FastLSH – A Similarity Search System

Abstract

Similarity search is an important component in Artificial Intelligence field. While existing solutions are usually problem specific, I proposed a flexible and high performance similarity Search Engine on High Dimensional Big data. The system can complete a query with one million base set in less than a hundred milliseconds. The system aims at providing a Fast, Flexible, Scalable and Extensible solution for similarity search by its seven different packages. By providing a wide range of interfaces, user may pick the one best suits their own needs for similarity search.