KnowItAll

Ars Technica links to an interesting research project (KnowItAll)at the U of Washington aimed at developing a search tool for extracting large collections of facts from the web. If you wanted to know about used car dealers in San Francisco, for instance, the tool would generate a list of dealers for you. The search engine uses an awareness of syntactic patterns to extract information from phrases such as "used-car dealers including XXX". DARPA and Google are reported to be funding the project.

From the abstract of an overview paper on the project's website:

Manually querying search engines in order to accumulate a large body of factual information is a tedious, error-prone process of piecemeal search. Search engines retrieve and rank potentially relevant documents for human perusal, but do not extract facts, assess confidence, or use information from multiple documents. This paper introduces KNOWITALL, a system that aims to automate the tedious process of extracting large collections of facts from the web in an autonomous, domain-independent, and scalable manner.