Training a Naive Bayes Classifier via the EM Algorithm with a Class Distribution Constraint

Combining a naive Bayes classifier with the EM algorithm is one of the promising approaches for making use of unlabeled data for disambiguation tasks when using local context features including word sense disambiguation and spelling correction. However, the use of unlabeled data via the basic EM algorithm often causes disastrous performance degradation instead of improving classification performance, resulting in poor classification performance on average. In this study, we introduce a class distribution constraint into the iteration process of the EM algorithm. This constraint keeps the class distribution of unlabeled data consistent with the class distribution estimated from labeled data, preventing the EM algorithm from converging into an undesirable state. Experimental results from using 26 confusion sets and a large amount of unlabeled data show that our proposed method for using unlabeled data considerably improves classification performance when the amount of labeled data is small.