Data processing: database and file management or data structures – Database design – Data structure types
Reexamination Certificate
1999-03-03
2002-03-12
Mizrahi, Diane D. (Department: 2171)
Data processing: database and file management or data structures
Database design
Data structure types
C707S793000, C707S793000, C707S793000
Reexamination Certificate
active
06356899
ABSTRACT:
BACKGROUND OF THE INVENTION
FIELD OF USE
This invention relates generally to a method for identifying, filtering, ranking and cataloging information elements; as for example, Internet, World Wide Web pages, considered in whole, in part, or in combination; and more particularly, to a method, preferably implemented in computer software, for interactively creating an information database including preferred information elements, the method including steps for enabling a user to interactively create a frame-based, hierarchical organizational structure for the information elements, and steps, thereafter, for identifying by iteration and automatically filtering and ranking by degree of relevance information elements, for populating the frames of the structure to form; for example, a searchable, World Wide Web page database. In further detail, the method features steps for enabling a user to interactively define a frame-based, hierarchical information structure for cataloging information, and, steps for identifying information elements to populate respective frames of the structure by iteration, the iteration including steps for: identifying a preliminary population of information elements with the use of a search query based on respective frame attributes, frame attributes selectively including classification designations, example pages, stop pages and/or control parameters used by conventional search engines, as required; supplementing preliminary population based on usage of example pages and/or stop pages; expanding the supplemented preliminary population to include related information; automatically filtering and computing information element ranking based on degree of relevance to the respective frame; and, thereafter, refining the identification with successive iterations of the steps described until identification is deemed complete, whereupon the hierarchical structure is populated with a user-defined portion of preferred information elements identified.
As a yet further problem, and potentially an even more perplexing one, not only has The computer revolution created a greater need for information, but, undeniably, it has created an abundance, indeed, an overabundance of information to meet that need. In fact, the computer revolution has spawned so much information, that it is now to the point where the amount of information available on most subjects is typically so large as to create the new and associated problems of going through that wealth of information, and selecting from it the items most relevant to the question at hand.
For example, in the case of the Internet's World Wide Web, if one were looking for information concerning something as straightforward as the restoration of an old car, there likely would be hundreds, if not thousands, of potential Web sites having as many if not more pages of information relating to the subject of old cars, and the parts, services and techniques for their restoration. Accordingly, one faced with the problem of developing information on the subject of automobile restoration, would potentially be required to locate and go through literally hundreds of Web pages in an attempt to find those few most suited to his needs.
In the past, the World Wide Web's approach to this problem has been to provide search facilities such as Yahoo®! and others, to assist Web users in finding the information; i.e., Web pages, they might be looking for. However, search facilities such as Yahoo! typically provide only generalized organizations of Web subject matter, those organizations being arranged as categories of Web pages, the categories and the things included in them being based on the nature of the Web sites, the subjective points of view of numerous staff classifiers working for the search facility, and the classification criteria they established. In accordance with this approach, organization of the information is, therefore, influenced by the respective points of view of the various classifiers, the providers of the search facilities, and the Web site providers. As a result, such Web subject matter organizations tend to be subjective and suffer from over inclusion and under inclusion of information, which, in turn, affects their relevance, accuracy and ease of use.
Moreover, and of yet greater concern, is the fact that formulating and maintaining organizations of Web subject matter in the fashion noted requires expenditure of substantial amounts of human time and effort and, accordingly, money. Particularly, continuous growth and change in Web makeup requires such organizations of Web information to be repeatedly supplemented and the existing framework revised to accommodate the introduction of new and changing information. Accordingly, such approaches are manpower intensive, leading to higher costs for creation and maintenance, and because of the extensive human involvement, are, as well, subject to error.
Still further, such search facilities, typically, are unable to group the information elements they return; e.g., Web pages, by their respective “relevance”, that is, the degree to which others have referred to; i.e., pointed to, the respective elements; e.g., pages, as sources of information on the subject matter in question. Pages that have many references pointing to them are termed herein “authorities”. In this scheme, and in the context of Web pages, “relevance” is a function of the number and quality of links to an authority page from various hub pages, referred to as the “authority weight” for the respective authority page, or, the number and quality of links from a hub page to various authority pages, referred to as the “hub weight” for the respective hub page. Moreover, and as will be appreciated, pages of higher relevance; i.e. higher authority weight or higher hub weight, are “preferred” where one is seeking information concerning particular subject matter. Accordingly, “preferred” information elements; e.g., Web pages, are considered to have higher relevance to some specific subject matter where the information elements; e.g., Web pages, have either, higher authority weight, or, higher hub weight with respect to the particular subject matter. And, as will also be appreciated, since information elements; e.g., Web pages, may both point to authority pages; i.e. function as a hub, and also be pointed to as an authority; i.e., function as an authority, such pages may be relevant either as a hub page or as authority page, or as both.
No prior references has proposed systems or methods for enabling a user to interactively create an information database of “preferred” data elements such as “preferred” Web pages; i.e., pages of either higher authority weight, or hub weight; i.e. “relevance”, or, procedures for removing spurious factors that arise during computation of hub and authority weights for the respective pages.
With regard to relevance; i.e. weight, computation, workers in the field have found that the computational accuracy is adversely affected by such factors as “self-promotion”, “related-page promotion,”, “hub redundancy”, “copied pages”, and “false authority.” Particularly, it has been found that during relevance computations pages with links to other pages of the same Web site can improperly confer authority upon themselves, thus giving rise to false promotion; i.e., “self-promotion,” and adversely affecting relevance computation accuracy. Further, it has been found that in addition to self-promotion, related pages from the same Web site, as for example, a home page and several sub-pages of the home page can improperly accumulate authority weights, giving rise to false promotion in the form of “related-page promotion”, which again adversely affects relevance computation accuracy.
Further still, workers have found that a page may have value only because of the hub links it contains; that is, its content may be otherwise irrelevant. In that case, if the hub links for such a page can be found in other pages, the hub links of such a page are redundant and may not be suitable for inclusion. It is to be note
Chakrabarti Soumen
Dom Byron Edward
Gibson David Andrew
Raghavan Prabhakar
Rajagopalan Sridhar
Mizrahi Diane D.
Rogitz John L.
LandOfFree
Method for interactively creating an information database... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for interactively creating an information database..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for interactively creating an information database... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2834055