Main Article Content
Abstract
We propose a collaborative multi-domain reviews classification by opinion mining approach to train reviews classifiers for multiple domains simultaneously. In our approach, the reviews information in different domains is shared to train more accurate and robust reviews classifiers for each domain when labeled data is scarce. Specifically, we decompose the reviews classifier of each domain into two components, a global one and a domain-specific one. The global model can capture the general reviews knowledge and is shared by various domains. The domain-specific model can capture the specific reviews expressions in each domain. In addition, we extract domain-specific reviews knowledge from both labeled and unlabeled samples in each domain and use it to enhance the learning of domain-specific reviews classifiers. Besides, we incorporate the similarities between domains into our approach as regularization over the domain-specific reviews classifiers to encourage the sharing of reviews information between similar domains. Two kinds of domain similarity measures are explored, one based on textual content and the other one based on reviews expressions. Moreover, we introduce two efficient algorithms to solve the model of our approach. Experimental results on benchmark datasets show that our approach can effectively improve the performance of multi-domain reviews classification and significantly outperform baseline methods.