Building effective question answering characters (2006)
| Venue: | In Proceedings of the 7th SIGdial Workshop on Discourse and Dialogue |
| Citations: | 25 - 17 self |
BibTeX
@INPROCEEDINGS{Leuski06buildingeffective,
author = {Anton Leuski and Ronakkumar Patel and David Traum},
title = {Building effective question answering characters},
booktitle = {In Proceedings of the 7th SIGdial Workshop on Discourse and Dialogue},
year = {2006},
pages = {18--27}
}
Bookmark
 |
 |
 |
 |
 |
 |
OpenURL
Abstract
In this paper, we describe methods for building and evaluation of limited domain question-answering characters. Several classification techniques are tested, including text classification using support vector machines, language-model based retrieval, and cross-language information retrieval techniques, with the latter having the highest success rate. We also evaluated the effect of speech recognition errors on performance with users, finding that retrieval is robust until recognition reaches over 50 % WER. 1
Citations
| 1370 | Latent Dirichlet allocation - Blei, Ng, et al. |
| 258 | Eliza: a computer program for the study of natural language communication between man and machine - Weizenbaum - 1966 |
| 236 | Viewing morphology as an inference process - Krovetz - 1993 |
| 216 | Training algorithms for linear text classi ers - Lewis, Shapire, et al. - 1996 |
| 211 | Support vector machine learning for interdependent and structured output spaces - Tsochantaridis, Hofmann, et al. - 2004 |
| 67 | Text segmentation by topic - Ponte, Croft - 1997 |
| 61 | Vector-based natural language call routing - Chu-Carroll, Carpenter - 1999 |
| 47 | INQUERY does battle with TREC-6 - Allan, Callan, et al. - 1998 |
| 38 | A Generative Theory of Relevance - Lavrenko |
| 18 | Probabilistic Latent Semantic Indexing - Hoffman - 1999 |
| 10 | Building topic specific language models from webdata using competitive models - Sethy, Georgiou, et al. - 2005 |
