Abstract
Most CRM systems include the text base response function through the Web, which apply the text mining technology. However, there is the critical problem of bad performance of the mining system; low hit rate of expected answers at the beginning stage. The problem is caused by limited knowledge in the system due to the lack of corpus and documents accumulated. Another cause is that the vocabulary is sometimes poor in the customer’s short questionnaire. The main purpose of this study is to improve the performance of mining systems by tuning from the user’s standpoint, not from the system provider. We experimented with a mining system. We populated corpus to the system and put some questions into the system repeatedly while changing corpus quantity and the effect of keywords. The results suggest that when the corpus quantity is not large enough, the system can be improved by repeating to input the same corpus several times.
Chapter PDF
Similar content being viewed by others
Keywords
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
References
Koudai Aman and Fukuya Ishino, Performance Improvement of Text Mining, General Conference of IEICE, 2006, p. 44.
Koudai Aman, Satoshi Watanabe and Fukuya Ishino, A Proposal for Better Performance of Text Mining, in: Proceedings of International Conference on Operations and Supply Chain Management, Bali International Convention Center, The Westin Resort, Nusa Dua, Bali, 2005, pp. 25–33.
Yasuhiro Takayama, Text mining based on Concept Extraction for eCRM, Technical report of IEICE No. 3, 2003, pp. 19–23.
J. L. Neto, A. D. Santos, C. A. A. Kaestner and A. A. Freitas, Document clustering and text summarization, in: 4 th International Conference on Practical Applications of Knowledge Discovery and Data Mining (PADD-2000), (The Practical Application Company, London, 2000), pp. 41–55.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2006 International Federation for Information Processing
About this paper
Cite this paper
Aman, K., Ishino, F. (2006). Application of Text Mining into Auto Answering System and Improvement of Mining Performance. In: Suomi, R., Cabral, R., Hampe, J.F., Heikkilä, A., Järveläinen, J., Koskivaara, E. (eds) Project E-Society: Building Bricks. IFIP International Federation for Information Processing, vol 226. Springer, Boston, MA. https://doi.org/10.1007/978-0-387-39229-5_14
Download citation
DOI: https://doi.org/10.1007/978-0-387-39229-5_14
Publisher Name: Springer, Boston, MA
Print ISBN: 978-0-387-39226-4
Online ISBN: 978-0-387-39229-5
eBook Packages: Computer ScienceComputer Science (R0)