Development of a Multi-Classifier Approach for Multilingual Text Categorization

- Research work related to applying text categorization methods to a monolingual corpus such as English text collections has been well established by several research teams in recent years. However, little attention has been paid to applying the techniques to classify the documents in multiple languages such as English and Chinese by means of a unified model. In this paper we propose a multi-classifier system platform to enable multilingual documents be effectively categorized. First, we utilized a number of selected corpora in multiple languages collected from internet to train several text classifiers based on the Support Vector Machines (SVM) model. Subsequently, the multilingual texts of unknown category were classified by the trained classifiers. Finally, we evaluated our experimental results by accuracy, recall, precision, and F1 measures. The preliminary results show that our platform model has the potential for multilingual text categorization.

• Type: Conference
• Year: 2006
• Where: DMIN
• Authors: Chung-Hong Lee, Hsin-Chang Yang, Ting-Chung Chen, Sheng-Min Ma
• Keywords: DMIN 2006, DMIN 2007, Multiple Languages, Text Categorization, Text Categorization Methods