Abstract
The Competition on legal information extraction/entailment (COLIEE) is an international information processing and retrieval competition. As an aid to future participants as well as question designers, this article describes how to connect legal questions taken from past Japanese bar exams to relevant statutes (articles of the Japanese Civil Code, Task 3) and how to construct a Yes/No question answering system for legal queries (Task 4) incorporating background materials on Japanese law. We restructured the given data to a dataset which contains all possible combinations of queries and articles as continuous strings as our samples. In this way, the difficult pairing task has been turned into a simpler classification task and samples for training became sufficient in number. Next, we used three BERT-based models to solve binary questions in order to achieve stable performance. As a result, the model achieved an F2-score of 0.6587 in Task 3 (ranked 1st) and an accuracy of 0.6161 in Task 4.
All authors contributed equally to this work.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Rabelo, J., Kim, M.-Y., Goebel, R., Yoshioka, M., Kano, Y., Satoh, K.: A summary of the COLIEE 2019 competition. In: Sakamoto, M., Okazaki, N., Mineshima, K., Satoh, K. (eds.) JSAI-isAI 2019. LNCS (LNAI), vol. 12331, pp. 34–49. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58790-1_3
Yoshioka, M., Kano, Y., Kiyota, N., Satoh, K.: Overview of Japanese statute law retrieval and entailment task at COLIEE-2018. In: The Proceedings of the 12th International Workshop on Juris-Informatics (JURISIN 2018), pp. 117–128. The Japanese Society of Artificial Intelligence (2018)
Kano, Y., et al.: COLIEE-2018: evaluation of the competition on legal information extraction and entailment. In: Kojima, K., Sakamoto, M., Mineshima, K., Satoh, K. (eds.) JSAI-isAI 2018. LNCS (LNAI), vol. 11717, pp. 177–192. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-31605-1_14
COLIEE organizer. COLIEE-2020. https://sites.ualberta.ca/~rabelo/COLIEE2020/
Cl-tohoku, BERT-base_mecab-ipadic-char-4k_whole-word-mask. https://github.com/cl-tohoku/bert-japanese
Alinear-corp, albert-japanese. https://github.com/alinear-corp/albert-japanese
Kingma, D.P., Ba, J.: Adam: a method for stochastic optimization. arXiv preprint arXiv:1412.6980 (2014)
Devlin, J., Chang, M.W., Lee, K., Toutanova, K.: BERT: pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805 (2018)
Clark, C., Lee, K., Chang, M.W., Kwiatkowski, T., Collins, M., Toutanova, K.: BoolQ: exploring the surprising difficulty of natural yes/no questions. arXiv preprint arXiv:1905.10044 (2019)
Barskar, R., Ahmed, G., Barskar, N.: An approach for extracting exact answers to question answering (QA) system for English sentences. Procedia Eng. 30, 1187–1194 (2012). https://doi.org/10.1016/j.proeng.2012.01.979
Li, X., et al.: Entity-relation extraction as multi-turn question answering. arXiv preprint arXiv:1905.05529 (2019)
Rabelo, J., Kim, M.Y., Goebel, R., Yoshioka, M., Kano, Y., Satoh, K.: COLIEE 2020: methods for legal document retrieval and entailment. In: Proceedings of the International Workshop on Juris-Informatics 2020 (JURISIN 2020), pp. 114–127 (2020)
Acknowledgments
This work was financially supported by: Hsuan-Lei Shao, “From Knowledge Genealogy to Knowledge Map-China Studies in Big Data and Machine Learning” (MOST 107-2410-H-003 -058 -MY3), Sieh-Chuen Huang, Center for Research in Econometric Theory and Applications (Grant no. NTU-110L900203) from The Featured Areas Research Center Program within the framework of the Higher Education Sprout Project by the Ministry of Education (MOE) and Ministry of Science and Technology (MOST 109-2634-F-002-045) in Taiwan.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Shao, HL., Chen, YC., Huang, SC. (2021). BERT-Based Ensemble Model for Statute Law Retrieval and Legal Information Entailment. In: Okazaki, N., Yada, K., Satoh, K., Mineshima, K. (eds) New Frontiers in Artificial Intelligence. JSAI-isAI 2020. Lecture Notes in Computer Science(), vol 12758. Springer, Cham. https://doi.org/10.1007/978-3-030-79942-7_15
Download citation
DOI: https://doi.org/10.1007/978-3-030-79942-7_15
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-79941-0
Online ISBN: 978-3-030-79942-7
eBook Packages: Computer ScienceComputer Science (R0)