A new relevant document retrieval algorithm via adaptive discrete stochastic optimization

Abstract

In recent years, information is increasing exponentially which makes it more and more difficult for people to find the needed information from the huge database. To fulfill this demanding, a high accurate and fast-time document retrieval algorithm is highly required for current applications. In this paper, based on the document similarity maximum criterion, we propose a new fast-time document retrieval algorithm based on the adaptive discrete stochastic optimization method. The designed adaptive step-size ensures the proposed algorithm converges fast to the relevant documents in the database and retrieve the optimal document. Numerical results demonstrate that the proposed algorithm gets better converge and retrieval performance than conventional methods in the huge database.