EXPLORING THE EFFECTIVENESS OF DATA AUGMENTATION TECHNIQUES IN MACHINE LEARNING AND DEEP LEARNING: APPLICATION TO IMAGE RECOGNITION

Dao Thi Le Thuy; Chu Ba Thanh

Dao Thi Le Thuy Faculty of Information Technology, University of Transport and Communications
Chu Ba Thanh Faculty of Information Technology, Hung Yen University of Technology and Education

Abstract

In this study, we explore data augmentation techniques aimed at enhancing the accuracy of image recognition models in machine learning and deep learning. By applying methods such as rotation, scaling, flipping, and advanced techniques like GAN-based augmentation, we address challenges related to data scarcity and imbalance. Experimental results on the Dogs vs. Cats dataset show significant improvements in model performance, with combined data augmentation techniques achieving the highest accuracy of 90.83%. This study underscores the critical role of data augmentation in improving model robustness and generalization, especially in scenarios with limited data. We also discuss the implications of these findings for future research in the field of image recognition.

References

Krizhevsky, I. Sutskever, and G. E. Hinton, “ImageNet Classification with Deep Convolutional Neural Networks,” in Advances in Neural Information Processing Systems, 2012.

C. Shorten and T. M. Khoshgoftaar, “A Survey on Image Data Augmentation for Deep Learning,” Journal of Big Data, 2019, vol. 6, no. 1.

J. Wei and K. Zou, “EDA: Easy Data Augmentation Techniques for Boosting Performance on Text Classification Tasks,” Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing, 2019.

S. Kobayashi, “Contextual Augmentation: Data Augmentation by Words with Paradigmatic Relations,” Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, 2018.

T. Ko, V. Peddinti, D. Povey, and S. Khudanpur, “Audio Augmentation for Speech Recognition,” in Proceedings of Interspeech 2015.

D. Tran, L. Bourdev, R. Fergus, L. Torresani, and M. Paluri, “Learning Spatiotemporal Features with 3D Convolutional Networks,” in Proceedings of the IEEE International Conference on Computer Vision (ICCV), 2015.

D. T. L. Thuy, T. V. Loan, C. B. Thanh, and N. H. Cuong, “Music genre classification using densenet and data augmentation”, Computer Systems Science and Engineering, 2023, vol. 47, no.1, pp. 657–674.

Dao Thi Le Thuy, Chu Ba Thanh, Trinh Van Loan, and Le Xuan Thanh, “Automatic Identification of Vietnamese Singer Voices Using Deep Learning and Data Augmentation”, Lecture Notes in Networks and Systems, 2023, 848, pp 237-252.

Loan Trinh Van, Quang Nguyen Hong, Thuy Dao Thi Le, “Emotion Recognition with Capsule Neural Network”, Computer Systems Science and Engineering, 2021, 41(3), pp 1083-1098.

Zhang, Xiang, Junbo Zhao, and Yann LeCun. “Character-level convolutional networks for text classification.” Advances in neural information processing systems, 2015, 28.

Fadaee, Marzieh, Arianna Bisazza, and Christof Monz. “Data augmentation for low-resource neural machine translation.” arXiv preprint arXiv:1705.00440, 2017.

Wei, Jason, and Kai Zou. “EDA: Easy data augmentation techniques for boosting performance on text classification tasks.” arXiv preprint arXiv:1901.11196, 2019.

Edunov, S., Ott, M., Auli, M., & Grangier, D. “Understanding Back-Translation at Scale.” In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing (EMNLP), 2018, pp. 489-500.

Yao, Liang, Chengsheng Mao, and Yuan Luo. “Graph convolutional networks for text classification.” Proceedings of the AAAI conference on artificial intelligence, 2019, Vol. 33, No. 01.

Kim, Kyung Geun, and Byeong Tak Lee. “Graph Structure Based Data Augmentation Method.” arXiv preprint arXiv:2205.14619, 2022.