Search

Data clustering is a typical data analysis approach that is utilised in a variety of domains, namely machine learning, pattern matching, and visual analytics. K-means clustering is a popular and straightforward solution to data clustering, although it has important shortcomings, including local optimum convergence and initial point sensitivities. To attend the challenge of local convergence of optimal clusters in this article a swam-based optimization technique is proposed. Firefly method is a swarm-based technique used for optimizing challenges. This research proposes a tale approach for clustering data using the firefly algorithm. It is demonstrated how the K-Meanstechnique may be applied to locate the centroidsfor the known initial cluster centres. The approach was later enhanced to improve centroids and clusters using firefly optimization. This novel algorithm is known as AFA. The experimental findings demonstrated the suggested method's efficiency and capabilities for data clustering and the conclusions show that the suggestedmodel outperform traditional K-means clustering.