Optimal Data Clustering through Hybrid CSO and CD-Fuzzy C means Optimization

Abstract

Clustering is a technique that separates a given set of items into groups in such a way that objects within a cluster are both extremely similar to and dissimilar from those in other clusters The KMH-CSO algorithm is a cross between the KMH-CSO (K means harmonic clustering CSO) and the more traditional cat swarm optimisation. The second method is called Fuzzy C and it involves grouping data based on the density of the clusters. (CD-FCM). The FCM algorithm is versatile enough to be utilised for a number of different data analysis tasks. A generalised multi-objective function is employed to aggregate subsets, and this serves as a clustering criterion for the data. The DB index, the XB index, the sym-index, and the stability measure are only few of the many objectives that are included among the characteristics. In order to tackle the optimisation problem, a relatively new metaheuristic technique called cat swarm optimisation (CSO), which simulates the behaviour of cats, will be used. For the objectives of clustering, many other qualities were chosen, including petroleum oil, iris, wine, glass, cancer, and vowels. The findings have been compiled into tables, and diagrams have been drawn up to show how the data breaks down. The methods that are proposed are much more effective in comparison to other possible options.