Features Grouping And Orientation Analysis Of Conditional Sentences

This paper mainly analyzes the orientation of features in product reviews. As different language using, different reviewers may use different feature expressions, and we need to group different feature expressions into suitable groups. Also, different reviewers may use different sentences, so we need to deal with some special sentences. Our work can be divided into two parts.-Feature grouping and orientation analysis of conditional sentences.On feature grouping, considering that, although the same feature may have several feature expressions, these expressions are always used with same opinion words in a sentence. Product feature expressions and opinion words are first extracted in pairs to build a bipartite graph, and then. Weight Normalized SimRank is used to compute similarity between different feature expressions in the bipartite graph, and the similarity is used to optimize the Bayesian classifier in Semi-Supervised Learning. Experimental results show; that the proposed method is useful.When we analyze the orientation of features in conditional sentences, we first identify conditional sentences, and then analyze the orientation. Conditional sentences usually contain conditional conjunctions, but some sentences don’t contain conditional conjunctions, and these sentences are called implicit conditional sentences. Implicit conditional sentences contain some words which are able to express conditional relationships, and these words are called implicit conditional words. Conditional sentences are identified by using conditional conjunctions and implicit conditional words and their tags and CSR rules. When analyzing the orientation of conditional sentences, conditional sentences are classified into four classes according to conditional conjunctions and implicit conditional words, and the classification is used for SVM classification. It is proved that the proposed method is useful according to the experimental results.