期刊名称：International Journal of Computer Trends and Technology
电子版ISSN：2231-2803
出版年度：2014
卷号：15
期号：4
页码：168-173
DOI：10.14445/22312803/IJCTT-V15P136
出版社：Seventh Sense Research Group
摘要：The proposed system focus on documentlevel, feature level opinion classification or general domains in conjunction with topic detection and opinion sentiment analysis, based on the semantic label annotation techniques. An additionally ATOM has been suggested. And the proposed system finally identifies whether the semantic orientation of the given text is positive, negative, or neutral. This can detect sentiment and topics simultaneously with active learning paradigm. The objective of the proposed system is providing and clustering data from book review domain using weakly supervised, active leaning and unsupervised learning process. An approach for semantic sentence and feature based clustering based on structured and sentence based clustering technique is the objective. At first the reviews and documents from the book review dataset are clustered in Static method using Active Learning Processing technique combine for opinion detection and identifying the exact topic and opinion has been used, all documents should be preprocessed in the initial stage.
关键词：semantic text bi-clustering; Machine Learning; support vector machine; Natural Language Processing; Porter Stemmer
