Hybrid K-means Clustering and DID Model for High-Dimensional Data Intervention Evaluation within a Multiple Linear Regression Framework
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
This paper proposes a hybrid analytical framework integrating K-means clustering and Difference-in-Differences (DID) modeling to address high-dimensional data distribution analysis and intervention effect evaluation. A multiple linear regression model is first established to capture the linear mapping between feature and target variables, optimized via gradient descent with a mean squared error cost function and MinMax normalization to mitigate feature scale heterogeneity, with statistical significance verified through an F-test. Subsequently, the K-means algorithm performs unsupervised partitioning of high-dimensional data, where the optimal cluster number is determined by the elbow method, and data points are iteratively assigned to centroids based on minimized Euclidean distance. The DID component employs a bidirectional fixed-effects model to quantify causal effects by comparing temporal changes between treatment and control groups, effectively controlling for confounders. Experimental results demonstrate that the proposed framework ensures robust clustering performance and accurate causal inference, offering a scalable and generalizable solution for complex data-driven intervention analysis.
Fields of Study & Subjects
Advanced Causal Inference Techniques
Advanced Clustering Algorithms Research
Data Mining and Machine Learning Applications
Publication details
- Journal / source
- IEEE International Conference on Power, Electronics and Green Energy (ICPEGE)
- Pages
- 817–822
- Publication date
- 28 January 2026
- Language
- EN