Aspect-level Sentiment Analysis based on a Generalized Probabilistic Topic and Syntax Model

In this research, we apply a generalized topic and syntax model named Part-of-Speech LDA (POSLDA) to sentiment analysis, and propose several feature selection schemes to separate entities and modifiers so that we can conduct sentiment analysis at both document and aspect levels. We also explore ways of optimizing the model parameters for POSLDA and the training of a classifier based on Maximum Entropy Modeling. The advantage of using POSLDA is that we can automatically separate semantic and syntactic classes, and easily extend it to aspect level sentiment analysis by mapping topics to aspects. However, the noun-related classes, which are also treated as semantic classes, should be removed as much as possible to reduce their impact on sentiment analysis. To evaluate the effectiveness of our solutions, we conducted experiments on two collections of review documents and received the accuracy results competitive to the previous work on sentiment analysis.