期刊名称：International Journal of Computer Science and Information Technologies
电子版ISSN：0975-9646
出版年度：2014
卷号：5
期号：2
页码：2471-2476
出版社：TechScience Publications
摘要：Correlation Preserving Indexing is a spectral clustering method which discovers intrinsic structures embedded in high-dimensional document space. But the problem is to predict the result of one variable based on another variable is not suitable for all the situations. So, the directed Ridge regression is used which computes the relationship among the variables based on the Eigen values to identify the similarity between the documents. But in these two methods the similarity is identified by taking terms. So, there is high computation and less clustering efficiency. Further to improve the cluster efficiency, in this manuscript an innovative technique is introduced which is called Sentence level document clustering in fuzzy relational spectral clustering (SCFSC). The spectral fuzzy can better handle clusters with a complex, nonlinear geometric structure and it does not need prior information on the number of clusters. In this method the similarity between the sentences are measured by using the standard similarity measure. By using the fuzzy relational spectral clustering the efficient clustering is achieved. An experimental results show that the proposed system achieves high clustering efficiency and less computation
