상대적인 시간정보의 규칙기반 정규화
Rule-based Normalization of Relative Temporal Information
2
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
문서에는 상대적인 시간표현이 포함되어있으며, 이를 다루기 위한 시간표현 체계를 구축하고 상대시간정보를 추출하는 모델을 개발하는 것은 중요하다. 본 연구에서는 timex3 태그에서 상대적인시간표현의 정규값을 담을 수 있도록 하기 위해 연, 월, 일, 주, 시, 분, 초 단위에 대하여 총 7가지의새로운 속성을 새롭게 제시하였으며, 이전, 이후, 번째 등에 대한 정규값의 기술방법도 함께 제시하였다. 또한, 새롭게 추가된 속성들의 정규값을 추출하는 규칙 모음을 설계하였다. 추가된 속성들을바탕으로 구축한 데이터셋은 일상대화, 뉴스, 역사와 관련된 총 1,041개의 문서를 포함하고 있으며, 본 연구에서 설계한 규칙 모음을 전체 데이터셋에 대하여 적용하여 전반적으로 70% 이상의 정확도를 보이는 것을 확인하였다. 특히, 데이터셋에 자주 등장한 상대시간표현인 year, day, week 속성에대한 성능이 비교적 높은 것을 확인할 수 있었다. 본 연구의 결과물인 추가적인 timex3 속성과 규칙기반 모델은 질의응답시스템, 챗봇 등의 서비스 개발에 유용하게 활용될 수 있을 것이다.
Abstract
Documents often contain relative time expressions, and it is important to define a schema of the relative time information and develop a system that extracts such information from corpus. In this study, to deal with the relative time expressions, we propose seven additional attributes of timex3: year, month, day, week, hour, minute, and second. We propose a way to represent normalized values of the relative time expressions such as before, after, and count, and also design a set of rules to extract the relative time information from texts. With a new corpus constructed using the new attributes that consists of dialog, news, and history documents, we observed that our rule-set generally achieved 70% accuracy on the 1,041 documents. Especially, with the most frequently appeared attributes such as year, day, and week, we got higher accuracies compared to other attributes. The results of this study, our proposed timex3 attributes and the rule-set, will be useful in the development of services such as question-answer systems and chatbots.
참고문헌 (17)
- "Naver Clova"[URL]
- "KT Giga gini"[URL]
- "IBM Watson"[URL]
- James Pustejovsky, "ISO-TimeML: An International Standard for Semantic Annotation", Proceedings of the Seventh International Conference on Language Resources and Evaluation, pp.394~397 (2010)
- 유현조, "한국어 TimeML - 텍스트의 사건 및 시간 정보 연구", 언어와 정보, 15(1), pp.31~62 (2011), 한국언어정보학회
- Young-Seob Jeong, "Korean TimeML and Korean TimeBank", Proceedings of the Tenth International Conference on Language Resources and Evaluation, pp.356~359 (2016)
- Tommaso Caselli, "Annotating Events, Temporal Expressions and Relations in Italian: the It-Timeml Experience for the Ita-TimeBank", Proceedings of the 5th Linguistic Annotation Workshop, pp.143~151 (2011)
- Arman Cohan, "Pretrained Language Models for Sequential Sentence Classification", Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing, pp.3693~3699 (2019)
- Gati L. Martin, "SwahBERT: Language Model of Swahili", Proceedings of the Annual Conference of the North American Chapter of the Association for Computational Linguistics, pp.314~324 (2022)
- Devendra Singh Sachan, "Effective Use of Bidirectional Language Modeling for Transfer Learning in Biomedical Named Entity Recognition", Proceedings of the Machine Learning for Healthcare Conference, pp.383~402 (2018)
- Touseef Iqbal, "The survey: Text generation models in deep learning", Journal of King Saud University -Computer and Information Sciences, 34(6), pp.2515~2528 (2020)[DOI]
- Chitwan Saharia, "Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding", ~ (2022)
- Alec Radford, "Robust Speech Recognition via Large-Scale Weak Supervision", ~ (2022)
- Ashish Vaswani, "Attention Is All You Need", Proceedings of the 31st Conference on Neural Information Processing Systems (2017)
- Aitor Lewkowycz, "Solving Quantitative Reasoning Problems with Language Models", ~ (2022)
- "ISO 8601"[URL]
- Young-Seob Jeong, "Temporal Information Extraction from Korean Texts", Proceedings of the 19th SIGNLL Conference on Computational Natural Language Learning, pp.279~288 (2015)