An evidential clustering algorithm by finding belief-peaks and disjoint neighborhoods

• Published in: Pattern recognition 2021-05, Vol.113, p.107751, Article 107751
• Main Authors: Gong, Chaoyu, Su, Zhi-gang, Wang, Pei-hong, Wang, Qian
• Format: Article
• Language: English
• Subjects: Belief-peaks; Disjoint neighborhood; Evidential clustering; Proximity data
• ISSN: 0031-3203; 1873-5142
• DOI: 10.1016/j.patcog.2020.107751

•A new evidential clustering algorithm based on finding the belief-peaks and disjoint neighborhoods, called BPDNEC, is proposed.•BPDNEC automatically detects cluster centers based on a new assumption, which concerns the size of neighborhood rather than delta metric.•BPDNEC determines the appropriate size of disjoint neighborhoods by solving an equation and thus, avoids problem of parameter setting for K.•BPDNEC creates a credal partition by minimizing an objective function and available for both proximity data and object data. In this paper, we introduce a new evidential clustering algorithm based on finding the belief-peaks and disjoint neighborhoods, called BPDNEC. The basic idea of BPDNEC is that each cluster center has the highest possibility of becoming a cluster center among its neighborhood and neighborhoods of those cluster centers are disjoint in vector space. Such possibility is measured by the belief notion in framework of evidence theory. By solving an equation related to neighborhood size, the size of such disjoint neighborhoods is determined and those objects having highest belief among their neighborhoods are automatically detected as cluster centers. Finally, a credal partition is created by minimizing an objective function concerning dissimilarity matrix of data objects. Experimental results show that BPDNEC can automatically detect cluster centers and derive an appropriate credal partition for both object data and proximity data. Simulations on synthetic and real-world datasets validate the conclusions.