一种基于语义关系的情感词典自动构建方法

构建英文情感词典研究相对成熟，形成了丰富可靠的词典资源。而针对中文的研究时间短，中文情感分析词典资源较少。借鉴现有可靠的英文词典资源，提出了基于语义关系的情感词典自动构建算法，算法先从HowNet的概念中进行中文义原和词语抽取及语义分析，再利用HowNet概念中DEF中英文属性值，在英文情感词典SentWordNet中进行义原和词语情感值查询，最后根据词语和义原之间的语义关系进行词语的情感值计算。算法直接利用现有的英文情感词典，无需人工标注，生成的情感词典记录了词语的语义关系、情感极性值等多种信息，弥补了现有词典的不足。评测实验结果表明，根据算法实现的情感词典相比其他词典在准确率接近的情况下，召回率和F值最高，取得了较好的评测性能。

An automatic constructing approach for sentiment lexicon based on semantic relations

Researches on Constructing English sentiment lexicon is relatively mature, and there have been abundant and reliable lexical resources. While for Chinese studies, the period is short, and there are only a few Chinese sentiment lexicon resources. In this paper, with reliable English sentiment lexicon as reference, an automatic constructing approach has been proposed based on semantic relationships. Firstly the Chinese sememes and words are extracted from the defination of concepts in HowNet and the semantic analysis is carried out upon them, secondly the sentimental value of each sememe and word is retrieved from English sentiment lexicon SentiWordNet according to the DEF attributes of concepts in HowNet, and the final sentimental value of each word is calculated on the semantic relations of the sememes and words. The ready English lexicon is used without mannual labeling in the mehod, and diverse informations of words are recorded in the final lexicon including semantic relations and sentimental valuses, which remedy the lacks of other lexicons. The experimental results show that the resulted sentiment lexicon achieves best performance on the recall and F value measurements while approaching other lexicons on the precision measurements.