Journal of Engineering, Project, and Production Management, 2026, 16(4), 2025-287
An Efficient Unsupervised Data Mining Method Using Adaptive Grid-Entropy Density Clustering
1 Lecturer,
School of Information Engineering, Henan Vocational College of
Agriculture, Zhengzhou, 451450, China,
E-mail: WeiWangw.w@outlook.com (corresponding author).
2 Lecturer, School of Information
Engineering, Henan Agricultural Vocational College of Agriculture,
Zhengzhou, 451450, China
Project Management
Received; revised April 29, 2026; May 8, 2026; accepted May 14, 2026
Available online May 29, 2026
Abstract: Density-based clustering algorithms in unsupervised data mining often suffer from heavy data processing loads, slow computation, and difficulties in determining density thresholds. Therefore, a modified form of clustering using an optimized algorithm for sorting point recognition is proposed herein. This algorithm initially involves adaptive partitioning of the whole space using density-based grid partitioning. This will be followed by the calculation of weighted information entropy considering the probability distribution of points in the grid. Finally, by merging adjacent high-density grids and extracting their weighted centroids, clustering analysis is performed on the centroids rather than the original data points to reduce computational complexity. The results demonstrate that the optimized algorithm reduces sample computations by over 90% in clustered datasets. When compared with mainstream density-based clustering algorithms, it achieves a response time of 19 seconds and a classification accuracy above 95% on datasets with 10,000 samples. When using a single node and running time as the benchmark, the acceleration ratio of the proposed algorithm reaches 4.6 when using two nodes. And in all five datasets, the contour coefficient remains above 0.8. From these results, it can be observed that the enhanced algorithm has a lot of advantages, such as good parallel processing capabilities, high efficiency in data mining, and high effectiveness in unsupervised learning. It can solve the problem of computational redundancy in density-based clustering algorithms.
Keywords: Adaptive grid refinement, density-based clustering, information entropy, parallel computing, unsupervised data mining.
Copyright © Journal of Engineering, Project, and Production Management (EPPM-Journal).
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License.
Requests for reprints and permissions at eppm.journal@gmail.com.
Citation: Wang, W. and Ran, C. (2026). An Efficient Unsupervised Data Mining Method Using Adaptive Grid-Entropy Density Clustering. Journal of Engineering, Project, and Production Management, 16(4), 2025-287.
