Research on emergency decision-making method for online public opinion on emergency events based on probabilistic linguistic term sets

doi:10.16265/j.cnki.issn1003-3033.2025.10.1691

Abstract

Abstract:

Aiming at the problems of uncertainty and imprecision of public opinion information in emergency response decision-making of emergency events network public opinion, a decision-making method for emergency response to online public opinion on emergency events based on PLTSs was proposed. Firstly, methods such as Python programming and machine learning were used to crawl, preprocess and analyze the sentiment tendency of online public opinion information to obtain the decision matrix characterized by PLTSs. Secondly, the CRITIC assignment method was used to objectively determine the weights of each attribute. Then, the comprehensive relevance of each case was obtained and ranked based on grey correlation analysis(GRA) model. Finally, the effectiveness and practicability of the proposed method were verified through typhoon disaster cases. The results of the study show that the proposed method is able to monitor the online public opinion of emergencies in real time, obtain decision-making data objectively and intelligently so as to realize the quantitative assessment of typhoon disasters. It provides good decision-making support for the relevant emergency departments to effectively respond to the online public opinion of emergencies.

Key words: probabilistic linguistic term sets(PLTSs), emergencies, online public opinion, emergency decision-making, criteria importance through intercriteria correlatio(CRITIC)

CLC Number:

X913.4安全系统工程

CHEN Chong, TAN Ruipu, ZHANG Wende. Research on emergency decision-making method for online public opinion on emergency events based on probabilistic linguistic term sets[J]. China Safety Science Journal, 2025, 35(10): 230-238.

Figures/Tables 10

Table 1

Fig.1

Fig.2

Table 2

Evaluation decision matrix characterized as PLTSs

方案	c₁	c₂	c₃	c₄
$Y 1$	${(s 0 (0.194), s 1 (0.182), s 2 (0.197), s 3 (0.103), s 4 (0.323)}$	${(s 0 (0.462), s 1 (0.151), s 2 (0.151), s 3 (0.087), s 4 (0.149)}$	${(s 0 (0.481), s 1 (0.133), s 2 (0.082), s 3 (0.068), s 4 (0.235)}$	${(s 0 (0.341), s 1 (0.259), s 2 (0.156), s 3 (0.115), s 4 (0.129)}$
$Y 2$	${(s 0 (0.192), s 1 (0.198), s 2 (0.209), s 3 (0.095), s 4 (0.305)}$	${(s 0 (0.418), s 1 (0.219), s 2 (0.166), s 3 (0.112), s 4 (0.086)}$	${(s 0 (0.586), s 1 (0.143), s 2 (0.090), s 3 (0.074), s 4 (0.107)}$	${(s 0 (0.362), s 1 (0.320), s 2 (0.196), s 3 (0.081), s 4 (0.040)}$
$Y 3$	${(s 0 (0.193), s 1 (0.162), s 2 (0.181), s 3 (0.113), s 4 (0.351)}$	${(s 0 (0.451), s 1 (0.189), s 2 (0.142), s 3 (0.085), s 4 (0.132)}$	${(s 0 (0.464), s 1 (0.140), s 2 (0.109), s 3 (0.091), s 4 (0.195)}$	${(s 0 (0.427), s 1 (0.226), s 2 (0.136), s 3 (0.111), s 4 (0.099)}$
$Y 4$	${(s 0 (0.179), s 1 (0.197), s 2 (0.202), s 3 (0.111), s 4 (0.310)}$	${(s 0 (0.421), s 1 (0.170), s 2 (0.170), s 3 (0.134), s 4 (0.105)}$	${(s 0 (0.649), s 1 (0.117), s 2 (0.074), s 3 (0.058), s 4 (0.102)}$	${(s 0 (0.333), s 1 (0.334), s 2 (0.189), s 3 (0.090), s 4 (0.054)}$
$Y 5$	${(s 0 (0.193), s 1 (0.186), s 2 (0.185), s 3 (0.103), s 4 (0.333)}$	${(s 0 (0.495), s 1 (0.195), s 2 (0.155), s 3 (0.090), s 4 (0.065)}$	${(s 0 (0.619), s 1 (0.120), s 2 (0.093), s 3 (0.072), s 4 (0.095)}$	${(s 0 (0.468), s 1 (0.222), s 2 (0.165), s 3 (0.091), s 4 (0.055)}$
$Y 6$	${(s 0 (0.215), s 1 (0.201), s 2 (0.219), s 3 (0.110), s 4 (0.255)}$	${(s 0 (0.425), s 1 (0.202), s 2 (0.171), s 3 (0.115), s 4 (0.087)}$	${(s 0 (0.523), s 1 (0.156), s 2 (0.112), s 3 (0.088), s 4 (0.121)}$	${(s 0 (0.434), s 1 (0.243), s 2 (0.188), s 3 (0.086), s 4 (0.048)}$
$Y 7$	${(s 0 (0.207), s 1 (0.164), s 2 (0.186), s 3 (0.124), s 4 (0.320)}$	${(s 0 (0.446), s 1 (0.169), s 2 (0.161), s 3 (0.120), s 4 (0.104)}$	${(s 0 (0.583), s 1 (0.128), s 2 (0.091), s 3 (0.073), s 4 (0.125)}$	${(s 0 (0.496), s 1 (0.223), s 2 (0.134), s 3 (0.078), s 4 (0.069)}$
$Y 8$	${(s 0 (0.187), s 1 (0.172), s 2 (0.189), s 3 (0.118), s 4 (0.334)}$	${(s 0 (0.521), s 1 (0.129), s 2 (0.156), s 3 (0.087), s 4 (0.106)}$	${(s 0 (0.506), s 1 (0.071), s 2 (0.146), s 3 (0.112), s 4 (0.165)}$	${(s 0 (0.329), s 1 (0.225), s 2 (0.174), s 3 (0.140), s 4 (0.132)}$
$Y 9$	${(s 0 (0.200), s 1 (0.175), s 2 (0.202), s 3 (0.103), s 4 (0.321)}$	${(s 0 (0.472), s 1 (0.179), s 2 (0.116), s 3 (0.106), s 4 (0.126)}$	${(s 0 (0.545), s 1 (0.129), s 2 (0.091), s 3 (0.069), s 4 (0.165)}$	${(s 0 (0.455), s 1 (0.211), s 2 (0.151), s 3 (0.078), s 4 (0.105)}$
$Y 10$	${(s 0 (0.195), s 1 (0.203), s 2 (0.225), s 3 (0.094), s 4 (0.283)}$	${(s 0 (0.399), s 1 (0.217), s 2 (0.178), s 3 (0.101), s 4 (0.105)}$	${(s 0 (0.643), s 1 (0.146), s 2 (0.086), s 3 (0.052), s 4 (0.073)}$	${(s 0 (0.311), s 1 (0.341), s 2 (0.207), s 3 (0.090), s 4 (0.050)}$
$Y 11$	${(s 0 (0.211), s 1 (0.185), s 2 (0.193), s 3 (0.099), s 4 (0.311)}$	${(s 0 (0.441), s 1 (0.196), s 2 (0.151), s 3 (0.078), s 4 (0.134)}$	${(s 0 (0.447), s 1 (0.153), s 2 (0.153), s 3 (0.071), s 4 (0.176)}$	${(s 0 (0.365), s 1 (0.250), s 2 (0.196), s 3 (0.095), s 4 (0.095)}$
$Y 12$	${(s 0 (0.213), s 1 (0.174), s 2 (0.186), s 3 (0.117), s 4 (0.309)}$	${(s 0 (0.430), s 1 (0.160), s 2 (0.210), s 3 (0.100), s 4 (0.100)}$	${(s 0 (0.504), s 1 (0.120), s 2 (0.064), s 3 (0.112), s 4 (0.200)}$	${(s 0 (0.353), s 1 (0.245), s 2 (0.216), s 3 (0.078), s 4 (0.108)}$
$Y 13$	${(s 0 (0.211), s 1 (0.172), s 2 (0.189), s 3 (0.100), s 4 (0.327)}$	${(s 0 (0.444), s 1 (0.186), s 2 (0.157), s 3 (0.095), s 4 (0.118)}$	${(s 0 (0.539), s 1 (0.099), s 2 (0.130), s 3 (0.042), s 4 (0.190)}$	${(s 0 (0.415), s 1 (0.207), s 2 (0.175), s 3 (0.087), s 4 (0.116)}$
$Y 14$	${(s 0 (0.178), s 1 (0.150), s 2 (0.430), s 3 (0.084), s 4 (0.159)}$	${(s 0 (0.357), s 1 (0.107), s 2 (0.321), s 3 (0.179), s 4 (0.036)}$	${(s 0 (0.429), s 1 (0.000), s 2 (0.429), s 3 (0.143), s 4 (0.000)}$	${(s 0 (0.346), s 1 (0.154), s 2 (0.192), s 3 (0.269), s 4 (0.038)}$
$Y 15$	${(s 0 (0.203), s 1 (0.201), s 2 (0.203), s 3 (0.103), s 4 (0.290)}$	${(s 0 (0.486), s 1 (0.201), s 2 (0.116), s 3 (0.086), s 4 (0.112)}$	${(s 0 (0.497), s 1 (0.116), s 2 (0.112), s 3 (0.076), s 4 (0.198)}$	${(s 0 (0.419), s 1 (0.204), s 2 (0.147), s 3 (0.114), s 4 (0.116)}$

Table 2

Fig.3

Table 3

Grey correlation coefficient matrix

方案	ξ⁺				ξ^-				$G i$
方案	$c 1$	$c 2$	$c 3$	$c 4$	$c 1$	$c 2$	$c 3$	$c 4$	$G i$
$Y 1$	0.892	0.930	0.712	0.890	0.478	0.634	0.416	0.680	0.656
$Y 2$	0.889	0.925	0.637	0.745	0.497	0.796	0.632	0.869	0.451
$Y 3$	0.875	0.912	0.744	0.852	0.447	0.679	0.461	0.794	0.623
$Y 4$	0.889	1.000	0.615	0.780	0.491	0.721	0.657	0.850	0.470
$Y 5$	0.876	0.839	0.625	0.769	0.469	0.921	0.667	1.000	0.398
$Y 6$	0.816	0.939	0.674	0.760	0.557	0.792	0.588	0.952	0.444
$Y 7$	0.872	0.963	0.656	0.755	0.484	0.745	0.592	0.964	0.458
$Y 8$	0.887	0.903	0.765	0.959	0.465	0.769	0.500	0.656	0.630
$Y 9$	0.897	0.899	0.692	0.788	0.479	0.695	0.514	0.791	0.521
$Y 10$	0.876	0.964	0.590	0.775	0.518	0.738	0.736	0.824	0.413
$Y 11$	0.872	0.914	0.770	0.831	0.495	0.671	0.485	0.788	0.590
$Y 12$	0.861	0.993	0.700	0.805	0.500	0.737	0.453	0.733	0.606
$Y 13$	0.880	0.939	0.731	0.817	0.475	0.717	0.474	0.740	0.560
$Y 14$	0.683	0.907	0.602	0.953	0.445	0.603	0.470	0.560	0.581
$Y 15$	0.859	0.859	0.741	0.869	0.519	0.748	0.461	0.733	0.551

Table 3

Table 4

Fig.4

Fig.5

Table 5

Comparative analysis results of different sorting methods

方案	传统 TOPSIS 方法 $P i$	排序	PLWA 算子方法 $Q i$	排序	文中方法 $G i$	排序
$Y 1$	0.609	1	0.314	2	0.607	1
$Y 2$	0.526	12	0.274	12	0.465	12
$Y 3$	0.587	4	0.304	6	0.582	3
$Y 4$	0.536	10	0.279	10	0.476	10
$Y 5$	0.499	15	0.260	15	0.429	15
$Y 6$	0.522	13	0.276	11	0.464	13
$Y 7$	0.532	11	0.274	13	0.471	11
$Y 8$	0.597	3	0.313	3	0.590	2
$Y 9$	0.567	9	0.285	9	0.514	9
$Y 10$	0.519	14	0.271	14	0.441	14
$Y 11$	0.582	6	0.304	5	0.563	5
$Y 12$	0.581	7	0.308	4	0.573	4
$Y 13$	0.584	5	0.298	7	0.542	7
$Y 14$	0.599	2	0.341	1	0.559	6
$Y 15$	0.579	8	0.291	8	0.537	8

Table 5

References 19

[1]	曾润喜. 网络舆情管控工作机制研究[J]. 图书情报工作, 2009, 53(18): 79-82.
	ZENG Runxi. Research on construction of network opinion control mechanism[J]. Library and Information Service, 2009, 53(18): 79-82.
[2]	中国互联网络信息中心. 第54次《中国互联网络发展状况统计报告》[EB/OL]. [2024-08-29]. https://www3.cnnic.cn/n4/2024/0829/c88-11065.html.
[3]	石川, 马铭君, 胡琳梅. 基于用户心理的网络舆情指标体系构建研究[J]. 北京邮电大学学报: 社会科学版, 2019, 21(5): 93-99.
	SHI Chuan, MA Mingjun, HU Linmei. Research on the construction of network public opinion indicator system based on user psychology[J]. Journal of Beijing University of Posts and Telecommunications:Social Sciences Edition, 2019, 21(5): 93-99.
[4]	LIU Zixuan, WU Xianwen. Structural analysis of the evolution mechanism of online public opinion and its development stages based on machine learning and social network analysis[J]. International Journal of Computational Intelligence Systems, 2023, 16(1): DOI:10.1007/s44196-023-00277-8.
[5]	CHEN Min, DU Wenhu. The predicting public sentiment evolution on public emergencies under deep learning and internet of things[J]. Journal of Supercomputing, 2023, 79(6): 6452-6470. doi: 10.1007/s11227-022-04900-x
[6]	陈鑫, 谢科范. 基于主动安全的重大事故网络舆情智能建模与仿真[J]. 中国安全科学学报, 2024, 34(1): 53-61. doi: 10.16265/j.cnki.issn1003-3033.2024.01.0741
	CHEN Xin, XIE Kefan. Intelligent modeling and simulation of online public opinion for major accidents based on proactive safety[J]. China Safety Science Journal, 2024, 34(1): 53-61. doi: 10.16265/j.cnki.issn1003-3033.2024.01.0741
[7]	刘婷, 鲁梓萌, 刘翕, 等. 基于灾害链的突发公共事件网络舆情风险评估研究[J]. 情报科学, 2025, 43(1): 69-78.
	LIU Ting, LU Zimeng, LIU Xi, et al. Network public opinion risk assessment based on disaster chain theory in public emergencies[J]. Information Science, 2025, 43(1): 69-78.
[8]	张桂蓉, 王雨晴. 数智赋能推进敏捷化应急情报体系研究[J]. 现代情报, 2024, 44(4): 3-10, 31.
	ZHANG Guirong, WANG Yuqing. Empowering digital intelligence to promote the agility of emergency intelligence system[J]. Journal of Modern Information, 2024, 44(4): 3-10, 31.
[9]	陈冲, 谭睿璞, 张文德, 等. 基于中智数的突发事件网络舆情辅助决策方法研究[J]. 中国安全生产科学技术, 2024, 20(5): 50-56.
	CHEN Chong, TAN Ruipu, ZHANG Wende, et al. Research on auxiliary decision-making method of online public opinion in emergencies based on neutrosophic number[J]. Journal of Safety Science and Technology, 2024, 20(5): 50-56.
[10]	DING Xuefeng, LIU Huchen, SHI Hua. A dynamic approach for emergency decision making based on prospect theory with interval-valued Pythagorean fuzzy linguistic variables[J]. Computers & Industrial Engineering, 2019, 131: 57-65. doi: 10.1016/j.cie.2019.03.037
[11]	RODRIGUEZ R M, MARTINEZ L, HERRERA F. Hesitant fuzzy linguistic term sets for decision making[J]. IEEE Transactions on Fuzzy Systems, 2012, 20(1): 109-119. doi: 10.1109/TFUZZ.2011.2170076
[12]	刘富鹏, 杨九, 吴世博, 等. 基于FDHHFLTS-BN的海底管道泄漏失效风险定量分析[J]. 中国安全科学学报, 2024, 34(1): 166-170. doi: 10.16265/j.cnki.issn1003-3033.2024.01.1247
	LIU Fupeng, YANG Jiu, WU Shibo, et al. Quantitative risk analysis on failure of submarine pipeline leakage based on FDHHFLTS-BN[J]. China Safety Science Journal, 2024, 34(1): 166-170. doi: 10.16265/j.cnki.issn1003-3033.2024.01.1247
[13]	PANG Qi, WANG Hai, XU Zeshui. Probabilistic linguistic term sets in multi-attribute group decision making[J]. Information Sciences, 2016,369:128-143.
[14]	HAN Xinru, ZHAN Jianming, XU Zeshui, et al. Trust risk test-based group consensus with probabilistic linguistic preference relations under social networks[J]. IEEE Transactions on FuzzySystems, 2024, 32(6): 3508-3520.
[15]	李宝德, 吕靖, 李晶. 基于前景理论和概率语言术语的突发事件应急响应决策方法[J]. 系统管理学报, 2023, 32(6): 1164-1175.
	LI Baode, LYU Jing, LI Jing. An emergency response decision-making method for emergencies based on prospect theory and probalilistic linguistic terms[J]. Journal of Systems & Management, 2023, 32(6): 1164-1175.
[16]	XIAO Huimin, GAO Xiaosong, YANG Peng, et al. A multi-attribute decision making method for TODIM based on mixed correlation coefficients for probabilistic linguistic term sets[J]. Journal of Intelligent & Fuzzy Systems, 2023, 45(4):6591-6604.
[17]	王增强, 蒲云. 基于多粒度概率语言和双参照点的应急决策方法[J/OL]. 中国管理科学: 1-12. [2024-02-19]. https://doi.org/10.16381/j.cnki.issn1003-207x.2023.0990.
	WANG Zengqiang, PU Yun. An emergency decision-making method based on multi-granularity probabilistic linguistic terms and double reference points[J/OL]. Chinese Journal of Management Science: 1-12. [2024-02-19]. https://doi.org/10.16381/j.cnki.issn1003-207x.2023.0990.
[18]	DIAKOULAKI D, MAVROTAS G, PAPAYANNAKIS L. Determining objective weights in multiple criteria problems: the CRITIC method[J]. Computers & Operations Research, 1995, 22(7): 763-770. doi: 10.1016/0305-0548(94)00059-H
[19]	谭学瑞, 邓聚龙. 灰色关联分析: 多因素统计分析新方法[J]. 统计研究, 1995(3): 46-48.

属性	指标差异性	指标冲突性	信息量	属性权重/%
C₁	0.012	3.514	0.042	22.762 7
C₂	0.012	3.074	0.036	19.483 8
C₃	0.028	2.218	0.062	34.008 2
C₄	0.020	2.206	0.044	23.745 2