Binary Feature Selection and Integration in Specialized Search Engines

We present a methodology for rapid implementation of specialized search engines. To catalog data, these search engines interpret and classify the content of w eb material to identify different represen- tations of common domain-related elements. While designers can typically develop multiple partial solutions for interpreting the data, acceptable relevance determination requires the appropriate integra- tion of all of these solutions. We present a method for automatically integrating such partial solutions in a Bayesian framework. The Bayesian framework produces a search engine where each user can control the false alarm rate in an intuitive yet rigorous fashion. We discuss the use of this technique in the con- struction of DEADLINER, a search engine that catalogs conference and seminar material found on the web.