A Rough Set Method for Co-training Algorithm

In recent years, semi-supervised learning has been a hot research topic in machine learning area. Different from traditional supervised learning which learns only from labeled data; semi-supervised learning makes use of both labeled and unlabeled data for learning purpose. Co-training is a popular semi-supervised learning algorithm which assumes that each example is represented by two or more redundantly sufficient sets of features (views) and additionally these views are independent given the class. To improve the performance and applicability of co-training, ensemble learning, such as bagging and random subspace has been used along with co-training. In this work, we propose to use the rough set based ensemble learning method with co-training algorithm (RSCO). Inherited the inherent characteristics of rough set, ensemble learning is expected to meet both the diversity and accuracy requirement. Finally experimental results on the UCI data sets demonstrate the promising performance of RSCO.