基于最大熵模型的观点句主观关系提取

Extraction of Subjective Relation in Opinion Sentences Based on Maximum Entropy Model

提出一种提取中文观点句中评价对象和评价词主观匹配关系的方法。分析观点句中评价词和评价对象的词性、词语位置,通过句法分析获取语义特征,将2类特征应用于最大熵模型,提取观点句的主观关系。实验结果证明,与取距离评价词语最近的词作为评价对象的Baseline方法相比,该方法大幅度提高了准确率和F测试值。

This paper presents a novel method of extracting subjective relation between opinion targets and opinion-bearing words in Chinese opinion sentences. This method analyzes lexical and part of speech information in the sentence. Syntactic analyzing is adopted to achieve syntactic path information which is regerded as semantic feature. The two kinds of feature are both applied in maximum entropy model. According to this model,all subjective relations in the sentence are extracted. Experimental results show that this method is better than Baseline method in precision rate and F-measure.