Clustering Methods Research Based On Evolutionary Algorithm

As an unsupervised learning method, clustering has been widely used in many fields such as machine learning, data mining, artificial intelligence and image processing. And, it has been a hotspot research in these fields. With the increasing attention on clustering, several problems need to be solved. Firstly, how to determine the number of clustering automatically. Secondly, how to cluster and get the global optimization solution. Thirdly, how to do clustering for arbitrarily shaped data sets. Fourthly, how to integrate various dif-ferent clustering methods together. Fifthly, how to take clustering method into application fields such as image processing. In this paper, our main contributions are as follows:1) We introduce an automatic clustering method based on evolutionary algorithms (EAs). The basic idea is to convert a clustering problem into a global optimization problem and tackle it by an EA. A new validity index, which balances the inter-cluster consistency and the intra-cluster consistency, is proposed to be the objective function. Three adaptive coding schemes, which can deal with variable-length optimization problems by using a fixed-length chromosome, are designed to detect the cluster number automatically. The validity index and adaptive coding schemes are incorporated in an EA for automatic clustering. Our approach is compared with some widely used validity indices and an adaptive coding scheme on some artificial data sets and two real world problems. The experimental results suggest that our method not only successfully detects the correct cluster numbers but also achieve stable results for most of test problems.2) We propose four methods for automatic detecting the number of clusters in arbitrar-ily shaped data sets. In the first method, path-based clustering is a well-known method for extracting arbitrarily shaped clusters. However, its high time complexity limits some pos-sible applications. In this scheme, we propose two new algorithms to speed up the original path-based clustering method. A basic method focuses on the path-distance calculation. A modified Floyd algorithm is applied to reduce the time complexity. A preprocessing proce-dure is used to reduce the number of data points to the path-based clustering algorithm. Moreover, this algorithm can automatic determine the number of clusters by a box cluster-ing. In the second method, an evolutionary arbitrarily shaped clustering (EASC) method is proposed for extracting arbitrarily shaped clusters. In EASC, the path distance is used to measure the similarity between data points and a modified Modularity index is utilized as the optimization objective. In the third method, we apply real-based coding scheme to represent solutions in search space and establish a new validity index to cluster arbitrarily shaped data sets. In the forth method, we propose a multiobjective evolutionary frame to mix the different clustering algorithms. The new approaches are applied to a variety of test data sets with arbitrarily shapes and the experimental results show that our method is efficient in dealing with the given problems.3) Undersegmentation or oversegmentation is a challenge faced by image segmentation methods, and it is extreme important to determine the current number of regions of an image in real-world applications. In this paper, we introduce an adaptive strategy to do so. The basic idea is to firstly oversegmentation an image by using the Mean-shift method; and then segment the obtained results by using an evolutionary algorithm. In the second stage, a feature is extracted for each region obtained by the Mean-shift method, and a new fitness function is designed to determine the proper number of clusters. The adaptive approach is applied to a variety of images, and the experimental results show that our method is both efficient and effective for image segmentation.4) Although various image segmentation methods have been proposed in recent decades, most of these methods are based on only a single feature space. How to combine various features to image segmentation is a challenge problem. To address this problem, we pro-pose to combine different features based on multiobjective evolutionary optimization. Two optimization objectives, which are based on color and texture features respectively, are therefore designed for image segmentation. The experiments show that our method is able to combine multiple features for image segmentation successfully.