K-Prototype clustering algorithm stays simple, ascendible and only modifiable into some spread of contexts and software domains. It integrates the choices of K-Means along with K-Modes clustering to package information with joint numerical and categorical values. Thus, this work concentrates on K-Prototype clustering algorithm to eliminate the restriction and also to preserve its effectiveness. As optimization established clustering will raise the accuracy and effectiveness, the Firefly algorithm was properly improve the clustering effects to acquire optimal answer. Hence, within this paper Firefly established K-Prototype clustering algorithm was enforced and assessed over five standard routine data collections which has been obtained out of UCI machine learning repository. The results are contrasted using Firefly and K-Prototype clustering calculations to show the effectiveness of the suggested algorithm.
Volume 11 | 07-Special Issue
Pages: 1781-1789