Research on product feature extraction for Chinese reviews
Guiying Wei, Wenming Cai, Lan Qin · 2016
Opinion mining technique aims at automatically obtaining valuable information by retrieving and processing product reviews on the web, product feature extraction is the key step for opinion mining. Currently, the product feature extraction has been studied a lot in the field of English. However, the research of Chinese feature extraction is insufficient. There exist many issues such as small size of datasets, low precision rate, etc. In this paper, we put forward a feature extraction algorithm for Chinese reviews based on Apriori algorithm on the basis of Chinese language characteristics. Experiments with a large number of real network reviews data are carried out and good results are obtained. The experiment results show that this algorithm is effective and has a certain practical value.