Temporal Information Extraction from Korean Texts

Proceedings of the 19th Conference on Computational Language Learning, pages 279–288, Beijing, China, July 30-31, 2015. c⃝2015 Association for Computational Linguistics Temporal Information Extraction from Korean Texts Young-Seob Jeong, Zae Myung Kim, Hyun-Woo Do, Chae-Gyun Lim, and Ho-Jin Choi School of Computing Korea Advanced Institute of Science and Technology 291 Daehak-ro, Yuseong-gu, Daejeon 305-701, South Korea {pinode,zaemyung,realstorm103,rayote,hojinc}@kaist.ac.kr Abstract As documents tend to contain temporal information, extracting such information is attracting much research interests re- cently. In this paper, we propose a hybrid method that combines machine-learning models and hand-crafted rules for the task of extracting temporal information from unstructured Korean texts. We address Korean-specific research issues and pro- pose a new probabilistic model to generate complementary features. The performance of our approach is demonstrated by exper- iments on the TempEval-2 dataset, and the Korean TimeBank dataset which we built for this study. 1 Introduction Due to the increasing number of unstructured documents available on the Web and from other sources, developing techniques that automatically extract knowledge from the documents has been of paramount importance. Among many aspects of extracting knowledge from documents, the ex- traction of temporal information is recently draw- ing much attention, since the documents usually incorporate temporal information that is useful for further applications such as Information Re- trieval (IR) and Question Answering (QA) sys- tems. Given a simple question, “who was the pres- ident of the U.S. 8 years ago?”, for example, a QA system may have a difficulty in finding the right answer without the correct temporal informa- tion about when the question is posed and what ‘8 years ago’ refers to. There have been many studies for temporal in- formation extraction, but most of them are appli- cable only to their target languages. The main rea- son for this limitation is that some parts of tem- poral information are difficult to predict without the use of language-specific processing. For ex- ample, the normalized value ‘1983-03-08’ can be represented by ‘March 8, 1983’ in English, while it can be represented by ‘1983년삼월8일’ in Ko- rean. The order of date representation in Korean is usually different from that of English, and the digit expression in Korean is more complex than that of English. This implies that it is necessary to inves- tigate language-specific difficulties for developing a method to extract temporal information. In this paper, we propose a method for tem- poral information extraction from Korean texts. The contributions of this paper are as follows: we (1) show how the Korean-specific issues (e.g., morpheme-level tagging, various ways of digit ex- pression, uses of lunar calendar, and so on) are ad- dressed, (2) propose a hybrid method that com- bines a set of hand-crafted rules and machine- learning models, (3) propose a data-driven proba- bilistic model to generate complementary features, and (4) create a new dataset, the Korean Time- Bank, that consists of more than 3,700 manually annotated sentences. The rest of the paper is organized as follows. Section 2 describes the background of the re- search. …

Content truncated to 500 words