A new method for creating a chain diagram of events that occur during disasters by extracting causal knowledge from Japanese newspaper articles and designing a causal network is proposed herein. Machine learning discriminant models were created for both conventional cue phrases and succession expressions with causation to extract causal sentences. We found that causal sentences can be extracted with a certain degree of accuracy from disaster articles. We were also able to create a causal network using sentences as nodes and links. The chain diagram using our new method extracted events and causal knowledge that were unavailable in a disaster chain diagram designed using conventional methods.

1. [1] Kajimatoshibosaikenkyukai, “Zishinbousai to Anzentoshi,” Kajima Institute Publishing Co., Ltd., 1996 (in Japanese).
2. [2] C. S. G. Khoo, J. Kornfilt, R. N. Oddy, and S. H. Myaeng, “Automatic Extraction of Cause-Effect Information from Newspaper Text Without Knowledge-Based Inferencing,” Literary and Linguistic Computing, Vol.13, No.4, pp. 177-186, 1998.
3. [3] R. Girju, “Automatic Detection of Causal Relations for Question Answering,” Proc. of the ACL 2003 Workshop on Multilingual Summarization and Question Answering (MultiSumQA’03), pp. 76-83, 2003.
4. [4] H. Sakaji and S. Masuyama, “A Method for Extracting Sentences Including Causal Relations from Newspaper Articles,” The IEICE Trans. on Information and Systems (Japanese Edition), Vol.J94-D, No.8, pp. 1496-1506, 2011 (in Japanese).
5. [5] H. Sakaji, H. Sakai, and S. Masuyama, “An Extraction Method of Causal Knowledge from Newspaper Corpus,” The J. of the Faculty of Science and Technology, Seikei University, Vol.51, No.2, pp. 23-28, 2014 (in Japanese with English abstract).
6. [6] H. Sakaji, R. Murono, H. Sakai, J. Bennett, and K. Izumi, “Discovery of Rare Causal Knowledge from Financial Statement Summaries,” Proc. of the 2017 IEEE Symp. Series on Computational Intelligence (SSCI), pp. 602-608, 2017.
7. [7] F. Sakahira and U Hiroi, “Study on the Method of Creating a Real-Time Disaster Chain Diagram,” Proc. of the 21st Conf. Japan Society for Disaster Information Studies, pp. 96-97, 2019 (in Japanese).
8. [8] T. Sato and M. Horita, “Assessing the Plausibility of Inference Based on Automated Construction of Causal Networks Using Web-mining,” Sociotechnica, Vol.4, pp. 66-74, 2006 (in Japanese).
9. [9] H. Aono and M. Ohta, “Construction of a Causal Network by Searching Factors,” IPSJ SIG Technical Report, Vol.2009-DBS-149, No.9, 8pp., 2009 (in Japanese with English abstract).
10. [10] H. Ishii, Q. Ma, and M. Yoshikawa, “Causal Network Construction using SVO Structure,” IPSJ SIG Technical Report, Vol.2009-DBS-149, No.10, 8pp., 2009 (in Japanese with English abstract).
11. [11] K. Nishimura, H. Sakaji, and K. Izumi, “Creation of Causal Relation Network Using Semantic Similarity,” The 32nd Annual Conf. of the Japanese Society for Artificial Intelligence, Session No.1P1-04, 4pp., 2018 (in Japanese with English abstract).
12. [12] T. Fujimori, M. Koyama, and J. Kiyono, “A Study of Circumstances of Disaster Victims According to Multiple Attributes Using Text Mining Method for Newspaper Articles Related to the 2011 Great East Japan Earthquake,” J. of Social Safety Science, No.23, pp. 55-64, 2014 (in Japanese with English abstract).
13. [13] H. Kato, N. Nojima, M. Koyama, and K. Tanaka, “Text Mining of Newspaper Articles Related to Lifeline Damage in the 2016 Kumamoto Earthquake: Comparison of Regional and National Daily Newspapers,” J. of Japan Society of Civil Engineers, Ser. A1 (Structural Engineering & Earthquake Engineering (SE/EE)), Vol.75. No.4, pp. I_443-I_453, 2019 (in Japanese with English abstract).
14. [14] The Asahi Shimbun Company, http://www.asahi.com/information/cd/gakujutsu.html [accessed April 17, 2021]
15. [15] National Institute of Information and Communications Technology, http://compling.hss.ntu.edu.sg/wnja/ [accessed April 17, 2021]
16. [16] A. Kobayashi, S. Masuyama, and S. Sekine, “A Method for Automatic Ontology Construction Using Wikipedia,” The IEICE Trans. on Information and Systems (Japanese Edition), Vol.J93-D, No.12, pp. 2597-2609, 2010 (in Japanese).
17. [17] CaboCha: Yet Another Japanese Dependency Structure Analyzer, https://taku910.github.io/cabocha/ [accessed April 17, 2021]
18. [18] MeCab: Yet Another Part-of-Speech and Morphological Analyzer, https://taku910.github.io/mecab/ [accessed April 17, 2021]
19. [19] GiNZA - Japanese NLP Library version 3.1.0, https://megagonlabs.github.io/ginza/ [accessed April 17, 2021]
20. [20] J. Devlin, M.-W. Chang, K. Lee, and K. Toutanova, “BERT: Pre-Training of Deep Bidirectional Transformers for Language Understanding,” Proc. of the 2019 Conf. of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Vol.1, pp. 4171-4186, 2019.
21. [21] Y. Niki, H. Sakaji, K. Izumi, and H. Matsushima, “Further Pretraining BERT for Causality Existence Classification in Financial Domain,” The 34th Annual Conf. of the Japanese Society for Artificial Intelligence, Session No.3Rin4-39, 4pp., 2020 (in Japanese with English abstract).