Automatic Determination of the Number of Clusters for Semi-Supervised Relational Fuzzy Clustering

다국어 초록 (Multilingual Abstract)

Semi-supervised clustering relies on both labeled and unlabeled data to steer the clustering process towards optimal categorization and escape from local minima. In this paper, we propose a novel fuzzy relational semi-supervised clustering algorithm based on an adaptive local distance measure (SSRF-CA). The proposed clustering algorithm utilizes side-information and formulates it as a set of constraints to supervise the learning task. These constraints are expressed using reward and penalty terms, which are integrated into a novel objective function. In particular, we formulate the clustering task as an optimization problem through the minimization of the proposed objective function. Solving this optimization problem provides the optimal values of different objective function parameters and yields the proposed semi-supervised clustering algorithm. Along with its ability to perform data clustering and learn the underlying dissimilarity measure between the data instances, our algorithm determines the optimal number of clusters in an unsupervised manner. Moreover, the proposed SSRF-CA is designed to handle relational data. This makes it appropriate for applications where only pairwise similarity (or dissimilarity) information between data instances is available. In this paper, we proved the ability of the proposed algorithm to learn the appropriate local distance measures and the optimal number of clusters while partitioning the data using various synthetic and real-world benchmark datasets that contain varying numbers of clusters with diverse shapes. The experimental results revealed that the proposed SSRF-CA accomplished the best performance among other state-of-the-art algorithms and confirmed the outperformance of our clustering approach.

목차 (Table of Contents)

• Abstract
• 1. Introduction
• 2. Related Works
• 3. Proposed Approach
• 4. Experimental Results
• 5. Conclusions
• References

참고문헌 (Reference)

1 N. Grira, "Unsupervised and semi-supervised clustering: a brief survey"

2 E. Frias-Martinez, "Survey of data mining approaches to user modeling for adaptive hypermedia" 36 (36): 734-749, 2006

3 R. Greenlaw, "Survey of clustering : algorithms and applications" 3 (3): 1-29, 2013

4 O. Bchir, "Semisupervised relational fuzzy clustering with local distance measure learning" 1-4, 2013

5 X. Yin, "Semi-supervised clustering with metric learning : an adaptive kernel method" 43 (43): 1320-1333, 2010

6 R. J. Hathaway, "Relational duals of the c-means clustering algorithms" 22 (22): 205-212, 1989

7 D. M. Duc, "Lagrange multipliers theorem and saddle point optimality criteria in mathematical programming" 323 (323): 441-455, 2006

8 M. Bilenko, "Integrating constraints and metric learning in semi-supervised clustering" 2004

9 O. Bchir, "Fuzzy clustering with Learnable Cluster dependent Kernels" 2521-2527, 2011

10 G. Raju, "Fuzzy clustering methods in data mining: a comparative case analysis" 489-493, 2008

11 C. Borgelt, "Finding the number of fuzzy clusters by resampling" 48-54, 2006

12 J. M. Yih, "FCM & FPCM algorithm based on unsupervised Mahalanobis distances with better initial values and separable criterion" 3 (3): 9-18, 2009

13 O. Nasraoui, "Extracting web user profiles using relational competitive fuzzy clustering" 9 (9): 509-526, 2000

14 H. Frigui, "Clustering by competitive agglomeration" 30 (30): 1109-1119, 1997

15 A. Skabar, "Clustering Sentence-level text using a novel fuzzy relational clustering algorithm" 25 (25): 62-75, 2013

16 S. Saha, "Brain image segmentation using semi-supervised clustering" 52 : 50-63, 2016

17 O. Arbelaitz, "An extensive comparative study of cluster validity indices" 46 (46): 243-256, 2013

18 Z. Yu, "Adaptive ensembling of semisupervised clustering solutions" 29 (29): 1577-1590, 2017

19 S. Xiong, "Active Learning of Constraints for Semi-Supervised Clustering" 26 (26): 43-54, 2014

20 S. Wazarkar, "A survey on image data analysis through clustering techniques for real world applications" 55 : 596-626, 2018

21 S. Zeng, "A study on semi-supervised FCM algorithm" 35 (35): 585-612, 2013

22 C. F. Gao, "A new semi-supervised clustering algorithm with pairwise constraints by competitive agglomeration" 11 (11): 5281-5291, 2011

23 D. Xu, "A comprehensive survey of clustering algorithms" 2 (2): 165-193, 2015