ArticleOpen Access http://dx.doi.org/10.26855/ea.2025.10.014
Research on the Evaluation of User Safety Intervention Measures Based on Causal Inference
Huisheng Liu
Operations Research, Columbia University, New York, NY 10027, USA.
*Corresponding author: Huisheng Liu
Published: December 24,2025
Abstract
With the growing accumulation of user behavioral data on digital platforms, safety intervention has become a crucial mechanism for maintaining ecosystem stability. Traditional evaluation methods often rely on empirical comparison, making it difficult to uncover the true causal relationship between interventions and safety outcomes. The adoption of causal inference provides a rigorous analytical framework for identifying intervention effectiveness. By applying models such as propensity score matching, difference-in-differences, and instrumental variables, confounding factors can be controlled, and the genuine effect mechanism revealed. Causal analysis based on behavioral data enables the calibration of intervention strategies, optimization of user segmentation, timing, and intensity, thus improving both precision and long-term efficiency in safety governance. This approach reshapes the evaluation paradigm through data-driven causal reasoning and offers methodological guidance for enhancing user protection and platform trust mechanisms.
Keywords
Causal inference; user safety; intervention measures; effectiveness evaluation; data governance
References
[1] Chen W, Hu Y, Sui R, et al. Competitive e-commerce platforms’ data provision and pricing strategies with different attribution behaviors of users. Expert Syst Appl. 298:129466.
[2] Campellone RT, Flom M, Montgomery MR, et al. Safety and user experience of a generative artificial intelligence digital mental health intervention: exploratory randomized controlled trial. J Med Internet Res. 2025;27:e67365.
[3] Cheng L, Guo Y. Privacy protection on social media platforms: overdisclosure of online behavioral data is labeling users. Int J Digit Law Gov. 2025;2(1):107-33.
[4] Taher R, Stahl D, Shergill S, et al. The safety of digital mental health interventions: findings and recommendations from a qualitative study exploring users’ experiences, concerns, and suggestions. JMIR Hum Factors. 2025;12:e62974.
[5] Yang L. Deep learning-based user behavior data mining in precise recommendation of e-commerce platforms. Appl Math Non-linear Sci. 2025;10(1).
[6] Anne C, Levana H. How social media are collecting more of users’ data: a behavioral model of platform retention strategies. SN Bus Econ. 2023;3(7).
[7] Long J. Research on purchasing behaviour prediction of e-commerce platform users based on multidimensional data mining. Int J Web Based Commun. 2023;19(4):305-19. DOI: 10.1504/IJWBC.2023.134867.
[8] Yuan X, Wei Y, Yao L. Research and application of user behavior data analysis technology for e-commerce. Appl Math Nonlinear Sci. 2025;10(1).
How to cite this paper
Research on the Evaluation of User Safety Intervention Measures Based on Causal Inference
How to cite this paper: Huisheng Liu. (2025). Research on the Evaluation of User Safety Intervention Measures Based on Causal Inference. Engineering Advances, 5(4), 212-218.
DOI: http://dx.doi.org/10.26855/ea.2025.10.014