Improving Chinese text categorization by outlier learning

Text categorization is one of the typical machine learning tasks that suffer from an incomplete training data problem. A main reason is the existence of outliers in training data, such as non-sense documents, documents mislabeled or lying on the border between different categories, and documents that are out of the defined categories, etc. Therefore, in a text categorization task, outlier learning technique could be adopted to improve text categorization. In this paper, an outlier learning based text categorization system is proposed, where AdaBoost algorithm is adopted for outlier identifying. Simulation results reveal that the new system is successful in improving learning performance for text categorization.Computer Science, Artificial IntelligenceComputer Science, Information SystemsCPCI-S(ISTP)