Building Effective Question Answering Characters (bibtex)
by Leuski, Anton, Patel, Ronakkumar, Traum, David and Kennedy, Brandon
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.
Reference:
Building Effective Question Answering Characters (Leuski, Anton, Patel, Ronakkumar, Traum, David and Kennedy, Brandon), In 7th SIGdial Workshop on Discourse and Dialogue, 2006.
Bibtex Entry:
@inproceedings{leuski_building_2006,
	address = {Sydney, Australia},
	title = {Building {Effective} {Question} {Answering} {Characters}},
	url = {http://ict.usc.edu/pubs/Building%20Effective%20Question%20Answering%20Characters.pdf},
	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.},
	booktitle = {7th {SIGdial} {Workshop} on {Discourse} and {Dialogue}},
	author = {Leuski, Anton and Patel, Ronakkumar and Traum, David and Kennedy, Brandon},
	month = jul,
	year = {2006},
	keywords = {Virtual Humans}
}
Powered by bibtexbrowser