An evolutionary chimp-based chimp-based metaheuristic of data clustering with intelligent learning system...

Area of Science:

• Computer Science
• Artificial Intelligence
• Data Mining

Background:

• Data clustering is crucial for decision-making in complex, large-scale, high-dimensional, and heterogeneous systems.
• Existing clustering methods may struggle with exploration-exploitation balance and premature convergence.

Purpose of the Study:

• To introduce an evolutionary adaptation of the Chimp Optimization Algorithm (EVCHOA) for enhanced data clustering.
• To improve the quality and strength of clustering in real-world applications, particularly intelligent learning systems.

Main Methods:

• Implemented an adaptive evolutionary update mechanism within the chimp-inspired search process.
• Tested EVCHOA on real-life datasets using a distributed computing framework.
• Compared EVCHOA against K-means, DBSCAN, hierarchical clustering, mean shift, and Gaussian mixture models.

Main Results:

• EVCHOA demonstrated superior clustering performance, yielding more coherent group structures.
• Performance was validated using Davies-Bouldin Index, Silhouette Score, Adjusted Rand Index (ARI), and Calinski-Harabasz Index.
• The algorithm showed improved exploration-exploitation ratio and avoided early convergence.

Conclusions:

• EVCHOA offers a robust and scalable solution for data clustering and decision support in operational research.
• In intelligent learning environments, EVCHOA enables accurate student profiling, targeted interventions, and adaptive resource distribution.
• Evolutionary metaheuristics provide valuable data-driven tools for complex decision-making challenges.