Customizing Sentiment Classifiers to New Domains: a Case Study

Submitted to RANLP-05, the International Conference on Recent Advances in Natural Language Processing |

Sentiment classification is a very domain specific problem; classifiers trained in one domain do not perform well in others. Unfortunately, many domains are lacking in large amounts of labeled data for fully-supervised learning approaches. At the same time, sentiment classifiers need to be customizable to new domains in order to be useful in practice. We attempt to address these difficulties and constraints in this paper, where we survey four different approaches to customizing a sentiment classification system to a new target domain in the absence of large amounts of labeled data. We base our experiments on data from four different domains. After establishing that naive cross-domain classification results in poor classification accuracy, we compare results obtained by using each of the four approaches and discuss their advantages, disadvantages and performance.