A New Hybrid Evolutionary-Based Data Clustering Using Fuzzy Particle Swarm Optimization

In this paper, a new hybrid evolutionary fuzzy-based particle swarm optimization algorithm is proposed for multi-dimensional data clustering. The proposed mechanism integrates an Evolutionary-based approach with a Fuzzy Particle Swarm data Clustering (EFPSC). Unlike other static and centralized clustering techniques, our proposed model can dynamically adapt to the changes and does not require a prior knowledge of the number of clusters in the datasets. It is more adaptive towards problems with dynamic changed information. It has a linear scaling behavior, which make it suitable for use on large data sets. In addition, swarm-based clustering has the capacity to work with any kind of data that can be described in terms of symmetric dissimilarities, and it imposes no assumptions on the shape of the clusters it works with. Finally, an important strength of the algorithm is its ability to automatically determine the number of clusters within the data. The adaptation scheme proposed in the algorithm make it possible to tune with structures exist within the data. One of the key advantages of the EFPSC is the selection of parameter values which offers the good combination of its setting that generates the efficient clustering results in the solution space. The principle of evolutionary approach including its properties like crossover and mutation made a concrete effect in the algorithm. In addition, the guiding rules, which are alignment, cohesion and separation rule, led to better solutions. Numerous experiments will be conducted using both synthetic and real datasets to evaluate the efficiency of the proposed model. Cluster validity approaches are used to quantitatively evaluate the results of the clustering algorithm.