Text Categorization for Service Request Classification
Yu-Chuan Huang 1, Liang-Chih Yu 2, and
I-Chun Lin 3
1. Manz Taiwan Limited / IT Department, Chung-Li, Taiwan, R.O.C.
2. Yuan Ze University / Department of Information Management, Chung-Li, Taiwan, R.O.C.
3. National Yunlin University of Science and Technology / Department of Industrial Management, Yunlin, Taiwan, R.O.C
2. Yuan Ze University / Department of Information Management, Chung-Li, Taiwan, R.O.C.
3. National Yunlin University of Science and Technology / Department of Industrial Management, Yunlin, Taiwan, R.O.C
Abstract—Maintaining high levels of customer satisfaction is an essential problem for many companies. Communication with internal customers is usually accomplished using electronic documents, but important processes can be delayed if documents are sent to the wrong person. This paper describes the use of text categorization technology to analyze internal service requests with the aim to reduce the probability of document delivery error. Testing is performed using the Bayesian Classifier, Support Vector Machine and Decision Tree Classifier. Experimental results indicate that the use of such techniques can reduce errors and thus reduce delays. In addition, accuracy is shown to be improved by reducing the number of categories of computer problems.
Index Terms—Text categorization, service request classification, natural language processing
Cite: Yu-Chuan Huang, Liang-Chih Yu, and I-Chun Lin, "Text Categorization for Service Request Classification," International Journal of Signal Processing Systems, Vol. 1, No. 1, pp. 54-58, June 2013.doi: 10.12720/ijsps.1.1.54-58
Index Terms—Text categorization, service request classification, natural language processing
Cite: Yu-Chuan Huang, Liang-Chih Yu, and I-Chun Lin, "Text Categorization for Service Request Classification," International Journal of Signal Processing Systems, Vol. 1, No. 1, pp. 54-58, June 2013.doi: 10.12720/ijsps.1.1.54-58