The feasible solution algorithm for fuzzy Least Trimmed Squares clustering

The issue of sensitivity to noise and outliers of LS minimization based clustering techniques, is addressed in this paper. A novel and robust clustering scheme based on the feasible solution algorithm that implements the Least Trimmed Squares (LTS) estimator, is developed, implemented and the results presented. The LTS estimator is known to be resistant to noise and has a high breakdown point. The feasible solution algorithm approach also guarantees convergence of the solution set to a global optima. Our experiments show the practicability of the proposed scheme in terms of computational requirements and in the attractiveness of its simplistic framework.