Journal of KIISE (정보과학회 논문지)

Korean Institute of Information Scientists and Engineers (한국정보과학회)
권호 표지
DOI QR코드

DOI QR Code

End-to-end Korean Document Summarization using Copy Mechanism and Input-feeding

복사 방법론과 입력 추가 구조를 이용한 End-to-End 한국어 문서요약

  • Received : 2016.12.08
  • Accepted : 2017.02.28
  • Published : 2017.05.15
⟨ Previous Next ⟩

Abstract

In this paper, the copy mechanism and input feeding are applied to recurrent neural network(RNN)-search model in a Korean-document summarization in an end-to-end manner. In addition, the performances of the document summarizations are compared according to the model and the tokenization format; accordingly, the syllable-unit, morpheme-unit, and hybrid-unit tokenization formats are compared. For the experiments, Internet newspaper articles were collected to construct a Korean-document summary data set (train set: 30291 documents; development set: 3786 documents; test set: 3705 documents). When the format was tokenized as the morpheme-unit, the models with the input feeding and the copy mechanism showed the highest performances of ROUGE-1 35.92, ROUGE-2 15.37, and ROUGE-L 29.45.

본 논문에서는 copy mechanism과 input feeding 추가한 RNN search 모델을 end-to-end 방식으로 한국어 문서요약에 적용하였다. 또한 시스템의 입출력으로 사용하는 데이터를 음절단위, 형태소단위, hybrid 단위의 토큰화 형식으로 처리하여 수행한 각각의 성능을 구하여, 모델과 토큰화 형식에 따른 문서요약 성능을 비교한다. 인터넷 신문기사를 수집하여 구축한 한국어 문서요약 데이터 셋(train set 30291 문서, development set 3786 문서, test set 3705문서)으로 실험한 결과, 형태소 단위로 토큰화 하였을 때 우수한 성능을 확인하였으며, GRU search에 input feeding과 copy mechanism을 추가한 모델이 ROUGE-1 35.92, ROUGE-2 15.37, ROUGE-L 29.45로 가장 높은 성능을 보였다.

Keywords

Acknowledgement

Grant : (엑소브레인-1 세부) 휴먼 지식증강 서비스를 위한 지능진화형 WiseQA 플랫폼 기술 개발

Supported by : 정보통신기술진흥센터

References

  1. Jae-Min Yoon, You-Jin Chung, Jong-Hyeok Lee, "Automatic Extractive Summarization of Newspaper Articles using Activation Degree of 5W1H," Journal of KIISE, SA, Vol. 31, No. 4, pp. 505-515, 2004.
  2. Chul-Won Kim, Sun Park, "Document Summarization using Pseudo Relevance Feedback and Term Weighting," Journal of Kiice, Vol. 16, No. 3, pp. 533-540, 2012.
  3. Sutskever, Ilya, Oriol Vinyals, and Quoc V. Le. "Sequence to sequence learning with neural networks," Advances in neural information processing systems, 2014.
  4. Luong, Minh-Thang, Hieu Pham, and Christopher D. Manning, "Effective approaches to attentionbased neural machine translation," arXiv preprint arXiv:1508.04025, 2015.
  5. Jiatao Gu, Zhengdong Lu, Hang Li, Victor O.K. Li. "Incorporating copying mechanism in sequence-tosequence learning," arXiv preprint arXiv:1603.06393, 2016.
  6. Lin, Chin-Yew, "ROUGE: a Package for Automatic Evaluation of Summaries," Proc. of the Workshop on Text Summarization Branches Out (WAS 2004), 2004.
  7. Dingding Wang, Tao Li, Shenghuo Zhu, Chris H. Q. Ding, "Multi-document summarization via sentence-level semantic analysis and symmetric matrix factorization," ACM SIGIR conference on Research and development in information retrieval, pp. 307-314, 2008.
  8. Insight, [Online]. Available: http://www.insight.co.kr/