M – Sarcasm: A Robust Deep Learning Framework For Enhanced Sarcasm Detection Of Marathi Tweets
DOI:
https://doi.org/10.29284/03x47387Keywords:
Marathi NLP, Dataset Augmentation, Deep Learning, Attention Mechanism, XAIAbstract
Detecting sarcasm in low-resource languages is challenging because of data scarcity, complexity of language, and cultural aspects. This paper fills the gap in detecting sarcasm automatically from Marathi tweets by using a framework that integrates dataset augmentation, advanced deep learning techniques with Attention Mechanisms. The methods for augmenting data to enhance the small size of the dataset included synonym replacement, back-translation, noise injection, and contextual augmentation, resulting in 13,438 total samples. The models were evaluated using deep learning approaches, with the proposed hybrid CNN-LSTM with attention model achieving the highest performance (Accuracy: 0.9948). Attention Mechanisms allowed the model to focus attention on opposing phrases or words that are exaggerated to indicate sarcasm. Explainable AI (LIME and SHAP) was used to understand the reasons for predictions made by the model and showed that irony, code-mixing, punctuation cues, and positive words are the top contributing features for sarcasm detection. This study shows that a combination of augmentation, advanced deep learning, and explainability can create high-performance interpretable systems for detecting sarcasm in low-resource languages, thus improving the ability to identify sentiment in social media content in Marathi.
References
1. Kumar, A., Singh, J. P., & Singh, A. K. (2023). Explainable BERT-LSTM stacking for sentiment analysis of covid-19 vaccination. IEEE Transactions on Computational Social Systems.
2. Kumar, A., & Singh, J. P. (2022). Deep neural networks for location reference identification from bilingual disaster-related tweets. IEEE Transactions on Computational Social Systems, 11(1), 880-891.
3. Kumar, A., Singh, J. P., Rana, N. P., & Dwivedi, Y. K. (2023). Multi-channel convolutional neural network for the identification of eyewitness tweets of disaster. Information Systems Frontiers, 25(4), 1589-1604.
4. Pandey, R., Kumar, A., Singh, J. P., & Tripathi, S. (2021). Hybrid attention-based long short-term memory network for sarcasm identification. Applied Soft Computing, 106, 107348.
5. Pandey, R., Kumar, A., Singh, J. P., & Tripathi, S. (2025). A hybrid convolutional neural network for sarcasm detection from multilingual social media posts. Multimedia Tools and Applications, 84(16), 15867-15895.
6. Yue, T., Shi, X., Mao, R., Hu, Z., & Cambria, E. (2024, May). SarcNet: a multilingual multimodal sarcasm detection dataset. In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024) (pp. 14325-14335).
7. Maynard, D. G., & Greenwood, M. A. (2014, March). Who cares about sarcastic tweets? investigating the impact of sarcasm on sentiment analysis. In Lrec 2014 proceedings. ELRA.
8. Bali, K., Sharma, J., Choudhury, M., & Vyas, Y. (2014, October). “i am borrowing ya mixing?" an analysis of english-hindi code mixing in facebook. In Proceedings of the first workshop on computational approaches to code switching (pp. 116-126).
9. Kim, Y., Yim, J., Yun, J., & Kim, J. (2019). Nlnl: Negative learning for noisy labels. In Proceedings of the IEEE/CVF international conference on computer vision (pp. 101-110).
10. Barbieri, F., Ballesteros, M., & Saggion, H. (2017). Are emojis predictable?. arXiv preprint arXiv:1702.07285.
11. Barman, U., Das, A., Wagner, J., & Foster, J. (2014, October). Code mixing: A challenge for language identification in the language of social media. In Proceedings of the first workshop on computational approaches to code switching (pp. 13-23).
12. Attardo, S. (2024). Linguistic theories of humor (Vol. 1). Walter de Gruyter GmbH & Co KG.
13. Ribeiro, Marco Tulio, Sameer Singh, and Carlos Guestrin. "" Why should i trust you?" Explaining the predictions of any classifier." In Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining, pp. 1135-1144. 2016.
14. Gibbs, R. W. (1986). On the psycholinguistics of sarcasm. Journal of experimental psychology: general, 115(1), 3.
15. Glucksberg, S. (1989). How to be sarcastic: The echoic reminder theory of verbal irony. Journal of Experimental Psychology: General, 118(4), 374-386.
16. Utsumi, A. (2000). Verbal irony as implicit display of ironic environment: Distinguishing ironic utterances from nonirony. Journal of pragmatics, 32(12), 1777-1806.
17. Gibbs, R. W., & Colston, H. L. (Eds.). (2007). Irony in language and thought: A cognitive science reader. Psychology Press.
18. Joshi, A., Bhattacharyya, P., & Carman, M. J. (2017). Automatic sarcasm detection: A survey. ACM Computing Surveys (CSUR), 50(5), 1-22.
19. Davidov, D., Tsur, O., & Rappoport, A. (2010, July). Semi-supervised recognition of sarcasm in Twitter and Amazon. In Proceedings of the fourteenth conference on computational natural language learning (pp. 107-116).
20. Reyes, A., Rosso, P., & Buscaldi, D. (2012). From humor recognition to irony detection: The figurative language of social media. Data & Knowledge Engineering, 74, 1-12.
21. Riloff, E., Qadir, A., Surve, P., De Silva, L., Gilbert, N., & Huang, R. (2013, October). Sarcasm as contrast between a positive sentiment and negative situation. In Proceedings of the 2013 conference on empirical methods in natural language processing (pp. 704-714).
22. Ptáček, T., Habernal, I., & Hong, J. (2014, August). Sarcasm detection on czech and english twitter. In Proceedings of COLING 2014, the 25th international conference on computational linguistics: Technical papers (pp. 213-223).
23. Bagate, R. A., Joshi, A. S., Kadam, A., Choubey, C. K., Sable, N., Kumar, A., ... & Nandan, D. (2025). Sarcasm Detection an Explainable AI Approach for Reddit Political Text. Mathematical Modelling of Engineering Problems, 12(1).
24. Pakray, P., Gelbukh, A., & Bandyopadhyay, S. (2025). Natural language processing applications for low-resource languages. Natural Language Processing, 31(2), 183-197.
25. Dhumpati, R., Sasi, A., Basha, S. J., Vadhri, S., Saroja, P., Alluri, S., & Vatambeti, R. (2025). Enhancing sarcasm detection in sentiment analysis for cyberspace safety using advanced deep learning techniques. Scientific Reports, 15(1), 22681.
26. Golbazi, J. (2025). Development and Application of a Hybrid Deep Learning Model for Aspect-Based Sentiment Analysis (ABSA): A Comparative Study of McDonald’s Tweets Across Two Time Periods.
27. Chauhan, S., & Kumar, A. (2024). A Transformer-Based Model for Detecting Multilingual Sarcasm in Social Media Posts. In Forum of Information Retrieval and Evaluation FIRE-2024.
28. Sree, M. R., Siddhartha, M., Reddy, P. V. V., & Belwal, M. (2024, June). Sentiment Analysis and Likes Prediction from Social Media Comments: An ML approach. In 2024 15th International Conference on Computing Communication and Networking Technologies (ICCCNT) (pp. 1-8). IEEE.
29. Patil, P. K., & Kolhe, S. R. (2023, June). Sarcasm Detection for Marathi and the role of emoticons. In International Conference on Data Intelligence and Cognitive Informatics (pp. 193-204). Singapore: Springer Nature Singapore.
30. Tan, Y. Y., Chow, C. O., Kanesan, J., Chuah, J. H., & Lim, Y. (2023). Sentiment analysis and sarcasm detection using deep multi-task learning. Wireless personal communications, 129(3), 2213-2237.
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution 4.0 International License.
This work is licensed under a Creative Commons Attribution 4.0 International License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
