Augmented weighted K-means grey wolf optimizer: An enhanced metaheuristic algorithm for data clustering problems.

Area of Science:

• Artificial Intelligence
• Machine Learning
• Optimization Algorithms

Background:

• Data clustering is crucial for organizing information into meaningful groups.
• Conventional Grey Wolf Optimizer (GWO) struggles with exploration and exploitation for effective clustering.
• Premature convergence limits the performance of standard metaheuristic algorithms.

Purpose of the Study:

• To enhance the Grey Wolf Optimizer (GWO) for improved data clustering performance.
• To address the limitations of GWO in exploration and exploitation capabilities.
• To introduce a novel algorithm, the K-means clustering-based GWO, for superior optimization.

Main Methods:

• Integration of K-means algorithm concepts to refine initial solutions.
• Inclusion of a new weight factor to enhance solution diversity.
• Utilizing a partitional clustering-inspired fitness function for evaluation.

Main Results:

• The K-means clustering-based GWO demonstrates superior performance over the standard GWO.
• Achieved approximately 34% performance improvement in numerical and data clustering tasks.
• Produced high-quality cluster centers efficiently across various datasets.

Conclusions:

• The K-means clustering-based GWO is a robust and dependable method for data clustering.
• Represents a significant advancement over conventional metaheuristic clustering techniques.
• Establishes a new benchmark for future research in metaheuristic clustering algorithms.