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：2015
卷号：3
期号：10
DOI：10.15680/IJIRCCE.2015.0310178
出版社：S&S Publications
摘要：Clustering is one of the most classical research problems in pattern recognition and data mining and ithas been widely explored and applied to various applications. In MTSC (Multitask Spectral Clustering) there were theinter task correlations are identified in the unsupervised way in the random matched correlations, so that the clusterslabels are most accurate. In this proposed study Sequential clustering is going to perform with the priority basedcorrelation clustering. This proposed idea provides to a most accurate spectral clustering results. This sequentialprediction finds the important labels between the multi tasks. There are many similar inter tasks are there but all theinter tasks are not important to make spectral clusters so that the prediction is performed by finding the most connectedinter task labels. Since this proposed system find the most important labels this study gives more accuracy than theexisting system and also it takes less time than the MTSC scheme. The experimental results and study shows that theproposed system outperforms the previous multi task clustering schemes.
关键词：Clustering; Multi task; spectral; Sequence; Priority
