Retrieval of relevant information categories

Details Retrieval of relevant information categories Retrieval of relevant information categories

2000-05-19

2004-03-09

Robinson, Greta (Department: 2177)

Data processing: database and file management or data structures

Database design

Data structure types

Reexamination Certificate

active

06704729

ABSTRACT:

TECHNICAL FIELD
The present invention relates to the field of electronic data storage and retrieval. More specifically, the present invention relates to selecting categories of information in an Internet search engine database, based on a user's query.
BACKGROUND OF THE INVENTION
The Internet is a vast store of information, permitting access to tens of millions of web sites. Although the ever-increasing number of web sites are creating record access to information, the sheer number of sites available make it difficult for an Internet user to locate desired information. As a result, Internet search engines have become a necessary and valuable tool for locating information on the Internet
Not all search engines employ the same search strategy. Some Internet search engines return a “flat” list of results indexed according to a web site's similarity to a user's query. Although these lists are useful, the vast expanse of the Internet has reduced their effectiveness. Other Internet search engines take a different approach and catalog individual web sites into hierarchical taxonomies of categories based on the site's content. These category-based search engines not only return the most relevant web sites, but also lists of matching categories that describe and encompass relevant web sites in order to help users focus their query. In addition, these category-based search engines often display the matching category along with their hierarchically related categories in “category paths” in order to place the matching category in a proper context.
For some time, these categorization techniques were sufficient to provide search engine users with intelligible suggestions. However, just as previous Internet growth necessitated categorization over “flat” result lists, the Internet's recent exponential growth has limited the effectiveness of these taxonomy techniques. In particular, the rapid growth of the Internet has caused a corresponding expansion in the number of categories found in today's search engine taxonomies. As a result, simply categorizing web sites and displaying category paths has become unwieldy and unintelligible, often resulting in tens or hundreds of returned category paths. Moreover, the lack of differentiation among the many returned category paths often results in logical redundancies and even irrelevant search results. Therefore, although today's category paths are better than their predecessor “flat” lists of web sites, they have been rendered ineffective by the Internet's exponential growth.
Therefore, it would be beneficial to provide a logical distinction among the many possible matching categories and their category paths. In this way, the user can more easily sort through the returned results and more quickly focus the search to obtain the desired results.
SUMMARY OF THE INVENTION
Large stores of information are often organized in a hierarchical taxonomy to aid a search and retrieval of the information. The hierarchical taxonomy generally consists of related categories of information, called “nodes,” that each may contain information relevant to the search. Each node is addressable according to its path in the hierarchical taxonomy. In information stores where the number of nodes having relevant information is extremely large, such as the Internet, providing a cohesive, intelligent, and organized display of the search results becomes extremely important to the success of a user traversing the store to find relevant information. The invention provides such search results by ranking each node of the taxonomy to determine which nodes are most likely to be relevant to the search request. The invention then creates a conceptually-related “cluster” of nodes by selecting a relevant “seed” node and relevant nodes related to the “seed” node.
More specifically, the invention provides a method, system and computer-readable medium for selecting nodes in a hierarchical taxonomy. The method comprises the steps of receiving a query and comparing characteristics of the nodes with the query. The method ranks the nodes based on a predetermined criteria and selects a first node based on the ranking. The predetermined criteria may be based on common structure between the nodes and the query. Also, based on the ranking, the method selects one or more nodes that are hierarchically related to the first node. The method may then display the first node and the related nodes in such a way that the relation of the nodes is apparent. The method may be repeated to select additional relevant nodes and their related nodes.


REFERENCES:
patent: 5819258 (1998-10-01), Vaithyanathan et al.
patent: 5875446 (1999-02-01), Brown et al.
patent: 5924090 (1999-07-01), Krellenstein
patent: 5940821 (1999-08-01), Wical
patent: 5991756 (1999-11-01), Wu
patent: 6012053 (2000-01-01), Pant et al.
patent: 6112203 (2000-08-01), Bharat et al.
patent: 6223145 (2001-04-01), Hearst
patent: 6366910 (2002-04-01), Rajaraman et al.
patent: 6385602 (2002-05-01), Tso et al.
patent: 6397209 (2002-05-01), Reed et al.
patent: 6397218 (2002-05-01), Stern et al.
YAHOO Website, http://search.yahoo.com/search, printed Jan. 12, 2000, 6 pages.

Affiliated with

Also associated with

Rating

Retrieval of relevant information categories does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Retrieval of relevant information categories, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Retrieval of relevant information categories will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3226717