Searching Large Textual Dataset With Limited Computational Resources

In this paper we propose a search approach that can process large volumes of textual data efficiently and effectively even in environments where computational resources are limited. The traditional search solution for large collections assumes availability of practically unlimited computational resources. For many applications and organization this assumption is not realistic. Empirical evaluation of the proposed approach using some of the largest available datasets demonstrates that the proposed search approach is substantially more efficient than the existing approach, is on par if not better in terms of effectiveness, and can operate using very few computational resources. AUDIENCE: [Information Retrieval/Large-scale Search] [Text Processing] [Advanced technical talk]