BROWSE
Detail
Conference Paper
Heterogeneous Answer Acquisition Methods in Encyclopedia QA
Cited 0 time in 
Share 
Share
- Authors
- Hyo-Jung Oh, Chung-Hee Lee, Chang ki Lee, Ji-Hyun Wang, Yi-Gyu Hwang, Hyeon-Jin Kim, Myung-Gil Jang
- Issue Date
- 2006-09
- Citation
- International Conference on Intelligent Data Engineering and Automated Learning (IDEAL) 2006 (LNCS 4224), v.4224, pp.346-354
- Publisher
- Springer
- Language
- English
- Type
- Conference Paper
- Abstract
- We propose an enhanced QA model with combination of heterogeneous answer acquisition methods. Our QA system is based on web encyclopedia in Korean. We investigated characteristic features of the encyclopedia and incorporate them in our answer acquisition methods. We defined three different types of answer extraction methods: learning-based, pattern-based, and traditional statistical methods. By empirical experiments, we obtained 59% improvement on MRR as well as 2.3 times speedy response. © Springer-Verlag Berlin Heidelberg 2006.
- KSP Keywords
- Answer Extraction, As 2, Characteristic features, Empirical experiments, Extraction method, Learning-based, Pattern-based, QA system, Statistical methods