Browse Prior Art Database

Web search engine that learns additional, possible key search phrases from differing pages visited by a user

IP.com Disclosure Number: IPCOM000010317D
Original Publication Date: 2002-Nov-20
Included in the Prior Art Database: 2002-Nov-20
Document File: 2 page(s) / 40K

Publishing Venue

IBM

Abstract

This disclosure proposes a method for a browser to learn additional key phrases within visited pages and suggest those words for future or additional searches.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 50% of the total text.

Page 1 of 2

Web search engine that learns additional, possible key search phrases from

differing pages visited by a user

The web has provided users of computer systems with the luxury of having any type of information at their fingertips, especially with the aid of "free" web search engines. The problem with this is, that in order for the user to efficiently use a search engine, they have to type in the correct key words by which they instruct the search engine to present to them (the user) possible links of interest that the user will then go and investigate in detail. Many search engines have tried different strategies in order to present the best possible chance of giving the user useful information.

However, there exists a paradox: How can one learn about a subject by searching for key words of a subject, if one does not know enough about the subject in order to input the appropriate key words? An example is that a person could want to learn about a model of a car called Jaguar, but the user does not know that much about cars. If they type in the word "jaguar" into a web search engine, they might get "hits" about the following: cars, racing, or the animal specie.

The user does not have an interest about racing, the user wants to know how it compares to other cars and what type of features it offers. Additionally, the user does not have an interest in links about the animal species of "Jaguar".

What is proposed by this disclosure is a method for a browser to learn additional key phrases within visited pages and suggest those words for future or additional searches. For example, using the search engine results above, the user begins by guessing at a good link to start with and visits a page. When they enter that page, in the background a browser plugin does the following:

a) strip out all the prepositions and nonessential words in the text of the...