Title: Research into a risk assessment model for online public opinions based on big data: random forest and logistic model

Authors: Danhui Dong

Addresses: School of Culture and Tourism, Wuxi Vocational College of Science and Technology, No. 8 Xinxi Road, Wuxi City, Jiangsu Province, 214000, China

Abstract: The current risk assessment index system for online public opinions has some deficiencies; therefore, the risk assessment method for online public opinions has some disadvantages. In order to overcome these disadvantages, this research attempts to propose a risk assessment model for online public opinions based on a random forest and logistic model, and then the risks of online public opinions can be evaluated effectively. With the incident of "patient relatives purposefully hurting doctors at Beijing Civil Aviation General Hospital" as the research object, a systematic analysis was conducted in this research on the model indexes. The critical risk indexes of online public opinions are confirmed; sensitivity, reporting speed, reporting frequency, emotional tendency, satisfaction, and timeliness of the processing process are the main factors affecting the risk of online public opinions. The results can provide practical reference for the development trend and risk assessment of online public opinions.

Keywords: online public opinion; risk assessment; random forest; logistic model; big data.

DOI: 10.1504/IJDS.2024.135966

International Journal of Data Science, 2024 Vol.9 No.1, pp.19 - 34

Received: 24 Mar 2023
Accepted: 29 Nov 2023

Published online: 10 Jan 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article