Abstract

KorQuA: Answer Extraction by Flexible Matching, Filtering and Interpretation

Kyung-Soon Lee (National Institute of Informatics), Jae-Ho Kim, and Key-Sun Choi (Korea Advanced Institute of Science and Technology)



This paper describes a Korean Question Answering (KorQuA) sys-types, and retrieve passages by matching terms according to these data types. In answer extraction, we filter non-fact answers with negative or uncertain con-texts, and interpret a relative expression as an absolute date answer for date type flexibly according to their characteristics, answer filter make a system more re-liable, and answer interpretation make a system more intelligent. We also de-scribe construction of a Korean question answering test collection.