Abstract
A cross-region transfer learning method is proposed to identify community (e.g. car parks, public spaces and shopping centers) fire hazards based on text input provided by community members. The key component of the method, which also accounts for data imbalance, is an improved transfer component analysis that is embedded with a local discriminant analysis to transfer non-local rich knowledge to the fire hazard identification of local communities with an insufficient number of samples. In addition, a fire hazard knowledge map is established and applied to supplement the missing key features for fire hazard identification, and ontology modeling is applied to standardize the text features and reduce the effect of semantic ambiguity brought by cross-region knowledge transfer. The proposed method is verified based on the text data of nine fire hazard classes from Lanzhou and Beidaihe in China. Machine learning experiments show that fire hazard identification performance of all nine classes were improved with the overall accuracy, precision, recall, F1 score and AUC increased by 12%, 15%, 16%, 15% and 15%, respectively. Under data imbalance scenarios, the proposed method outperforms the state of the art methods, such as sampling-based methods, FastText and ULMFiT. The results also show that the proposed method can achieve desired performance with only half of the training samples. These findings illustrate that the proposed method can assist regions by improving fire identification results significantly through knowledge transfer. The proposed approach can be followed to build smart systems for community fire risk management with reasonable performance and high efficiency.
Similar content being viewed by others
Data Availability
The data in this paper is confidential and can only be shared with permission of the relevant governments.
References
CTIF World Fire Statistics Center (2020) World fire statistics. https://www.ctif.org/world-fire-statistics. Accessed 23 Jun 2020
Home Office (2019) Fire and rescue incident statistics: England, year ending June 2019. https://www.gov.uk/government/statistics/fire-and-rescue-incident-statistics-england-year-ending-june-2019.Accessed. Accessed 17 Jan 2020
U.S. Fire Administration (2016) U.S. fire statistics. https://www.usfa.fema.gov/data/statistics. Accessed 17 Jan 2020
Ministry of Emergency Management of the People’s Republic of China (2019) Fire statistics, China. http://www.119.gov.cn/xiaofang/hztj/36306.htm. Accessed 17 Jan 2020
Yung DT, Benichou N (2002) How design fires can be used in fire hazard analysis. Fire Technol 38(3):231–242
Gehandler J, Eymann L, Regeffe M (2015) Limit-based fire hazard model for evaluating tunnel life safety. Fire Technol 51(3):585–614
Aziz A, Ahmed S, Khan FI (2019) An ontology-based methodology for hazard identification and causation analysis. Process Saf Environ 123:87–98
Xin J, Huang C (2013) Fire risk analysis of residential buildings based on scenario clusters and its application in fire risk management. Fire Saf J 62:72–78
Xu G, Zhang YM, Zhang QX, Lin GH, Wang Z, Jia Y, Wang JJ (2019) Video smoke detection based on deep saliency network. Fire Saf J 105:277–285
Xin PW, Khan F, Ahmed S (2017) Dynamic hazard identification and scenario mapping using Bayesian network. Process Saf Environ 105:143–155
Crawley F, Tyler B (2015) HAZOP: guide to best practice. Elsevier, Amsterdam
Horváth I, van Beeck J, Merci B (2013) Full-scale and reduced-scale tests on smoke movement in case of car park fire. Fire Saf J 57:35–43
Moinuddin KAM, Innocent J, Keshavarz K (2019) Reliability of sprinkler system in Australian shopping centres—a fault tree analysis. Fire Saf J 105:204––215
Bai XY, Hanif MI, Li FS, Hanif MS, Gu YH (2017) An empirical study on application and efficiency of gridded management in public service supply of Chinese Government. J Sci Technol Policy Manag 8(1):2–15
Wang YF, Zhou HR, Jing ZL, Xiang LH, Cai WJ (2006) The exploration of urban gridded management in e-government. In: Proceeding of the fifth international conference on grid and cooperative computing workshops, Hunan, China
Soyata T, Habibzadeh H, Ekenna C, Nussbaum B, Lozano J (2019) Smart city in crisis: technology and policy concerns. Sustain Cities Soc 50:101566
Chen DT, Bourlard H, Thiran JP (2001) Text identification in complex background using SVM. In: Proceeding of the 2001 IEEE computer society conference on computer vision and pattern recognition, Kauai, HI, USA.
Koziarski M (2020) Radial-based undersampling for imbalanced data classification. Pattern Recognit 102:1–11
Grubinger T, Birlutiu A, Schöner H, Heskes T (2017) Multi-domain transfer component analysis for domain generalization. Neural Process Lett 46(3):845–855
Devlin J, Chang MW, Lee K, Toutanova K (2015) BERT: pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805v1
Howard J, Ruder S (2018) Universal language model fine-tuning for text classification. In: Proceedings of the 56th annual meeting of the association for computational linguistics, Melbourne, Australia
Zhang DW, Xu H, Su ZC, Xu YF (2015) Chinese comments sentiment classification based on word2vec and SVMperf. Expert Syst Appl 42(4):1857–1863
Quanz B, Huan J (2009) Large margin transductive transfer learning. In: Proceedings of the 18th ACM conference on information and knowledge management, Hong Kong, China
Xiao M, Guo Y (2012) Semi-supervised kernel matching for domain adaptation. In: Proceedings of the 26th AAAI conference on artificial intelligence, Toronto, Canada
Dai WY, Yang Q, Xue GR, Yu Y (2018) Boosting for transfer learning. In: Proceeding of the 24th international conference on machine learning, Corvallis, Oregon, USA
Pan SJ, Kwok JT, Yang Q (2008) Transfer learning via dimensionality reduction. In: Proceedings of the 23rd national conference on artificial intelligence, Chicago, USA
Pan SJ, Tsang IW, Kwok JT, Yang Q (2011) Domain adaptation via transfer component analysis. IEEE Trans Neural Netw 22(2):199–210
Pareek NK, Patidar V, Sud KK (2006) Image encryption using chaotic logistic map. Image Vis Comput 24(9):926–934
Li M, Lu XZ, Chen LS, Wang J (2020) Knowledge map construction for question and answer archives. Expert Syst Appl 141:112923
Wang DS, Tiwari P, Garg S, Zhu HY, Bruza P (2019) Structural block driven enhanced convolutional neural representation for relation extraction. Appl Soft Comput 105913
Balubaid MA, Manzoor U (2015) Ontology based SMS controller for smart phones. Int J Adv Comput Sci Appl 6(1):133–139
Ma J, Xu W, Sun YH, Turban E (2012) An ontology-based text-mining method to cluster proposals for research project selection. IEEE Trans Syst Man Cybern Syst 42(3):784–790
Xinhua News Agency (2014) Interpretation of decision of the Third Plenary Session of the eighteenth central committee: Improving the way of social governance. http://www.gov.cn/jrzg/2014-02/17/content_2606543.html. Accessed 17 Apr 2019
Hodges JH, Lattimer BY (2019) Wildland fire spread modeling using convolutional neural networks. Fire Technol 55(6):2115–2142
Mikolov T, Sutskever I, Chen K, Corrado CS (2013) Distributed representations of words and phrases and their compositionality. Adv Neural Inf Process Syst 26:1–9
Cong YN, Chan YB, Phillips CA, Langston MA (2017) Robust inference of genetic exchange communities from microbial genomes using TF-IDF. Front Microbiol 8(15):1–11
Tong H, Ng M (2018) Analysis of regularized least squares for functional linear regression model. J Complex 49:85–94
Joulin A, Grave E, Bojanowski P, Mikolov T (2017) Bag of tricks for efficient text classification. In: Proceedings of the 15th conference of the european chapter of the association for computational linguistics, Valencia, Spain
Liang XW, Jiang AP, Li T, Xue YY, Wang GT (2020) LR-SMOTE—an improved unbalanced data set oversampling based on K-means and SVM. Knowl Based Syst 196:1–10
El Hindi K, AlSalman H, Qasem S, Al Ahmadi S (2018) Building an ensemble of fine-tuned naive Bayesian classifiers for text classification. Entropy 20(11):1–13
Kamkarhaghighi M, Makrehchi M (2017) Content tree word embedding for document representation. Expert Syst Appl 90:241–249
Meng JN, Lin HF, Li YP (2011) Knowledge transfer based on feature representation mapping for text classification. Expert Syst Appl 38(8):10562–10567
Xu Z, Mei L, Lv ZH, Hu CP, Luo XF, Zhang H, Liu YH (2019) Multi-Modal description of public safety events Using surveillance and social media. IEEE Trans Big Data 5(4):529–539
Yu CY, Fang J, Wang JJ, Zhang YM (2010) Erratum to: video fire smoke detection using motion and color features. Fire Technol 46(3):763
Xiang EW, Cao B, Hu DH, Yang Q (2010) Bridging domains using world wide knowledge for transfer learning. IEEE Trans Knowl Data Eng 22(6):770–783
Acknowledgements
This work is supported by the Major Research Project of Nation Natural Science Foundation of China named “Big data Driven Management and Decision-making Research” (No. 91746207), the General Program of Nation Natural Science Foundation of China (No. 71774043) and the Emergency Management Major Research Project of Nation Natural Science Foundation of China (No. 91024028).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflicts of interest
None.
Code Availability
The code is open to the public. Anyone who want the code can contact the authors.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Liu, ZG., Li, XY. & Jomaas, G. Identifying Community Fire Hazards from Citizen Communication by Applying Transfer Learning and Machine Learning Techniques. Fire Technol 57, 2809–2838 (2021). https://doi.org/10.1007/s10694-020-01035-4
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10694-020-01035-4