Unsupervised domain adaptation with fuzzy similarity and fuzzy C-means for classification applications

Jhang, Jyun-Yu; Lin, Cheng-Jian; Lin, Xue-Qian; Cheng, Han

doi:10.22111/ijfs.2026.53085.9388

Articles in Press

Document Type : Original Manuscript

Authors

¹ Department of Computer Science and Information Engineering, National Taichung University of Science and Technology

² National Chin-Yi University of Technology

³ Department of Computer Science and Information Engineering, National Cheng Kung University

⁴ Department of Computer Science and Information Engineering, National Chin-Yi University of Technology

https://doi.org/10.22111/ijfs.2026.53085.9388

Abstract

As deep learning applications continue to expand, the performance of network models becomes increasingly critical. To enhance model accuracy, training often relies on a large amount of labeled data. However, manually labeling data is both labor-intensive and expensive. In recent years, Domain Adaptation (DA) techniques have gained significant attention as they enable models to be trained with limited labeled data while maintaining performance on unlabeled data. In particular, Unsupervised Domain Adaptation (UDA) faces a major challenge due to differences in data distributions. Addressing the problem of domain shift and identifying domain-invariant features with limited information remains a crucial task. This paper proposes a fuzzy-theory-based unsupervised domain adaptation approach called the Deep Fuzzy- Similarity-based Local Domain Adaptation Network (DFS-LDAN). The proposed method aims to mitigate domain shift and reduce annotation costs. DFS-LDAN integrates fuzzy similarity and Maximum Mean Discrepancy (MMD), allowing the model to retain more soft information during learning. This enhances its flexibility in handling uncertainties in target domain data. Additionally, this study introduces a Local Domain Adaptation strategy that aligns distributions at the class level. It employs Fuzzy C-Means (FCM) clustering to adaptively adjust feature weights based on membership degrees, further improving the model’s generalization capability. The effectiveness of DFS-LDAN is evaluated on three public datasets: Office-31, ImageCLEF-DA, and Office-Home. Comparative experiments with various domain adaptation methods demonstrate the superiority of DFS-LDAN. The results show that the proposed method, across the best-performing variant, achieves average accuracies of 89.1% (via Product-Sum), 90.4% (via Product-Sum), and 68.4% (via Min-Max), respectively.

Keywords

Main Subjects

Computer Science

References

[1] B. Caputo, H. Müller, J. Martinez-Gomez, et al., ImageCLEF 2014: Overview and analysis of the results, Information Access Evaluation. Multilinguality, Multimodality, and Interaction. CLEF 2014. Lecture Notes in Computer Science, 8685, Springer, Cham., (2014), 192-211. https://doi.org/10.1007/978-3-319-11382-1_18

[2] M. Eftekhari, A. Mehrpooya, F. Saberi-Movahed, V. Torra, How fuzzy concepts contribute to machine learning, Studies in Fuzziness and Soft Computing, (2022). https://doi.org/10.1007/978-3-030-94066-9

[3] Y. Ganin, E. Ustinova, H. Ajakan, et al., Domain-adversarial training of neural networks, Domain Adaptation in Computer Vision Applications. Advances in Computer Vision and Pattern Recognition. Springer, Cham., (2017). https://doi.org/10.1007/978-3-319-58347-1_10

[4] O. Gilo, J. Mathew, S. Mondal, R. K. Sanodiya, RDAOT: Robust unsupervised deep sub-domain adaptation through optimal transport for image classification, IEEE Access, 11 (2023), 102243-102260. https://doi.org/10.1109/ ACCESS.2023.3313623

[5] M. Golhar, T. L. Bobrow, M. P. Khoshknab, et al., Improving colonoscopy lesion classification using semi-supervised deep learning, IEEE Access, 9 (2021), 631-640. https://doi.org/10.1109/access.2020.3047544

[6] I. Goodfellow, J. Pouget-Abadie, M. Mirza, et al., Generative adversarial nets, Communications of the ACM, 63(11) (2020), 139-144. https://doi.org/10.1145/3422622

[7] A. Gretton, K. M. Borgwardt, M. J. Rasch, et al., A kernel two-sample test, Journal of Machine Learning Research, 13(25) (2012), 723-773. https://doi.org/10.5555/2188385.2188410

[8] M. M. Hasan, M. M. Hossain, M. M. Rahman, et al., FP-CNN: Fuzzy pooling-based convolutional neural network for lung ultrasound image classification with explainable AI, Computers in Biology, Medicine, 165 (2023), 107407. https://doi.org/10.1016/j.compbiomed.2023.107407

[9] K. He, X. Zhang, S. Ren, J. Sun, Deep residual learning for image recognition, IEEE Conference on Computer Vision and Pattern Recognition (CVPR), (2016), 770-778. https://doi.org/10.1109/CVPR.2016.90

[10] L. Huang, W. Zhao, Y. Liu, et al., An evidential multi-target domain adaptation method based on weighted fusion for cross-domain pattern classification, IEEE Transactions on Neural Networks, Learning Systems, 35(10) (2024), 14218-14232. https://doi.org/10.1109/TNNLS.2023.3275759

[11] T. Kim, S. Lee, A novel unsupervised clustering and domain adaptation framework for rotating machinery fault diagnosis, IEEE Transactions on Industrial Informatics, 19(9) (2023), 9404-9412. https://doi.org/10.1109/ TII.2022.3228395

[12] J. Lee, G. Lee, Unsupervised domain adaptation based on the predictive uncertainty of models, Neurocomputing, 520 (2023), 183-193. https://doi.org/10.1016/j.neucom.2022.11.070

[13] K. Li, J. Lu, H. Zuo, G. Zhang, Multi-source contribution learning for domain adaptation, IEEE Transactions on Neural Networks and Learning Systems, 33(10) (2022), 5293-5307. https://doi.org/10.1109/TNNLS.2021. 3069982

[14] K. Li, J. Lu, H. Zuo, G. Zhang, Source-free multidomain adaptation with fuzzy rule-based deep neural networks, IEEE Transactions on Fuzzy Systems, 31(12) (2023), 4180-4194. https://doi.org/10.1109/TFUZZ.2023. 3276978

[15] M. Long, Y. Cao, Z. Cao, et al., Transferable representation learning with deep adaptation networks, IEEE Transactions on Pattern Analysis and Machine Intelligence, 41(12) (2019), 3071-3085. https://doi.org/10.1109/TPAMI. 2018.2868685

[16] M. Long, Y. Cao, J. Wang, et al., Learning transferable features with deep adaptation networks, Proceedings of the 32nd International Conference on Machine Learning, 37 (2015), 97-105. https://doi.org/10.5555/3045118. 3045130

[17] M. Long, Z. Cao, J. Wang, M. I. Jordan, Conditional adversarial domain adaptation, NIPS’18: Proceedings of the 32nd International Conference on Neural Information Processing Systems, (2018), 1647-1657.

[18] M. Long, H. Zhu, J. Wang, M. I. Jordan, Deep transfer learning with joint adaptation networks, Proceedings of the 34th International Conference on Machine Learning, 70 (2017), 2208-2217. https://doi.org/10.5555/3305890. 3305909

[19] Q. Lou, W. Sun, W. Zhang, et al., Rules-based heterogeneous feature transfer learning using fuzzy inference, IEEE Transactions on Fuzzy Systems, 32(1) (2024), 306-321. https://doi.org/10.1109/TFUZZ.2023.3298147

[20] J. Lu, V. Behbood, P. Hao, et al., Transfer learning using computational intelligence: A survey, Knowledge-Based Systems, 80 (2015), 14-23. https://doi.org/10.1016/j.knosys.2015.01.010

[21] Y. Lu, H. Huang, B. Zeng, et al., Multi-source and multi-target domain adaptation based on dynamic generator with attention, IEEE Transactions on Multimedia, 26 (2024), 6891-6905. https://doi.org/10.1109/TMM.2024. 3358062

[22] J. Lu, H. Zuo, G. Zhang, Fuzzy multiple-source transfer learning, IEEE Transactions on Fuzzy Systems, 28(12) (2020), 3418-3431. https://doi.org/10.1109/TFUZZ.2019.2952792

[23] A. Ma, J. Li, K. Lu, et al., Adversarial entropy optimization for unsupervised domain adaptation, IEEE Transactions on Neural Networks and Learning Systems, 33(11) (2022), 6263-6274. https://doi.org/10.1109/TNNLS.2021. 3073119

[24] G. Ma, J. Lu, G. Zhang, Multisource domain adaptation with interval-valued target data via fuzzy neural networks, IEEE Transactions on Fuzzy Systems, 32(5) (2024), 3094-3106. https://doi.org/10.1109/TFUZZ.2024.3367456

[25] C. Qin, L. Wang, Q. Ma, et al., Semi-supervised domain adaptive structure learning, IEEE Transactions on Image Processing, 31 (2022), 7179-7190. https://doi.org/10.1109/TIP.2022.3215889

[26] K. Saenko, B. Kulis, M. Fritz, T. Darrell, Adapting visual category models to new domains, European Conference on Computer Vision (ECCV), (2010), 213-226. https://doi.org/10.1007/978-3-642-15561-1_16

[27] E. Tzeng, J. Hoffman, N. Zhang, et al., Deep domain confusion: Maximizing for domain invariance, arXiv:1412.3474, (2014). https://doi.org/10.48550/arXiv.1412.3474

[28] H. Venkateswara, J. Eusebio, S. Chakraborty, S. Panchanathan, Deep hashing network for unsupervised domain adaptation, IEEE Conference on Computer Vision and Pattern Recognition (CVPR), (2017), 5385-5394. https: //doi.org/10.1109/CVPR.2017.572

[29] Z. Wang, L. Chen, F. Y. Wang, Fuzzy inference attention module for unsupervised domain adaptation, IEEE Transactions on Fuzzy Systems, 32(4) (2024), 1706-1718. https://doi.org/10.1109/TFUZZ.2023.3332751

[30] J. Wang, W. Feng, Y. Chen, et al., Visual domain adaptation with manifold embedded distribution alignment, Proceedings of the 26th ACM International Conference on Multimedia, (2018), 402-410. https://doi.org/10. 1145/3240508.3240512

[31] X. Wang, L. Li, W. Ye, et al., Transferable attention for domain adaptation, in Proceedings of the AAAI Conference on Artificial Intelligence, 33(01) (2019), 5345-5352. https://doi.org/10.1609/aaai.v33i01.33015345

[32] X. Wei, Z. Ji, B. Wen, et al., Unsupervised multi-target domain adaptation incremental method based on contrastive learning, IEEE International Conference on Multimedia, (2024), 1-6. https://doi.org/10.1109/ICME57554. 2024.10687405

[33] J. Wu, S. Wang, J. Sun, AMMD: Attentive maximum mean discrepancy for few-shot image classification, Pattern Recognition, 155 (2024), 110680. https://doi.org/10.1016/j.patcog.2024.110680

[34] H. Xiao, W. Wang, H. Ogai, et al., Adversarial fuzzy-weighted deep transfer learning for intelligent damage diagnosis of bridge with multiple new damages, IEEE Sensors Journal, 22(17) (2022), 17005-17021. https://doi.org/10. 1109/JSEN.2022.3192307

[35] P. Xu, Z. Deng, J. Wang, et al., Transfer representation learning with TSK fuzzy system, IEEE Transactions on Fuzzy Systems, 29(3) (2021), 649-663. https://doi.org/10.1109/TFUZZ.2019.2958299

[36] Y. Ye, T. Pan, Q. Meng, et al., Online unsupervised domain adaptation via reducing inter- and intra-domain discrepancies, IEEE Transactions on Neural Networks and Learning Systems, 35(1) (2024), 884-898. https:// doi.org/10.1109/TNNLS.2022.3177769

[37] H. Yu, M. Hu, S. Chen, Multi-target unsupervised domain adaptation without exactly shared categories, arXiv:1809.00852, (2018). https://doi.org/10.48550/arXiv.1809.00852

[38] F. Zhang, F. Chen, Z. Su, J. Wei, Unsupervised domain adaptation via dynamic clustering and co-segment attentive learning for video-based person re-identification, IEEE Access, 12 (2024), 29583-29595. https://doi.org/10. 1109/ACCESS.2024.3365583

[39] Y. Zhu, F. Zhuang, J. Wang, et al., Deep subdomain adaptation network for image classification, IEEE Transactions on Neural Networks and Learning Systems, 32(4) (2021), 1713-1722. https://doi.org/10.1109/TNNLS.2020. 2988928

Iranian Journal of Fuzzy Systems

Unsupervised domain adaptation with fuzzy similarity and fuzzy C-means for classification applications

References

References

Articles in Press, Accepted Manuscript
Available Online from 01 October 2026

Unsupervised domain adaptation with fuzzy similarity and fuzzy C-means for classification applications

References

References

Articles in Press, Accepted Manuscript Available Online from 01 October 2026

Articles in Press, Accepted Manuscript
Available Online from 01 October 2026