Kybernetika 61 no. 2, 168-184, 2025

Fuzzy clustering of fuzzy data considering the shape of the membership functions using a novel representation learning technique

Alireza Khastan and Elham EskandariDOI: 10.14736/kyb-2025-2-0168

Abstract:

Most existing distance measures for fuzzy data do not capture differences in the shapes of the left and right tails of membership functions. As a result, they may calculate a distance of zero between fuzzy data even when these differences exist. Additionally, some distance measures cannot compute distances between fuzzy data when their membership functions differ in type. In this paper, inspired by human visual perception, we propose a fuzzy clustering method for fuzzy data using a novel representation technique that is capable of detecting small differences in the shapes of the left and right tails of membership functions. Moreover, it effectively clusters fuzzy data even when their membership functions differ in type. By utilizing the pre-trained ResNet50 network as a feature extractor and applying the FCM clustering method to the output from the last convolutional layer, our approach achieves high accuracy in clustering both synthetic and real data sets. Experimental results demonstrate that our method achieved a Rand Index of 0.9965, outperforming state-of-the-art methods, making it particularly suitable for applications that require high clustering accuracy.

Keywords:

fuzzy data, fuzzy clustering, resnet, representation learning, convolutional neural networks

Classification:

62H30, 03E72, 68T10

paper.pdf

References:

  1. R. M. Alguliyev, R. M. Aliguliyev and R. G. Alakbarov: Constrained k-means algorithm for resource allocation in mobile cloudlets. Kybernetika 59 (2023), 1, 88-109.   CrossRef
  2. B. Bede: Mathematics of Fuzzy Sets and Fuzzy Logic. Springer, Berlin, Heidelberg 2013.   CrossRef
  3. J. C. Bezdek: Pattern Recognition with Fuzzy Objective Function Algorithms. Springer, New York 1981.   CrossRef
  4. G. Castellano and G. Vessio: A deep learning approach to clustering visual arts. Int. J. Computer Vision 130 (2022), 11, 2590-2605.   DOI:10.1007/s11263-022-01664-y
  5. Y. Chen, S. Zhou, X. Zhang, D. Liu and C. Fu: Improved fuzzy c-means clustering by varying the fuzziness parameter. Pattern Recogn. Lett. 157 (2022), 60-66.   DOI:10.1016/j.patrec.2022.03.017
  6. R. Coppi, P. D'Urso and P. Giordani: A note on the algorithms for determining the model structureFuzzy and possibilistic clustering for fuzzy data. Comput. Statist. Data Analysis 56 (2012), 4, 915-927.   DOI:10.1016/j.csda.2010.09.013
  7. J. Deng, W. Dong, R. Socher, L. J. Li, K. Li and L. Fei-Fei: Imagenet: A largescale hierarchical image database. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2009, pp. 248-255.   CrossRef
  8. W. Ding, M. Abdel-Basset, H. Hawash and W. Pedrycz: Multimodal infant brain segmentation by fuzzy-informed deep learning. IEEE Trans. Fuzzy Systems 30 (2021), 4, 1088-1101.   DOI:10.1109/TFUZZ.2021.3052461
  9. P. D'Urso and L. De Giovanni: Robust clustering of imprecise data. Chemometr. Intell. Labor. Systems 136 (2014), 58-80.   DOI:10.1016/j.chemolab.2014.05.004
  10. P. D'Urso and P. Giordani: A weighted fuzzy c-means clustering model for fuzzy data. Comput. Statist. Data Analysis 50 (2006), 6, 1496-1523.   DOI:10.1016/j.csda.2004.12.002
  11. P. D'Urso and J. M. Leski: Fuzzy clustering of fuzzy data based on robust loss functions and ordered weighted averaging. Fuzzy Sets Systems 389 (2020), 1-28.   DOI:10.1016/j.fss.2019.03.017
  12. E. Eskandari and A. Khastan: A robust fuzzy clustering model for fuzzy data based on an adaptive weighted L1 norm. Iranian J. Fuzzy Systems 20 (2023), 6, 1-20.   CrossRef
  13. E. Eskandari and A. Khastan: Investigating the effect of using different loss functions on the performance of the fuzzy clustering model for fuzzy data in the presence of outlier data. Fuzzy Systems Appl. 7 (2024), 1, 109-123.   CrossRef
  14. E. Eskandari, A. Khastan and S. Tomasiello: Improved determination of the weights in a clustering approach based on a weighted dissimilarity measure between fuzzy data. In: IEEE International Conference on Fuzzy Systems (FUZZ-IEEE) 2022, pp. 1-6.   CrossRef
  15. Q. Feng, L. Chen, C. L. P. Chen and L. Guo: Deep fuzzy clustering - a representation learning approach. IEEE Trans. Fuzzy Systems 28 (2020), 7, 1420-1433.   DOI:10.1109/TFUZZ.2020.2998378
  16. M. B. Ferraro and P. Giordani: Possibilistic and fuzzy clustering methods for robust analysis of non-precise data. Int. J. Approx. Reasoning 88 (2017), 23-38.   DOI:10.1016/j.ijar.2017.05.002
  17. J. C. Figueroa-Garcia, C. A. Varón-Gaviria and J. L. Barbosa-Fontecha: Fuzzy random variable generation using $\alpha$-cuts. IEEE Trans. Fuzzy Systems 29 (2019), 3, 539-548.   CrossRef
  18. Y. Gao, W. Wang, J. Xie and J. Pan: A new robust fuzzy c-means clustering method based on adaptive elastic distance. Knowledge Based Systems 237 (2022), 107769.   DOI:10.1016/j.knosys.2021.107769
  19. K. He, X. Zhang, S. Ren and J. Sun: Deep residual learning for image recognition. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2016, pp. 770-778.   CrossRef
  20. K. Huang, Y. Zhang, H. D. Cheng, P. Xing and B. Zhang: Semantic segmentation of breast ultrasound image with fuzzy deep learning network and breast anatomy constraints. Neurocomputing 450 (2021), 319-335.   DOI:10.1016/j.neucom.2021.04.012
  21. W. Hung and M. Yang: Fuzzy clustering on LR-type fuzzy numbers with an application in Taiwanese tea evaluation. Fuzzy Sets Systems 150 (2005), 3, 561-577.   DOI:10.1016/j.fss.2004.04.007
  22. N. Kartli, E. Bostanci and M. S. Guzel: Heuristic algorithm for an optimal solution of fully fuzzy transportation problem. Computing 106 (2024), 10, 3195-3227.   DOI:10.1007/s00607-024-01319-5
  23. W. M. Rand: Objective criteria for the evaluation of clustering methods. J. American Statist. Assoc. 66 (1971), 336, 846-850.   DOI:10.1080/01621459.1971.10482356
  24. L. Su and X. Cao: Fuzzy autoencoder for multiple change detection in remote sensing images. J. Appl. Remote Sensing 12 (2018), 3, 035014.   DOI:10.1117/1.jrs.12.035014
  25. X. Wei, D. Lu, X. Cao, L. Su, L. Wang, H. Guo, Y. Hou and X. He: A fuzzy artificial neural network-based method for Cerenkov luminescence tomography. AIP Advances 9 (2019), 6, 065105.   CrossRef
  26. M. Yang and C. Ko: On a class of fuzzy c-numbers clustering procedures for fuzzy data. Fuzzy Sets Systems 84 (1996), 1, 49-60.   DOI:10.1016/0165-0114(95)00308-8