ISSN 0253-2778

CN 34-1054/N

open
Free to Publish. Free to Read.
Open AccessOpen Access JUSTC Engineering& Materials/Info.& Intelligence 17 May 2022

Application of a newly developed naive Bayes algorithm in fire alarm

  • State Key Laboratory of Fire Science, University of Science and Technology of China, Hefei 230027, China

Cite this:
https://doi.org/10.52396/JUSTC-2021-0258
More Information
  • Author Bio:

    Xiangyong He is a postgraduate student under the supervision of Prof. Yong Jiang at the University of Science and Technology of China. His research mainly focuses on the application of machine learning in fire science

    Yong Jiang is a doctoral supervisor and professor at the University of Science and Technology of China, and is currently the director of Computer Simulation Research Office, State Key Laboratory of Fire Science. His research interests mainly include precision diagnostic experimental technology of fire and combustion, computer simulation and emulation of fire and combustion, measurement and model of combustion reaction kinetics, thermal safety and artificial intelligence in energy utilization

  • Corresponding author: E-mail: yjjiang@ustc.edu.cn
  • Received Date: 03 December 2021
  • Accepted Date: 20 March 2022
    • Available Online: 17 May 2022
    Abstract

  • Abstract

    To address the problems of low recognition accuracy of traditional early fire warning systems in actual scenarios, a newly developed naive Bayes (NB) algorithm, namely, improved naive Bayes (INB), was proposed. An optimization method based on attribute weighting and an orthogonal matrix was used to improve the NB algorithm. Attribute weighting considers the influence of different values of each attribute on classification performance under every decision category; the orthogonal matrix weakens the linear relationship between the attributes reducing their correlations, which is more closely related to the conditional independence assumption. Data from the technology report of the National Institute of Standards and Technology (NIST) regarding fire research were used for the simulation, and eight datasets of different sizes were constructed for INB training and testing after filtering and normalization. A ten-fold cross-validation suggests that INB has been effectively trained and demonstrates the stable ability in fire alarms when the dataset contains 190 sets of samples; namely, the INB can be fully trained by using small datasets. A support vector machine (SVM), a back propagation (BP) neural network, and NB were selected for comparison. The results showed that the recognition accuracy, average precision, average recall, and average $\rm{F}_{1}$ measure of INB were 96.1%, 97.3%, 97.2%, and 97.3%, respectively, which is the highest among the four different algorithms. Additionally, INB has a better performance compared to NB, SVM, and BP neural networks when the training time is short . In conclusion, INB can be used as a core algorithm for fire alarm systems with excellent and stable fire alarm capabilities.

    Graphical abstract

    With reasonable improvements, the naive Bayes algorithm can be the core processing algorithm of a fire alarm system.

    Abstract

    To address the problems of low recognition accuracy of traditional early fire warning systems in actual scenarios, a newly developed naive Bayes (NB) algorithm, namely, improved naive Bayes (INB), was proposed. An optimization method based on attribute weighting and an orthogonal matrix was used to improve the NB algorithm. Attribute weighting considers the influence of different values of each attribute on classification performance under every decision category; the orthogonal matrix weakens the linear relationship between the attributes reducing their correlations, which is more closely related to the conditional independence assumption. Data from the technology report of the National Institute of Standards and Technology (NIST) regarding fire research were used for the simulation, and eight datasets of different sizes were constructed for INB training and testing after filtering and normalization. A ten-fold cross-validation suggests that INB has been effectively trained and demonstrates the stable ability in fire alarms when the dataset contains 190 sets of samples; namely, the INB can be fully trained by using small datasets. A support vector machine (SVM), a back propagation (BP) neural network, and NB were selected for comparison. The results showed that the recognition accuracy, average precision, average recall, and average $\rm{F}_{1}$ measure of INB were 96.1%, 97.3%, 97.2%, and 97.3%, respectively, which is the highest among the four different algorithms. Additionally, INB has a better performance compared to NB, SVM, and BP neural networks when the training time is short . In conclusion, INB can be used as a core algorithm for fire alarm systems with excellent and stable fire alarm capabilities.

    • Public Summary

    • A fire alarm algorithm based on naive Bayes was proposed.
    • Attribute weighting and orthogonal matrix methods were introduced to improve naive Bayes.
    • The improved naive Bayes algorithm has better performance and does not rely on a large amount of training data.

    FullText(HTML)
  • loading
    • References(22)
  • [1]
    Baek J, Alhindi T J, Jeong M K, et al. Real-time fire detection algorithm based on support vector machine with dynamic time warping kernel function. Fire Technology, 2021, 57 (6): 2929–2953. doi: 10.1007/s10694-020-01062-1
    [2]
    Jafari M J, Pouyakian M, Khanteymoori A, et al. Reliability evaluation of fire alarm systems using dynamic Bayesian networks and fuzzy fault tree analysis. Journal of Loss Prevention in the Process Industries, 2020, 67: 104229. doi: 10.1016/j.jlp.2020.104229
    [3]
    Shokouhi M, Nasiriani K, Cheraghi Z, et al. Preventive measures for fire-related injuries and their risk factors in residential buildings: A systematic review. Journal of Injury & Violence Research, 2019, 11 (1): 1–14. doi: https://doi.org/10.5249/jivr.v11i1.1057
    [4]
    Chow W K, Wan E T K, Cheung K P. Possibility of using laser-fibre optics as a fire detection system. Optics and Lasers in Engineering, 1997, 27 (2): 201–210. doi: 10.1016/S0143-8166(96)00002-4
    [5]
    Wu L S, Chen L, Hao X R. Multi-sensor data fusion algorithm for indoor fire early warning based on bp neural network. Information, 2021, 12 (2): 59. doi: 10.3390/info12020059
    [6]
    Sarwar B, Bajwa I S, Jamil J, et al. An intelligent fire warning application using IoT and an adaptive neuro-fuzzy inference system. Sensors, 2019, 19 (14): 3150. doi: 10.3390/s19143150
    [7]
    Saeed F, Paul A, Karthigaikumar P, et al. Convolutional neural network based early fire detection. Multimedia Tools and Applications, 2020, 79 (13-14): 9083–9099. doi: 10.1007/s11042-019-07785-w
    [8]
    Liang S, Zhang H G, You Y M, et al. Towards fire prediction accuracy enhancements by leveraging an improved naïve bayes algorithm. Symmetry, 2021, 13 (4): 530. doi: 10.3390/sym13040530
    [9]
    Kuo H C, Chang H K. A real-time shipboard fire-detection system based on grey-fuzzy algorithms. Fire Safety Journal, 2003, 38 (4): 341–363. doi: 10.1016/S0379-7112(02)00088-7
    [10]
    Wei L M, Dong T H, Zhang Y X, et al. Research on fire alarm system based on bayesian algorithm. Fire Science and Technology, 2021, 40 (8): 1199–1205. doi: 10.3969/j.issn.1009-0029.2021.08.021
    [11]
    Sulistian G, Abdurohman M, Putrada A G, et al. Comparison of classification algorithms to improve smart fire alarm system performance. In: 2019 International Workshop on Big Data and Information Security (IWBIS). IEEE, 2019: 119–124.
    [12]
    Bahrepour M, Meratnia N, Havinga P. Use of AI techniques for residential fire detection in wireless sensor networks. In: Proceedings of the Workshops of the 5th IFIP Conference on Artificial Intelligence Applications & Innovations (AIAI-2009), Thessaloniki, Greece, 2009: 311–321.
    [13]
    Ebadati O M E, Ahmadzadeh F. Classification spam email with elimination of unsuitable features with hybrid of GA-naive Bayes. Journal of Information & Knowledge Management, 2019, 18 (1): 1950008. doi: https://doi.org/10.1142/S0219649219500084
    [14]
    Li Z, Li R, Jin G H. Sentiment analysis of danmaku videos based on Naive Bayes and sentiment dictionary. IEEE Access, 2020, 8: 75073–75084. doi: 10.1109/ACCESS.2020.2986582
    [15]
    Zaidi N A, Cerquides J, Carman M J, et al. Alleviating naive Bayes attribute independence assumption by attribute weighting. Journal of Machine Learning Research, 2013, 24: 1947–1988.
    [16]
    Jiang L X, Zhang L G, Yu L J, et al. Class-specific attribute weighted naive Bayes. Pattern Recognition, 2019, 88: 321–330. doi: 10.1016/j.patcog.2018.11.032
    [17]
    Li F X, Wang J M, Liang J C, et al. Optimization of naive Bayesian classification algorithm for discrete attributes. Journal of Chinese Computer Systems, 2022, 43 (5): 897–901. doi: doi:10.20009/j.cnki.21-1106/TP.2020-1041
    [18]
    Bukowski R W, Peacock R D, Averill J D, et al. Performance of home smoke alarms analysis of the response of several available technologies in residential fire settings. Gaithersburg, MD: National Institute of Standards and Technology, 2003.
    [19]
    Alessandri A, Bagnerini P, Gaggero M, et al. Parameter estimation of fire propagation models using level set methods. Applied Mathematical Modelling, 2021, 92: 731–747. doi: 10.1016/j.apm.2020.11.030
    [20]
    Zhang J J, Ye Z Y, Li K F. Multi-sensor information fusion detection system for fire robot through back propagation neural network. PLoS ONE, 2020, 15 (7): e0236482. doi: 10.1371/journal.pone.0236482
    [21]
    Yang X, Zhang K, Chai Y, et al. A multi-sensor characteristic parameter fusion analysis based electrical fire detection model. In: Proceedings of 2018 Chinese Intelligent Systems Conference. Singapore: Springer, 2018: 397–410.
    [22]
    Savitzky A, Golay M J. Smoothing and differentiation of data by simplified least squares procedures. Analytical Chemistry, 1964, 36 (8): 1627–1639. doi: 10.1021/ac60214a047
    • Supplements(0)
    • Related Articles
    • Track Citations
  • 加载中

Catalog

    |
    Get Citation
    PDF
    XML

    Figure  1.  Comparison of the data before and after filtering.

    Figure  2.  Classification accuracy of four algorithms under different size of dataset.

    Figure  3.  Comparing classification accuracy of four algorithms.

    Figure  4.  Comparing precision of four algorithms.

    Figure  5.  Comparing recall of four algorithms.

    [1]
    Baek J, Alhindi T J, Jeong M K, et al. Real-time fire detection algorithm based on support vector machine with dynamic time warping kernel function. Fire Technology, 2021, 57 (6): 2929–2953. doi: 10.1007/s10694-020-01062-1
    [2]
    Jafari M J, Pouyakian M, Khanteymoori A, et al. Reliability evaluation of fire alarm systems using dynamic Bayesian networks and fuzzy fault tree analysis. Journal of Loss Prevention in the Process Industries, 2020, 67: 104229. doi: 10.1016/j.jlp.2020.104229
    [3]
    Shokouhi M, Nasiriani K, Cheraghi Z, et al. Preventive measures for fire-related injuries and their risk factors in residential buildings: A systematic review. Journal of Injury & Violence Research, 2019, 11 (1): 1–14. doi: https://doi.org/10.5249/jivr.v11i1.1057
    [4]
    Chow W K, Wan E T K, Cheung K P. Possibility of using laser-fibre optics as a fire detection system. Optics and Lasers in Engineering, 1997, 27 (2): 201–210. doi: 10.1016/S0143-8166(96)00002-4
    [5]
    Wu L S, Chen L, Hao X R. Multi-sensor data fusion algorithm for indoor fire early warning based on bp neural network. Information, 2021, 12 (2): 59. doi: 10.3390/info12020059
    [6]
    Sarwar B, Bajwa I S, Jamil J, et al. An intelligent fire warning application using IoT and an adaptive neuro-fuzzy inference system. Sensors, 2019, 19 (14): 3150. doi: 10.3390/s19143150
    [7]
    Saeed F, Paul A, Karthigaikumar P, et al. Convolutional neural network based early fire detection. Multimedia Tools and Applications, 2020, 79 (13-14): 9083–9099. doi: 10.1007/s11042-019-07785-w
    [8]
    Liang S, Zhang H G, You Y M, et al. Towards fire prediction accuracy enhancements by leveraging an improved naïve bayes algorithm. Symmetry, 2021, 13 (4): 530. doi: 10.3390/sym13040530
    [9]
    Kuo H C, Chang H K. A real-time shipboard fire-detection system based on grey-fuzzy algorithms. Fire Safety Journal, 2003, 38 (4): 341–363. doi: 10.1016/S0379-7112(02)00088-7
    [10]
    Wei L M, Dong T H, Zhang Y X, et al. Research on fire alarm system based on bayesian algorithm. Fire Science and Technology, 2021, 40 (8): 1199–1205. doi: 10.3969/j.issn.1009-0029.2021.08.021
    [11]
    Sulistian G, Abdurohman M, Putrada A G, et al. Comparison of classification algorithms to improve smart fire alarm system performance. In: 2019 International Workshop on Big Data and Information Security (IWBIS). IEEE, 2019: 119–124.
    [12]
    Bahrepour M, Meratnia N, Havinga P. Use of AI techniques for residential fire detection in wireless sensor networks. In: Proceedings of the Workshops of the 5th IFIP Conference on Artificial Intelligence Applications & Innovations (AIAI-2009), Thessaloniki, Greece, 2009: 311–321.
    [13]
    Ebadati O M E, Ahmadzadeh F. Classification spam email with elimination of unsuitable features with hybrid of GA-naive Bayes. Journal of Information & Knowledge Management, 2019, 18 (1): 1950008. doi: https://doi.org/10.1142/S0219649219500084
    [14]
    Li Z, Li R, Jin G H. Sentiment analysis of danmaku videos based on Naive Bayes and sentiment dictionary. IEEE Access, 2020, 8: 75073–75084. doi: 10.1109/ACCESS.2020.2986582
    [15]
    Zaidi N A, Cerquides J, Carman M J, et al. Alleviating naive Bayes attribute independence assumption by attribute weighting. Journal of Machine Learning Research, 2013, 24: 1947–1988.
    [16]
    Jiang L X, Zhang L G, Yu L J, et al. Class-specific attribute weighted naive Bayes. Pattern Recognition, 2019, 88: 321–330. doi: 10.1016/j.patcog.2018.11.032
    [17]
    Li F X, Wang J M, Liang J C, et al. Optimization of naive Bayesian classification algorithm for discrete attributes. Journal of Chinese Computer Systems, 2022, 43 (5): 897–901. doi: doi:10.20009/j.cnki.21-1106/TP.2020-1041
    [18]
    Bukowski R W, Peacock R D, Averill J D, et al. Performance of home smoke alarms analysis of the response of several available technologies in residential fire settings. Gaithersburg, MD: National Institute of Standards and Technology, 2003.
    [19]
    Alessandri A, Bagnerini P, Gaggero M, et al. Parameter estimation of fire propagation models using level set methods. Applied Mathematical Modelling, 2021, 92: 731–747. doi: 10.1016/j.apm.2020.11.030
    [20]
    Zhang J J, Ye Z Y, Li K F. Multi-sensor information fusion detection system for fire robot through back propagation neural network. PLoS ONE, 2020, 15 (7): e0236482. doi: 10.1371/journal.pone.0236482
    [21]
    Yang X, Zhang K, Chai Y, et al. A multi-sensor characteristic parameter fusion analysis based electrical fire detection model. In: Proceedings of 2018 Chinese Intelligent Systems Conference. Singapore: Springer, 2018: 397–410.
    [22]
    Savitzky A, Golay M J. Smoothing and differentiation of data by simplified least squares procedures. Analytical Chemistry, 1964, 36 (8): 1627–1639. doi: 10.1021/ac60214a047
    Recommended articles
    TrendMD
    Volume 52 Issue 6 page: 5
    Cover

    Article Metrics

    Article views (495) PDF downloads(2289)
    Proportional views

    Copyright © Editorial Office of JUSTC, All Rights Reserved. 皖ICP备05002528号

    Supported by: Beijing Renhe Information Technology Co. Ltd    

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return