Research On Chinese Opinion Interpretation Classification Method For Online Comments

Recent years has seen many advances made in the area of opinion mining.However,most current opinion mining systems are only concerned with mining positive or negative opinions from opinionated texts,and ignore the detailed explanation of the underlying reasons,facts,suggestions and conditions for the opinions.Actually,such explanatory information is extremely important for decision-making and information service based on opinion mining.To overcome this challenge,in this paper,we develop methods for opinion explanations classification by exploring the characteristic of opinion explanations in Chinese product reviews.Specifically,our research concerns the following three aspects:(1)Construction and analysis of Chinese opinion explanation classification corpus:In view of the fact that there is no public explanatory opinion resource libary available,this paper first constructs the opinion explanation classification corpus of phone domain and hotel domain based on the construction of classification systems and the relevant annotation norms for opinion explanation.Then,we compare the characteristics of opinion explanations in different domains.The analysis results show that there is no one-to-one correspondence between opinions and opinion explanations,and the characteristics of opinion explanations of each category are very different.(2)Chinese opinion explanation classification based on SVM:Under the framework of SVM,this paper explores the influence of different feature selection and representation methods on opinion explanation classification,especially the basic internal features of opinion explanation,and the contextual features such as attribute,evaluation,and polarity that are associated with opinion explanation.The experimental results show that the multi-feature fusion can effectively improve the performance of opinion explanation classification.(3)Chinese opinion explanation classification based on neural networks:In order to further improve the performance of opinion explanation classification,this paper attempts to solve the classification problem by using different neural network models such as CNN,RNN and LSTM on the basis of SVM method based on feature engineering.The experimental results show that the classification methods based on neural network model are better than the traditional SVM method,and the CNN method is superior to the RNN and LSTM method.