Opinion mining of camera reviews based on Semantic Role Labeling

Based on the analysis of the existing opinion mining techniques, mining accuracy and complexity, this paper proposes a opinion mining system which mines useful opinion information from camera reviews by utilizing Semantic Role Labeling (SRL) and polarity computing method. Feature lexicon and sentiment lexicon are constructed to mine features and emotional items. In the end, the comparison between positive and negative opinion are presented visually. Our experimental results show the system is feasible and effective.