Computer science student wins first place awards in Text Retrieval Conference competition

Computer science assistant professor Alexander Kotov and his Ph. D. student Saeid Balaneshin-kordan won two first place awards at the Text REtrieval Conference (TREC) Information Retrieval Competition in Gaithersburg, Maryland. The conference was hosted by the National Institute of Standards and Technology (NIST), one of the nation's oldest physical science laboratories.

The conference consists of multiple workshops that present challenges in different information retrieval (IR) environments. IR is used to help narrow down information to find the most relevant data, such as Google or a library database.

Kotov and Saeid received first place in Task A, automatic and manual runs, and Task B, manual runs, in the clinical decision support track, beating out schools such as John Hopkins University, University of Cambridge and University of Michigan. They developed a search system for medical articles that can process very long queries (medical case descriptions) and uses
knowledge base for semantic matching of those queries to relevant PubMed articles.