A modified and efficient shuffled frog leaping algorithm (MSFLA) for unsupervised data clustering

Abstract

Shuffled frog leaping Algorithm (SFLA) is a new memetic, population based, meta-heuristic algorithm, has emerged as one of the fast, robust with efficient global search capability. In order to enhance the algorithm's stability and the ability to search the global optimum, the conventional SFL Algorithm has been modified in our work by using the local best value of each memeplex instead of generating a new frog, to enhance the effectiveness of the SFLA. This paper implements the application of Modified SFLA in Partitional clustering of the unlabelled data. This algorithm is applied on various classification problems and the simulated results demonstrate that, this modified SFLA has outperformed the conventional SFL Algorithm. © 2011 Springer-Verlag.