The user can initiate a query or search for content objects that are contextually related to the base content object within the base object area by inputting keyword(s) into the search box at step 303. Additionally, the user may limit the scope of the search by defining the filtering criteria at step 304 and/or the selection criteria or parameters at step 305.

The client application 301 transfers the user query (the keywords, the filtering criteria and the selection criteria) to a system server 300 and the system server 300 locates potential OLINKs in the base object area that satisfies the selection criteria at step

306. The system server 300 eliminates or filters any potential OLINKs and their associated content objects that do not include the keyword(s) at step 307 and do not meet the filtering criteria at step 308 to provide selected OLINKs. That is, the system server 300 eliminates a potential OLINK if the keywords(s) are not found in the embedded data of such potential OLINK and/or in the content objects associated with such potential OLINK.

The system server 300 sorts the selected OLLNKs based on the relevancy of such OLINKs to the base content object at step 309. In sorting the selected OLINKs, the system server 310 may consider other factors or parameters, such as location of the OLINK or the content object, ranking or rating of the OLINK chain or path, etc. The system server 300 sends the sorted or ordered list to the client application 301 and the client application 301 displays the received sorted list to the user.

In accordance with another embodiment of the present invention, any web site or server, such as Internet servers 902 in Fig. 8, can be enabled to communicate with the system server 901 (Fig. 8). That is, the enabled server or web site functions similar to the client application 900 (Fig. 8) and permits any user of the enabled server to search for related content objects using the searching system of the present invention.

Preferably, the user query transferred to the system server 300 includes the base content object (or the base object), the keywords, the selection criteria, and the filtering criteria. The searching system of the present invention searches the database for OLINKs that contain the identification (ID) of the base object as the source or first object. Turning now to an example in Fig. 4, if the base object ID is 400, the searching system searches for OLINKs in the first layer that are contextually related to the base object, such as an OLINK 410 since its source object includes the base object ID. The searching system searches the second layer for OLINKs that are connected to the OLINK 410 or contextually related to the target object of the OLINK 410, such as an OLINK 420 since its source object includes the content object ID of 401. Since OLINK 420 suggests other contextually related OLLNKs, the searching system searches the third layer for OLINKs that are connected to the OLINK 420 or contextually related to the target object of the OLINK 420, such as an OLINK 430 since its source object includes the content object ID of 402.

The selection criteria includes one or more of the following parameters: time factor, e.g., limit the entire search process to 30 seconds or limit the search to 10 seconds per layer, absolute maximum distance, minimum link rate, average link rate, the location information, textual similarity, and other relevant and comparable parameters. Preferably, the selection parameters can be weighted to favor one parameter over the other parameter. The selecting parameters can be weighed in accordance with a system default setting or can be set by the user. The time factor may vary depending on the layer, e.g., search layer 1 for 20 seconds but search layer 2 for only 10 seconds.

The absolute maximum distance permits the user to limit scope of the search to a number of layers, a particular layer or range of layers, such as searching two layers (layers 1 and 2), only layer 2 or layers 2-3. For example, in Fig. 2, if the user limits the search to two layers, then the searching system searches only OLINKs 201-203 in layer 1 and OLINKs 211 -216 in layer 2.

The user can use the minimum link rate criteria to select only OLLNKs with a link rate greater or equal to the minimum link rate. In other words, the searching system eliminates any OLINK chain containing an OLINK with a link rate less than the minimum link rate. An OLINK chain contains one or more OLINKs starting from the base object to the last or final layer in a predefined base object area. Since each layer can have more than one OLINK, the searching system selects only OLINKs meeting or satisfying the selection criteria, i.e., the minimum link rate and limits the search to those OLINKs in the next layer that are contextually related to the selected OLINK. For example, if the user selects a minimum link rate of 80% and the base object A has two OLLNKs B and C in the first layer having link rates of 70% and 85%, respectively. The searching system will select OLINK C as the preferred path and will only evaluate

OLINKs in the second layer that are contextually related to OLINK C. If C also has two OLINKs D and E, and only E's link rate is higher than 80%, the searching system will then select OLINK E as the preferred path and will only evaluate OLINKs in the third layer that are contextually related to OLINK E. Accordingly, the searching system will consider OLINKs within the A-C-E chain as potential candidates.

Turning now to Fig. 5, there are illustrated chains of OLINKs with their corresponding link rates. In accordance with an embodiment of the present invention, for each OLINK stored in the database, the associated embedded data includes the corresponding link rate. For example, if the base object is the content object 500 and the user selects a minimum link rate of 80%, the searching system will locate and identify OLINKs 501 and 502 in the first layer as potential candidates because their link rates exceeds the selected minimum link rate. Additionally, the searching system will locate and identify OLLNKs 51 land 512 in the second layer, OLINKs 520 and 522 in the third layer, and OLINK 531 in the fourth layer as potential candidates. Thus, the searching system designates two OLINK chains, 500-501-511-522-531 and 500-502-512-520

(shaded in Fig. 5), as potential candidates and displays the OLLNKs within the designated chains to the user as containing contextually related content objects. Preferably, the user can select one of the related OLINKs (501, 511, 512, 520, 522 or 531) as a new base object and initiate another search.

The user can use the average link rate criteria or parameter to select OLINKs based on the quality of the chain rather than the quality of the individual components, i.e., OLLNKs, of the chain. The searching system proceeds on a path or chain only if the resulting average link rate is greater than or equal to the average link rate parameter. Preferably, the minimum average link rate varies with the length of path or chain. The average link rate is computed by summing the link rates of all OLLNKs within a chain or path divided by the number of the OLLNKs in the chain. Turning now to Figs. 6A-6C, there are illustrated chains of OLLNKs with their corresponding link rates and tables depicting the results of the average link rate calculation for the two chains. The application of the minimum average link rate as the search criteria will now described in conjunction with Figs. 6A-6C. The varying minimum link rate for each layer is delineated in the tables in Figs. 6A-6B, 60%) for adding any OLLNKs in the first layer, 65%o for adding any OLLNKs in the second layer, and 80%) for adding any OLLNKs in the third layer. The searching system will locate and identify the OLINK 600 and 602 as potential candidates because their average link rate is above the selected minimum link rate of 60% for layer 1. The searching system will proceed on separate paths to the OLINKS 603 and 604 because the resulting average link rate of 70% and 80%, respectively, exceeds the selected minimum link rate of 65% for adding second layer

OLLNKs. However, the searching system will not proceed to the OLINK 605 because the resulting average link rate of 73.3% is below the selected minimum link rate of 80% for adding a third layer OLINK. Thus, the searching system designates two OLINK chains, 600-601-603 and 600-602-604, as potential candidates and displays the OLLNKs within the designated chains to the user as containing contextually related content objects. Preferably, the user can select one of the related OLINKs (601, 602, 603 or 604) as a new base object and initiate another search. In accordance with another aspect of the present invention, the searching system includes keyword crawler. The user can use the keyword crawler to search for OLLNKs containing a particular word(s). The user browsing the content object can select the object as the base content object and input word(s) or text(s) representing a query into the keyword crawler. The keyword crawler then searches the database for any OLINKs in the base object containing the keyword or words similar to the keyword. Preferably, the key word crawler may include an algorithm for evaluating the textual similarity of words so it can identify OLINKs containing words similar to the keyword. This can be also used to sort the selected OLINKs by their textual relevancy to the base content object. It is appreciated that the textual search can be performed within a content object or within an OLINK.

In accordance with an aspect of the present invention, the searching system includes a sorting system comprising a sorting algorithm for sorting the selected OLINKs based on one or more sorting parameters. The sorting parameters or criteria can include, but not limited to, textual similarity measurement, OLINK ranking or rating, time factor, OLINK's absolute distance, and OLINK's accessibility. It is appreciated that the searching and sorting system can interchangeably utilize the selection and sorting criteria or parameters. In other words, the sorting system can sort the OLINKs based on the selection parameter and the searching system can search the database in accordance with the sorting parameter. Preferably, the sorting parameters can be weighted to favor one parameter over the other parameter. The sorting parameters can be weighed in accordance with a system default setting or can be set by the user.

The sorting algorithm calculates the selected parameter for each selected OLINK. An OLINK accessibility represents the number of OLINK chains that are connected to the selected OLINK. Preferably, the OLINK accessibility represents the number of OLINK chains that are connected to the selected OLINK and contextually related to the base content object. Hence, the sorting system considers OLINKs with a higher accessibility number (or highly accessible OLLNKs) to be more relevant to the base content object than OL NKs with a lower accessibility number. Also, in determining the OLINK accessibility, the sorting system considers the length of chain, i.e., the number of OLINKs in the chain. Turning now to Fig. 7, the OLINK 801 is considered highly accessible because there are two chains connecting the base content object 800 to the OLINK 801, chains 800-801 and 800-831-801. It is appreciated that the accessibility of the OLINK correlates directly with the OLINK's popularity.

Various modifications to the preferred embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other embodiments and applications without departing from the spirit and scope of the invention. Thus, the present invention is not intended to be limited to embodiments shown, but is to be accorded the widest scope consistent with the principles and features

disclosed herein.

Claims

WHAT IS CLAIMED:

1. A method for searching an on-line database, wherein contextually relevant content objects are connected by OLLNKs, comprising the steps of: receiving a base content object from a user; and searching the database for all content objects having direct and indirect links to said base content object to provide related content objects.

2. The method of claim 1 , further comprising the step of sorting said related content objects in accordance with relevancy of said related content objects to said base content to provide an ordered list.

3. The method of claim 1, further comprising the step of displaying said related content objects to the user.

4. The method of claim 1, wherein the step of searching includes the step of selecting content objects in accordance with a filtered criteria specified by the user to provide said related content objects.

5. The method of claim 1, each OLINK having at least a link rate specifying the relevancy of the content objects connected by said OLINK.

6. The method of claim 5, wherein said indirect links includes distance information representing the number of OLINKs between a related object and said base

content object.

7. The method of claim 6, wherein the step of searching includes the step of selecting OLINKs in accordance with a searching criteria to provide said related content objects.

8. The method of claim 7, wherein said searching criteria includes one or more of time factor, maximum absolute distance, minimum link rate and average link rate.

9. The method of claim 8, wherein each of said searching criteria is weighted in accordance with a predetermined setting.

10. The method of claim 8, wherein each of said searching criteria is weighted by the user.

11. The method of claim 2, wherein the step of sorting sorts in accordance with the relevancy of said related content objects to said base content and a sorting criteria to provide said ordered list.

12. The method of claim 11, wherein said sorting criteria includes one or more of time factor, maximum absolute distance, minimum link rate and average link rate.

13. The method of claim 12, wherein each of said sorting criteria is weighted in accordance with a predetermined setting.

14. The method of claim 12, wherein each of said sorting criteria is weighted

by the user.

15. The method of claim 4, wherein said filtered criteria includes one or more of content object type, origination date, originator identity, keyword, location information

and OLINK category.

16. The method of claim 15, wherein each of said filtered criteria is weighted

by the user.

17. The method of claim 2, wherein the step of sorting sorts in accordance with relevancy of said related content objects to said base content and a filtered criteria specified by the user to provide said ordered list.

18. N system for searching an on-line database over a computer network, wherein contextually relevant content objects are connected by OLINKs, comprising: a computer associated with a user having an input device; a client application residing in said computer for selecting a base content object; and a system server, connected to said computer over said computer network, for receiving said base object and searching the database for all content objects having direct and indirect links to said base object to provide related content objects.

19. The system of claim 18, wherein said server is operable to sort said related content objects in accordance with relevancy of said related content objects to said base content to provide an ordered list.

20. The system of claim 18, wherein said system server is operable to transmit said related content objects to said client application; and wherein said client application is operable to display said related content objects to the user.

21. The system of claim 18, wherein said system server is operable to select the content objects in accordance with a filtered criteria specified by the user to provide said related content objects.

22. The system of claim 18, each OLINK having at least a link rate specifying the relevancy of the content objects connected by said OLINK.

23. The system of claim 22, wherein said indirect links includes distance information representing the number of OLINKs between a related object and said base content object.

24. The system of claim 23, wherein said server is operable to select the OLINKs in accordance with a searching criteria to provide said related content objects.

25. The system of claim 24, wherein said searching criteria includes one or more of time factor, maximum absolute distance, minimum link rate and average link rate.

26. The system of claim 25, wherein each of said searching criteria is weighted in accordance with a predetermined setting.

27. The system of claim 25, wherein each of said searching criteria is weighted by the user.

28. The system of claim 19, wherein said server is operable to sort in accordance with the relevancy of said related content objects to said base content and a sorting criteria to provide said ordered list.

29. The system of claim 28, wherein said sorting criteria includes one or more of time factor, maximum absolute distance, minimum link rate and average link rate.

30. The system of claim 29, wherein each of said sorting criteria is weighted

in accordance with a predetermined setting.

31. The system of claim 29, wherein each of said sorting criteria is weighted

by the user.

32. The system of claim 21, wherein said filtered criteria includes one or more of content object type, origination date, originator identity, keyword, location information and OLINK category.

33. The system of claim 32, wherein each of said filtered criteria is weighted

by the user.

34. The system of claim 19, wherein said server is operable to sort in accordance with relevancy of said related content objects to said base content and a filtered criteria specified by the user to provide said ordered list.