TopX - Efficient and Versatile Top-k Query Processing for Text, Semistructured, and Structured Data

This paper presents a comprehensive overview of the TopX search engine, an extensive framework for unified indexing and querying large collections of unstructured, semistructured, and structured data. Residing at the very synapse of database(DB) engineering and information retrieval (IR), it integrates efficient scheduling algorithms for top-k-style ranked retrieval with powerful scoring models, as well as dynamic and self-throttling query expansion facilities.