Search engine for a computer network
AssigneeClaria Corporation
InventorAnthony Martin
Life of this patent
Estimating the term…
Abstract
In one embodiment, client-side programs are employed to observe the navigation of consumers to various websites. Addresses of web pages viewed by consumers may be used to fetch the web pages. A fetched web page may be parsed for one or more keywords. The relevance of the web page to a keyword may be ranked according to consumer preferences, which may relate to consumer interaction with the web page. Web pages and their ranking information may be stored in an index. The index may be consulted to find links to web pages relevant to a keyword employed in a search request.
Claims20
A method of responding to a search request, the method comprising:
- using client-side programs to monitor navigation of consumers on the Internet;
- building a search engine index using navigation data from the client-side programs;
- receiving a search request for a keyword;
- and
- sifting through the index to find links to web pages responsive to the keyword.
A search engine comprising:
- a crawler configured to receive addresses viewed by consumers as observed by client-side programs and to fetch the viewed web pages;
- an indexer configured to parse web pages fetched by the crawler and to store links to the fetched web pages in an index;
- and
- a sifter configured to sift through the index to find a stored link relevant to a keyword employed in a search request.
A method of responding to a search request, the method comprising:
- observing navigation of consumers to websites to generate navigation data;
- using the navigation data to build an index for a search engine;
- receiving a keyword as part of a search request;
- and
- responding to the search request by searching the index for links to web pages relevant to the keyword.
Dependencies are inferred from the claim text.