BM25-AH: Enhanced BM25 Algorithm for Domain-Specific Search Engine

Abstract

The Virginia Military Institute (VMI) uses Google search to provide webpage search service in the VMI website. As Google search is a general-purpose service, it does not consider VMI-specific information and in turn, often fails to retrieve relevant information. To address the limitation, in this paper, we propose the BM25-AH algorithm, an extension of the BM25 ranking algorithm that utilizes domain-specific knowledge by adding four new features to the existing BM25 algorithm. The implementation results show us the potential of the proposed algorithm and the new VMI search engine.