Data Cleaning publication categorizer

CiteSeer began as the first search engine for scientific litera-
ture to incorporate Autonomous Citation Indexing, and has since grown
to be a well-used, open archive for computer and information science pub-
lications, currently indexing over 730,000 academic documents. However,
CiteSeer currently faces significant challenges that must be overcome in
order to improve the quality of the service and guarantee that Cite-
Seer will continue to be a valuable, up-to-date resource well into the
foreseeable future. This paper describes a new architectural framework