Abstract
Clustering algorithms have attracted a lot of attentions recently in real-world applications. However, the traditional clustering algorithms still have plenty of defects which are not yet resolved. In this paper, a kernel-based intuitionistic fuzzy C-means clustering using improved multi-objective artificial immune algorithm (KIFCM-IMOIA) is proposed. In our algorithm, the kernel trick and the intuitionistic fuzzy entropy (IFE) are introduced into the objective functions, which improves the robustness to noises. In addition, an improved multi-objective optimization immune algorithm (IMOIA), which simultaneously optimizes the intra-cluster compactness and inter-cluster separation, is proposed to prevent the algorithm from falling into local optimum. The proposed IMOIA uses a novel active antibody selection strategy, a hybrid differential evolution strategy, and an adaptive mutation operator to maintain better distribution of the solutions with better convergence. Finally, we performed experiments using 14 UCI datasets and compared our algorithm with six clustering methods on three performance metrics. The experimental results show that our algorithm performs better than other algorithms.
Highlights
As an unsupervised classification method, clustering algorithm is a research hotspot in recent decades and is widely used in pattern recognition [1], data mining [2], [3], image segmentation [4]–[6] and so on
In order to further improve the performance of the algorithm, the intuitionistic fuzzy C-means (IFCM) clustering algorithm was proposed in [12]
A KERNEL-BASED INTUITIONISTIC FUZZY C-MEANS CLUSTERING USING IMPROVED MULTI-OBJECTIVE IMMUNE ALGORITHM we present the basic idea and the general framework of the proposed algorithm KIFCM-improved multi-objective optimization immune algorithm (IMOIA)
Summary
As an unsupervised classification method, clustering algorithm is a research hotspot in recent decades and is widely used in pattern recognition [1], data mining [2], [3], image segmentation [4]–[6] and so on. The above FCM algorithms have achieved some promising results, they are sensitive to noises and outliers seeing that the Euclidean distance is used as the similarity measurement Aiming at this problem, various improved methods have been proposed, in which the kernel-based clustering method has attracted extensive attention. By incorporating the intuitionistic fuzzy set and the kernel trick, two completely independent fuzzy objective functions JKIFCM and SKIFCM , which consider the intra-cluster compactness and the inter-cluster separation respectively, are proposed. This is done to improve the robustness to noises.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.