期刊名称：International Journal of Innovative Research in Computer and Communication Engineering
印刷版ISSN：2320-9798
电子版ISSN：2320-9801
出版年度：2014
卷号：2
期号：1
出版社：S&S Publications
摘要：Similarities play a vital role in clustering text on the prediction, in order to produce an efficient result when compared to the existing algorithms li ke k-modes, ROCK and STIRR. Future selection is important for making a subset according to the dataset. In order to overcome the problems in the existing system, single cluster and multiple clustering methods are proposed in order to cluster the famous quo tes with multiple semantic associations. But the problems on overlapping between the quotes are analyzed and the sentence similarities for information retrieval are measured. A FUZZY logic in finding the similarities to form a cluster, based on the relational prototypes has been proposed. A semantic clustering and FUZZY based pruning approach is practiced to bring more accuracy in mining process. FUZZY makes possible on using more complex prototypes that should be represent on the clustered text. The algorithm identifies the semantically related sentences and avoids duplication on the given data set. The information retrieval based on the keyword in which filtering is processed on the benchmark dataset. The result states the information retrieval based on the FU ZZY algorithm maximizes the effectiveness.
关键词：Benchmark dataset; Feature subset; filtering; FUZZY based clustering
