Hot-spot passage retrieval in question answering

Authors:
Jian Huang;Xuanjing Huang;Lide Wu
Affiliations:
Dept. of Computer Science and Engineering, Fudan University, Shanghai, P.R. China;Dept. of Computer Science and Engineering, Fudan University, Shanghai, P.R. China;Dept. of Computer Science and Engineering, Fudan University, Shanghai, P.R. China
Venue:
ICADL'04 Proceedings of the 7th international Conference on Digital Libraries: international collaboration and cross-fertilization
Year:
2004

Citing 5
Cited 0

Passage-level evidence in document retrieval

SIGIR '94 Proceedings of the 17th annual international ACM SIGIR conference on Research and development in information retrieval
Scaling question answering to the web

ACM Transactions on Information Systems (TOIS)
On the MSE robustness of batching estimators

Proceedings of the 33nd conference on Winter simulation
Performance issues and error analysis in an open-domain question answering system

ACM Transactions on Information Systems (TOIS)
Quantitative evaluation of passage retrieval algorithms for question answering

Proceedings of the 26th annual international ACM SIGIR conference on Research and development in informaion retrieval

Quantified Score

Hi-index	0.00

Visualization

Abstract

Question Answering has been the recent focus of information retrieval research; many systems just incorporate a search engine as a black box and most effort has been devoted to the question analyzer and the answer identifier. In the context of QA, however, passage provides an ideal medium between the document collection and an exact answer. And passage retrieval is a finer-grain approach than the traditional document retrieval both for the answer identifier and a human reader. In this paper, distinctions are first made between document retrieval and passage retrieval. And the Hot-Spot Passage Retrieval algorithm, which takes into account the measures of blurred BM25, coverage and height, is examined in detail. For evaluation, an isolated test is conducted and the algorithm gains 18.3% better answer redundancy and 4.8% better coverage rate than Okapi's original passage retrieval algorithm.