A new hybrid approach for document clustering using tabu search and particle swarm optimization (TSPSO)

Abstract

Clustering of text documents is the quickest developing research area, because of the availability of vast amount of information in an electronic form. To solve this document cluster analysis difficulties more efficiently and quickly, this paper proposes a hybrid method using tabu search particle swarm optimization (TSPSO). First, the automatic merging optimization clustering (AMOC) algorithm was performed for the formation of clusters and then implemented the optimization model using the variance ratio criterion (VRC) as fitness function.Second, this paper combines TS and PSO algorithm to use the exploration of both algorithms and to avoid flaws of both algorithms.The testing of TSPSO algorithm is performed on several standard datasets, and the results are compared with PSO and TS. So, the proposed TSPSO is efficient and effective for the problem of document clustering; we have tested PSO, TS, and our proposed TSPSO algorithm on various text document collections.