随着互联网门户网站的崛起,以及人们在选购商品时对于互联网的依赖,人们往往会在选购商品前在网络上查找商品的评价信息。但是面对海量的评论信息,感觉无从下手,需要一种能够自动识别评价语句情感倾向性的方法。针对这种情况,陈豪,刘功申等人选定酒店、笔记本电脑、书籍这3种商品,根据商品的特殊性,提出一种基于句法分析的商品情感倾向性分析算法,这种算法的创新之处在于利用句法分析的方法获得评价对象集,并且对评价对象进行权值的分配,再利用依赖关系找到并且处理从句,以及隐含的评价对象,最后得出整句评价语句的正负面评价结果。
With the development of e-commerce sites, more and more people are willing to make a survey on the internet before buying products. However, there are too many comments about the internet, and it is impossible to look up all of them, so the automatic identification of the comments is needed. To know the emotional tendency of every comment, CHEN Hao and LIU Gong-shen employ the syntactic parsing algorithm to analyze the comments, and the hotel, book and laptop become the research objects. According to the particularity of products, the object collection of evaluation is extracted, the allocation of weights done, and then the entire evaluation sentence emotional tendency acquired, and finally the positive and negative results of the products evaluation are concluded. The innovations of this algorithm lie in that the parsing is used to get the objects and deal with the clause and the implied objects.