Education, Science, Technology, Innovation and Life
Open Access
Sign In

An Early Warning Model for Social Stability Based on the PSO-C4.5 Decision Tree Model

Download as PDF

DOI: 10.23977/acss.2023.070416 | Downloads: 22 | Views: 388

Author(s)

Hengfeng Shen 1, Henghua Shen 2, Ning Wang 1

Affiliation(s)

1 Guangzhou Huashang College, Guangzhou, 522000, China
2 Capital University of Economics and Business, Jieyang, 522000, China

Corresponding Author

Ning Wang

ABSTRACT

Since ancient times, social stability has always been an issue that people attach importance to, and changes in political, economic, cultural and other factors have brought new challenges and made the country fragile for social stability, so the study of maintaining social stability has become particularly important. This paper establishes an early warning model of social stability, first based on the topsis entropy weight method to calculate the comprehensive score of social stability in 215 countries, and at the same time encode the data based on the score to form three social stability grades of strong, medium and weak, and then establish the C4.5 decision tree model, after hyperparameter optimization through PSO, the optimal parameter social stability early warning model is established. The advantage of the decision tree model is that it plots the decision-making process for each metric. The decision tree chart can clearly display the threshold of each indicator when dividing the social stability level, which is conducive to early warning of the indicators of the social stability index system.

KEYWORDS

Social stability, Early warning model, Topsis, Decision tree

CITE THIS PAPER

Hengfeng Shen, Henghua Shen, Ning Wang. An Early Warning Model for Social Stability Based on the PSO-C4.5 Decision Tree Model. Advances in Computer, Signals and Systems (2023) Vol. 7: 110-118. DOI: http://dx.doi.org/10.23977/acss.2023.070416.

REFERENCES

[1] Wang Yuanyuan, Bai Hongkun, Wang Shiqian, Bu Feifei, Wu Xiong, Li Haoyu. Power User Behavior Portrait Based on Information Gain and Spearman Correlation Coefficient. Electric Power Engineering Technology, 2022, 41 (04): 220-228. (in Chinese)
[2] Chen H. Analysis of water quality change characteristics in Shanghai based on Spearman rank correlation coefficient method. Environmental science and technology, 2020 (3): 28 and 33, DOI: 10.19824 / j.carol carroll nki cn32-1786 / x. 2020.0037.
[3] Li Ying, Ge Xizhen. Construction of cucumber downy mildew early warning system in greenhouse based on PCA and multiple regression algorithm. Journal of Anhui Agricultural Sciences, 2022, 50 (21): 232-234.
[4] Mo Yun, Guo Yan, Mo Hesheng, Lu Zhongwei, Zhang Shaorong. Hybrid feature selection method based on feature filtering and PCA dimension reduction. Journal of Guilin University of Astronautics Technology, 2022, 27 (02): 145-151. (in Chinese)
[5] Hou Meien, Wang Xiangbin. Extraction of biomechanical characteristics of gait in KOA patients by PCA. Medical Biomechanics, 2021, 36 (S1): 224.
[6] Wang Xueli. A study on the Effect of Online celebrities' Live streaming on Consumers' Purchase Intention. Gansu Science and Technology, 2020, 49 (11): 76-80.
[7] Cong Xiaoqi. Research on Entrepreneurial self-efficacy of accounting based on hierarchical regression. Journal of Science and Technology Entrepreneurship, 2020, 33 (08): 129-131.
[8] Liu Xiaojun. Study on Comprehensive Benefit Evaluation of Green construction Land saving Technology based on topsis entropy weight method. Mathematics practice and understanding: 1-10 [2023-02-05].
[9] Wang Zitao, Wang Jianping, Han Guang. Evaluation of groundwater richness based on entropy weight TOPSIS method in Qaidam Basin. Salt Lake Research, 2022, 30 (02): 42-51.
[10] Hu Mingqi, Zhang Senchang. Leaf node weighted random forest algorithm based on PSO optimization. Modern Computer, 2022, 28 (04): 1-4.

Downloads: 15496
Visits: 269508

Sponsors, Associates, and Links


All published work is licensed under a Creative Commons Attribution 4.0 International License.

Copyright © 2016 - 2031 Clausius Scientific Press Inc. All Rights Reserved.