Journal of the Korea Convergence Society (한국융합학회논문지)

Korea Convergence Society (한국융합학회)
권호 표지
DOI QR코드

DOI QR Code

Meta Learning based Global Relation Extraction trained by Traditional Korean data

전통 문화 데이터를 이용한 메타 러닝 기반 전역 관계 추출

  • Kim, Kuekyeng (Department of Computer Science and Engineering, Korea University) ;
  • Kim, Gyeongmin (Department of Computer Science and Engineering, Korea University) ;
  • Jo, Jaechoon (Department of Computer Science and Engineering, Korea University) ;
  • Lim, Heuiseok (Department of Computer Science and Engineering, Korea University)
  • Received : 2018.08.31
  • Accepted : 2018.11.20
  • Published : 2018.11.28
Download PDF
⟨ Previous Next ⟩

Abstract

Recent approaches to Relation Extraction methods mostly tend to be limited to mention level relation extractions. These types of methods, while featuring high performances, can only extract relations limited to a single sentence or so. The inability to extract these kinds of data is a terrible amount of information loss. To tackle this problem this paper presents an Augmented External Memory Neural Network model to enable Global Relation Extraction. the proposed model's Global relation extraction is done by first gathering and analyzing the mention level relation extraction by the Augmented External Memory. Additionally the proposed model shows high level of performances in korean due to the fact it can take the often omitted subjects and objectives into consideration.

최근 존재하는 대부분의 관계 추출 모델은 언급 수준의 관계 추출 모델이다. 이들은 성능은 높지만, 장문의 텍스트에 존재하는 다수의 문장을 처리할 때, 문서 내에 주요 개체 및 여러 문장에 걸쳐서 표현되는 전역적 개체 관계를 파악하지 못한다. 그리고 이러한 높은 수준의 관계를 정의하지 못하는 것은 데이터의 올바른 정형화를 막는 중대한 문제이다. 이 논문에서는 이러한 문제를 해결하고 전역적 관계를 추출하기 위하여 외부 메모리 신경망 모델을 이용하는 새로운 방식의 전역관계 추출 모델을 제안한다. 제안하는 모델은 1차적으로는 단편적인 관계 추출을 실행한 뒤, 외부메모리 신경망을 이용하여 단편적인 관계들을 분석 및 종합하여 텍스트 전체로부터 전역적 관계들을 추출한다. 또한 제안된 모델은 외부 메모리를 통하여 전역적 관계 추출 외에도 주어와 목적어 생략이 잦은 한국어 관계 추출에도 뛰어난 성능을 보인다.

Keywords

OHHGBW_2018_v9n11_23_f0001.png 이미지

Fig. 1. LSTM based Local Relation Extraction

OHHGBW_2018_v9n11_23_f0002.png 이미지

Fig. 2. Meta Learning via Augmented External Memory Neural Network

Table 1. Performances Comparisons on Local and Global Relation Extraction

OHHGBW_2018_v9n11_23_t0001.png 이미지

References

  1. Konstantinova, N. (2014, April). Review of relation extraction methods: What is new out there?. In International Conference on Analysis of Images, Social Networks and Texts_x000D_Springer, Cham, 15-28.
  2. Sujeong Kwak, Bogyum Kim, Jae Sung Lee. (2014). Considerations for Relation Extraction using Korean Dependency Parsing. Proceedings of the KISS conference, 831-833.
  3. Pawar, S., Palshikar, G. K., & Bhattacharyya, P. (2017). Relation Extraction: A Survey. arXiv preprint arXiv:1712.05191.
  4. Moschitti, A., Patwardhan, S., & Welty, C. (2013). Long-distance time-event relation extraction. In Proceedings of the Sixth International Joint Conference on Natural Language Processing, 1330-1338.
  5. Zeng, X., He, S., Liu, K., & Zhao, J. (2018). Large scaled relation extraction with reinforcement learning. Relation, 2, 3.
  6. Vanschoren, J. (2018). Meta-Learning: A Survey. arXiv preprint arXiv:1810.03548.
  7. Lemke, C., Budka, M., & Gabrys, B. (2015). Metalearning: a survey of trends and technologies. Artificial intelligence review, 44(1), 117-130. https://doi.org/10.1007/s10462-013-9406-y
  8. D. Y. Lee, W. H. Yu & H. S. Lim. (2017). Bi-directional LSTM-CNN-CRF for Korean Named Entity Recognition System with Feature Augmentation. Journal of the Korea Convergence Society, 8(12), 55-62. https://doi.org/10.15207/JKCS.2017.8.12.055
  9. Shen, W., Wang, J., & Han, J. (2015). Entity linking with a knowledge base: Issues, techniques, and solutions. IEEE Transactions on Knowledge and Data Engineering, 27(2), 443-460. https://doi.org/10.1109/TKDE.2014.2327028
  10. Sorokin, D., & Gurevych, I. (2017). Context-Aware Representations for Knowledge Base Relation Extraction. In Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, 1784-1789.
  11. Santoro, Adam, et al. "Meta-learning with memory-augmented neural networks." International conference on machine learning. 2016.
  12. Hyun, Kim et al. 2017. "Digital humanities - Research Administration of Archives and Humanities" Proceedings at Digital humanities. DOI: http://dh.aks.ac.kr/Encyves/wiki
  13. Nguyen, T. H., & Grishman, R. (2015). Relation extraction: Perspective from convolutional neural networks. In Proceedings of the 1st Workshop on Vector Space Modeling for Natural Language Processing (pp. 39-48).
  14. Miwa, M., & Bansal, M. (2016). End-to-end relation extraction using lstms on sequences and tree structures. arXiv preprint arXiv:1601.00770.
  15. Han, X., & Sun, L. (2016, March). Global distant supervision for relation extraction. In Thirtieth AAAI Conference on Artificial Intelligence.