Extracting Natech Reports from Large Databases: Development of a Semi-Intelligent Natech Identification Framework

Abstract

Natural hazard-triggered technological accidents (Natechs) refer to accidents involving releases of hazardous materials (hazmat) triggered by natural hazards. Huge economic losses, as well as human health and environmental problems are caused by Natechs. In this regard, learning from previous Natechs is critical for risk management. However, due to data scarcity and high uncertainty concerning such hazards, it becomes a serious challenge for risk managers to detect Natechs from large databases, such as the National Response Center (NRC) database. As the largest database of hazmat release incidents, the NRC database receives hazmat release reports from citizens in the United States. However, callers often have incomplete details about the incidents they are reporting. This results in many records having incomplete information. Consequently, it is quite difficult to identify and extract Natechs accurately and efficiently. In this study, we introduce machine learning theory into the Natech retrieving research, and a Semi-Intelligent Natech Identification Framework (SINIF) is proposed in order to solve the problem. We tested the suitability of two supervised machine learning algorithms, namely the Long Short-Term Memory (LSTM) and the Convolutional Neural Network (CNN), and selected the former for the development of the SINIF. According to the results, the SINIF is efficient (a total number of 826,078 records were analyzed) and accurate (the accuracy is over 0.90), while 32,841 Natech reports between 1990 and 2017 were extracted from the NRC database. Furthermore, the majority of those Natech reports (97.85%) were related to meteorological phenomena, with hurricanes (24.41%), heavy rains (19.27%), and storms (18.29%) as the main causes of these reported Natechs. Overall, this study suggests that risk managers can benefit immensely from SINIF in analyzing Natech data from large databases efficiently.

Keywords

Data extraction method / Machine learning / Natechs / Natural hazards / NRC database

References

[1] Antonioni G, Bonvicini S, Spadoni G, Cozzani V. Development of a framework for the risk assessment of Na-Tech accidental events. Reliability Engineering and System Safety, 2009, 94(9): 1442-1450.

[2] Antonioni G, Landucci G, Necci A, Gheorghiu D, Cozzani V. Quantitative assessment of risk due to NaTech scenarios caused by floods. Reliability Engineering and System Safety, 2015, 142: 334-345.

[3] Antonioni G, Spadoni G, Cozzani V. A methodology for the quantitative risk assessment of major accidents triggered by seismic events. Journal of Hazardous Materials, 2007, 147(1–2): 48-59.

[4] Bashar, S.S., and A.A. Mahmud. 2019. A machine learning approach for heart rate estimation from PPG signal using random forest regression algorithm. In Proceedings of 2019 International Conference on Electrical, Computer and Communication Engineering (ECCE), 24–25 July 2019, Swat, Pakistan, 1–5.

[5] Bishop CM. Pattern recognition and machine learning, 2006, Berlin: Springer

[6] Blanzieri E, Bryl A. A survey of learning-based techniques of email spam filtering. Artificial Intelligence Review, 2008, 29: 63-92.

[7] Bureau for Analysis of Industrial Risk and Pollution. 2019. Analysis, Research and Information on Accidents (ARIA). French Ministry of Ecology and Sustainable Development, Bureau for Analysis of Industrial Risk and Pollution, France. https://www.aria.developpement-durable.gouv.fr. Accessed 10 Jan 2019 (in French).

[8] Campadelli P, Lanzarotti R, Lipori G. Face detection in color images of generic scenes. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2004, 24: 97-103.

[9] Campedel M, Cozzani V, Garcia-Agreda A, Salzano E. Extending the quantitative assessment of industrial risks to earthquake effects. Risk Analysis, 2008, 28: 1231-1246.

[10] Chowdhury GG. Natural language processing. Annual Review of Information Science and Technology, 2003, 37: 51-89.

[11] Cireşan, D., U. Meier, and J. Schmidhuber. 2012. Multi-column deep neural networks for image classification. arXiv preprint arXiv:1202.2745.

[12] Clow JC. The Coast Guard’s National Response Center. Environment International, 1980, 3(2): 151-153.

[13] Cozzani V, Campedel M, Renni E, Krausmann E. Industrial accidents triggered by flood events: Analysis of past accidents. Journal of Hazardous Materials, 2010, 175: 501-509.

[14] Cruz AM, Krausmann E. Hazardous-materials releases from offshore oil and gas facilities and emergency response following Hurricanes Katrina and Rita. Journal of Loss Prevention in the Process Industries, 2009, 22(1): 59-65.

[15] Cruz AM, Okada N. Consideration of natural hazards in the design and risk management of industrial facilities. Natural Hazards, 2008, 44(2): 213-227.

[16] Cruz AM, Steinberg LJ, Vetere-Arellano AL. Emerging issues for natech disaster risk management in Europe. Journal of Risk Research, 2006, 9(5): 483-501.

[17] Diale M, Celik T, Van Der Walt C. Unsupervised feature learning for spam email filtering. Computers and Electrical Engineering, 2019, 74: 89-104.

[18] Emanuel K. Environmental factors affecting tropical cyclone power dissipation. Journal of Climate, 2007, 20(22): 5497-5509.

[19] European Commission. 2019a. Natech accident database. Joint Research Centre, Institute for the Protection and Security of the Citizen, Italy. http://enatech.jrc.ec.europa.eu. Accessed 11 Jan 2019.

[20] European Commission. 2019b. eMars (Major Accident Reporting System) database. European Commission, Joint Research Centre, Institute for the Protection and Security of the Citizen, Italy. http://emars.jrc.ec.europa.eu. Accessed 11 Jan 2019.

[21] Fernández, S., A. Graves, and J. Schmidhuber. 2007. An application of recurrent neural networks to discriminative keyword spotting. In Proceedings of International Conference on Artificial Neural Networks, 9–13 September 2007, Porto, Portugal, 220–229.

[22] Fukushima K. Neocognitron: A self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in position. Biological Cybernetics, 1980, 36(4): 193-202.

[23] Girgin, S., and E. Krausmann. 2014. Analysis of pipeline accidents induced by natural hazards: Final report. JRC88410. Joint Research Centre, European Commission, Italy.

[24] Girgin S, Krausmann E. Historical analysis of U.S. onshore hazardous liquid pipeline accidents triggered by natural hazards. Journal of Loss Prevention in the Process Industries, 2016, 40: 578-590.

[25] Graves, A., and J. Schmidhuber. 2009. Offline handwriting recognition with multidimensional recurrent neural networks. In Proceedings of the 21st International Conference on Neural Information Processing Systems, ed. D. Koller, D. Schuurmans, Y. Bengio, and L. Bottou, 545–552. Red Hook, NY: Curran Associates Inc.

[26] Hochreiter S, Schmidhuber J. Long short-term memory. Neural Computation, 1997, 9(8): 1735-1780.

[27] Huang, G.-B., H. Zhou, X. Ding, and R. Zhang. 2011. Extreme learning machine for regression and multiclass classification. IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics) 42(2): 513–529.

[28] Ji S, Xu W, Yang M, Yu K. 3D convolutional neural networks for human action recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(1): 221-231.

[29] Khakzad N, Van Gelder P. Fragility assessment of chemical storage tanks subject to floods. Process Safety and Environmental Protection, 2017, 111: 75-84.

[30] Khakzad N, Van Gelder P. Vulnerability of industrial plants to flood-induced natechs: A Bayesian network approach. Reliability Engineering and System Safety, 2018, 169: 403-411.

[31] Khakzad N, Dadashzadeh M, Reniers G. Quantitative assessment of wildfire risk in oil facilities. Journal of Environmental Management, 2018, 223: 433-443.

[32] Knutson TR, Tuleya RE, Garner ST, Bender MA, Vecchi GA, Held IM, Sirutis JJ. Modeled impact of anthropogenic warming on the frequency of intense Atlantic hurricanes. Science, 2010, 327(5964): 454-458.

[33] Krausmann E, Cruz AM. Impact of the 11 March 2011, Great East Japan earthquake and tsunami on the chemical industry. Natural Hazards, 2013, 67(2): 811-828.

[34] Krausmann E, Mushtaq F. A qualitative Natech damage scale for the impact of floods on selected industrial facilities. Natural Hazards, 2008, 46(2): 179-197.

[35] Krausmann E, Salzano E. Krausmann E, Cruz AM, Salzano E. Lessons learned from Natech events. Natech risk assessment and management: Reducing the risk of natural-hazard impact on hazardous installations, 2017, Amsterdam: Elsevier 33-52.

[36] Krausmann E, Renni E, Campedel M, Cozzani V. Industrial accidents triggered by earthquakes, floods and lightning: Lessons learned from a database analysis. Natural Hazards, 2011, 59(1): 285-300.

[37] Kumasaki M, Hara T, Nakajima N, Wada Y, Makino R. The classification of physical effects from natural hazards for Natech risk assessment based on a Japanese database. Journal of Loss Prevention in the Process Industries, 2017, 50(B): 308-316.

[38] Landsea CW, Franklin JL. Atlantic hurricane database uncertainty and presentation of a new database format. Monthly Weather Review, 2013, 141(10): 3576-3592.

[39] Le, Q.V., and T. Mikolov. 2014. Distributed representations of sentences and documents. In Proceedings of the 31st International Conference on Machine Learning, 21–26 June 2014, Beijing, China. JMLR:W&CP 32: 1–5.

[40] LeCun Y, Bengio Y. Arbib MA. Convolutional networks for images, speech, and time series. The handbook of brain theory and neural networks, 1997, Cambridge, MA: The MIT Press 276-279.

[41] LeCun Y, Bengio Y, Hinton G. Deep learning. Nature, 2015, 521(7553): 436-444.

[42] Lu D, Weng Q. A survey of image classification methods and techniques for improving classification performance. International Journal of Remote Sensing, 2007, 28(5): 823-870.

[43] Manevitz LM, Yousef M. One-class SVMs for document classification. Journal of Machine Learning Research, 2001, 2: 139-154.

[44] Murphy KP. Machine learning: A probabilistic perspective, 2012, Cambridge, MA: The MIT press

[45] Özgür, A., L. Özgür, and T. Güngör. 2005. Text categorization with class-based and corpus-based keyword selection. In Proceedings of 20th International Symposium on Computer and Information Sciences, 26–28 October 2005, Istanbul, Turkey, 606–615.

[46] Ranjan R, Patel VM, Chellappa R. HyperFace: A deep multi-task learning framework for face detection, landmark localization, pose estimation, and gender recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2019, 41: 121-135.

[47] Rubin TN, Chambers A, Smyth P, Steyvers M. Statistical topic models for multi-label document classification. Machine Learning, 2012, 88(1–2): 157-208.

[48] Sboev A, Litvinova T, Gudovskikh D, Rybka R, Moloshnikov I. Machine learning models of text categorization by author gender using topic-independent features. Procedia Computer Science, 2016, 101: 135-142.

[49] Sengul H, Santella N, Steinberg LJ, Cruz AM. Analysis of hazardous material releases due to natural hazards in the United States. Disasters, 2012, 36(4): 723-743.

[50] Shah AA, Ye J, Abid M, Khan J, Amir SM. Flood hazards: Household vulnerability and resilience in disaster-prone districts of Khyber Pakhtunkhwa Province, Pakistan. Natural Hazards, 2018, 93(1): 147-165.

[51] Shin, J., Y. Kim, S. Yoon, and K. Jung. 2018a. Contextual-CNN: A novel architecture capturing unified meaning for sentence classification. In Proceedings of 2018 IEEE International Conference on Big Data and Smart Computing, 15–18 January 2018, Shanghai, China, 491–494.

[52] Shih, C.H., B.C. Yan, S.H. Liu, and B. Chen. 2018b. Investigating Siamese LSTM networks for text categorization. In Proceedings of the 9th Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, 12–15 November 2018, Honolulu, Hawaii, USA, 641–646.

[53] Showalter PS, Myers MF. Natural disasters as the cause of technological emergencies: A review of the decade, 1980–1989, 1992, Boulder, CO: Univercity of Colorada

[54] Sokolova M, Lapalme G. A systematic analysis of performance measures for classification tasks. Information Processing and Management, 2009, 45(4): 427-437.

[55] Sotthisopha, N., and P. Vateekul. 2018. Improving short text classification using fast semantic expansion on multichannel convolutional neural network. In Proceedings of the 19th IEEE/ACIS International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD), 27–29 June 2018, Busan, South Korea, 182–187.

[56] Sudharshan PJ, Petitjean C, Spanhol F, Oliveira LE, Heutte L, Honeine P. Multiple instance learning for histopathological breast cancer image classification. Expert Systems with Applications, 2019, 117: 103-111.

[57] TNO Industrial and External Safety. 2019. Failure and ACcidents Technical information System (FACTS). the Unified Industrial & Harbour Fire Department in Rotterdam-Rozenburg, the Netherlands. http://www.factsonline.nl/. Accessed 14 Jan 2019.

[58] Torn RD, Snyder C. Uncertainty of tropical cyclone best-track information. Weather and Forecasting, 2012, 27(3): 715-729.

[59] United States Coast Guard. 2017. United States National Response Center (NRC) database. Washington, DC: United States Coast Guard. http://www.nrc.uscg.mil/. Accessed 17 Oct 2017.