In a question-answering (QA) task, "real" information retrieval, rather than document retrieval is required. Effective QA thus involves complicated and time-consuming processing, such as natural language processing and named-entity processing. To reduce the amount of processing needed, the quantity of documents in a database can be narrowed down during an initial stage of the answering procedure. This paper proposes a new evaluation measurement and compares the retrieval accuracy of initial-stage searching that uses "overall" document retrieval and "partial" passage retrieval with the TREC QA data set. The initial search and final result accuracy for various cutoff points defined according to the number of documents or words that are output is evaluated. A variety of experiments demonstrate that middle-length passage-retrieval is effective for QA, and short-length passage-retrieval could improve the accuracy of the final result for a specific question type.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Toru TAKAKI, "Comparison of Document Retrieval and Passage Retrieval Using a New Measurement in an Initial Narrowing-Down Search for Question Answering" in IEICE TRANSACTIONS on Information,
vol. E86-D, no. 9, pp. 1648-1657, September 2003, doi: .
Abstract: In a question-answering (QA) task, "real" information retrieval, rather than document retrieval is required. Effective QA thus involves complicated and time-consuming processing, such as natural language processing and named-entity processing. To reduce the amount of processing needed, the quantity of documents in a database can be narrowed down during an initial stage of the answering procedure. This paper proposes a new evaluation measurement and compares the retrieval accuracy of initial-stage searching that uses "overall" document retrieval and "partial" passage retrieval with the TREC QA data set. The initial search and final result accuracy for various cutoff points defined according to the number of documents or words that are output is evaluated. A variety of experiments demonstrate that middle-length passage-retrieval is effective for QA, and short-length passage-retrieval could improve the accuracy of the final result for a specific question type.
URL: https://global.ieice.org/en_transactions/information/10.1587/e86-d_9_1648/_p
Copy
@ARTICLE{e86-d_9_1648,
author={Toru TAKAKI, },
journal={IEICE TRANSACTIONS on Information},
title={Comparison of Document Retrieval and Passage Retrieval Using a New Measurement in an Initial Narrowing-Down Search for Question Answering},
year={2003},
volume={E86-D},
number={9},
pages={1648-1657},
abstract={In a question-answering (QA) task, "real" information retrieval, rather than document retrieval is required. Effective QA thus involves complicated and time-consuming processing, such as natural language processing and named-entity processing. To reduce the amount of processing needed, the quantity of documents in a database can be narrowed down during an initial stage of the answering procedure. This paper proposes a new evaluation measurement and compares the retrieval accuracy of initial-stage searching that uses "overall" document retrieval and "partial" passage retrieval with the TREC QA data set. The initial search and final result accuracy for various cutoff points defined according to the number of documents or words that are output is evaluated. A variety of experiments demonstrate that middle-length passage-retrieval is effective for QA, and short-length passage-retrieval could improve the accuracy of the final result for a specific question type.},
keywords={},
doi={},
ISSN={},
month={September},}
Copy
TY - JOUR
TI - Comparison of Document Retrieval and Passage Retrieval Using a New Measurement in an Initial Narrowing-Down Search for Question Answering
T2 - IEICE TRANSACTIONS on Information
SP - 1648
EP - 1657
AU - Toru TAKAKI
PY - 2003
DO -
JO - IEICE TRANSACTIONS on Information
SN -
VL - E86-D
IS - 9
JA - IEICE TRANSACTIONS on Information
Y1 - September 2003
AB - In a question-answering (QA) task, "real" information retrieval, rather than document retrieval is required. Effective QA thus involves complicated and time-consuming processing, such as natural language processing and named-entity processing. To reduce the amount of processing needed, the quantity of documents in a database can be narrowed down during an initial stage of the answering procedure. This paper proposes a new evaluation measurement and compares the retrieval accuracy of initial-stage searching that uses "overall" document retrieval and "partial" passage retrieval with the TREC QA data set. The initial search and final result accuracy for various cutoff points defined according to the number of documents or words that are output is evaluated. A variety of experiments demonstrate that middle-length passage-retrieval is effective for QA, and short-length passage-retrieval could improve the accuracy of the final result for a specific question type.
ER -