1 of 11

A Hybrid Approach to Sarcasm Detection in Dravidian

Code-Mixed Texts

By: Prasun Maity, Dipanjan Saha, Sayan Das, Sainik Kumar Mahata, Dipankar Das

Paper ID: 29

Track: Sarcasm Identification of Dravidian Languages Tamil & Malayalam (DravidianCodeMix)

at,

Forum for Information Retrieval (FIRE) - 2024

3 of 11

Datasets (contd...)

4 of 11

System Workflow

5 of 11

Parameters	Value/Description
Embedding Dimensions	128
CNN Filters	128
Kernel Size	5
Pooling Method	MaxPooling
LSTM Configuration	Bidirectional
Dropout Rate	0.5 or 50%
Dense Layer Units	128
Optimizer	Adam
Batch Size	32
Epochs	20

System Parameters

6 of 11

Dataset	Class	Precision	Recall	F1-Score	Accuracy
Tamil	Non-sarcastic	0.96	0.93	0.94	0.92
Tamil	Sarcastic	0.88	0.79	0.83	0.92
Malayalam	Non-sarcastic	0.97	0.98	0.97	0.96
Malayalam	Sarcastic	0.90	0.88	0.89	0.96

System Results

7 of 11

System Results (contd...)

8 of 11

Aspect	Observations	Proposed Future Work
Misclassification	Higher errors in Sarcastic class due to subtle or indirect sarcasm.	To use Transformer models like BERT for a better context understanding.
Class Imbalance	Non-sarcastic sentences dominate, leading to potential bias.	To apply data augmentation to balance classes.
Code-Mixed Nature	Challenges in handling grammar and syntax variability in Tami-English and Malayalam-English.	To use pre-trained embeddings tailored for code-mixed languages.
Overall Strengths	High accuracy for Non-sarcastic class; effective feature extraction using CNN + Bi-LSTM.	To expand the dataset by including more diverse sarcastic examples for a better generalization.

Error Analysis and Discussions

9 of 11

References

[1] B. R. Chakravarthi, S. N, B. B, N. K, T. Durairaj, R. Ponnusamy, P. K. Kumaresan, K. K. Ponnusamy, C. Rajkumar, Overview of sarcasm identification of dravidian languages in dravidiancodemix@fire-2024, in: Forum of Information Retrieval and Evaluation FIRE - 2024, DAIICT , Gandhinagar, 2024.

[2] N. Sripriya, T. Durairaj, K. Nandhini, B. Bharathi, K. K. Ponnusamy, C. Rajkumar, P. K. Kumaresan, R. Ponnusamy, C. Subalalitha, B. R. Chakravarthi, Findings of shared task on sarcasm identification in code-mixed dravidian languages, FIRE 2023 16 (2023) 22.

[3] B. R. Chakravarthi, N. Sripriya, B. Bharathi, K. Nandhini, S. C. Navaneethakrishnan, T. Durairaj, R. Ponnusamy, P. K. Kumaresan, K. K. Ponnusamy, C. Rajkumar, Overview of the shared task on sarcasm identification of dravidian languages (malayalam and tamil) in dravidiancodemix, in: Forum of Information Retrieval and Evaluation FIRE-2023, 2023.

[4] A. Rajadesingan, R. Zafarani, H. Liu, Sarcasm detection on twitter: A behavioral modeling approach, in: Proceedings of the eighth ACM international conference on web search and data mining, 2015, pp. 97–106.

[5] A. Joshi, V. Sharma, P. Bhattacharyya, Harnessing context incongruity for sarcasm detection, in: Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing (Volume 2: Short Papers), 2015, pp. 757–762.

10 of 11

References (contd...)

[6] A. Ghosh, T. Veale, Fracking sarcasm using neural network, in: Proceedings of the 7th workshop on computational approaches to subjectivity, sentiment and social media analysis, 2016, pp. 161–169.

[7] A. Ghosh, G. Li, T. Veale, P. Rosso, E. Shutova, J. Barnden, A. Reyes, Semeval-2015 task 11: Sentiment analysis of figurative language in twitter, in: Proceedings of the 9th international workshop on semantic evaluation (SemEval 2015), 2015, pp. 470–478.

[8] A. Joshi, P. Bhattacharyya, M. J. Carman, Automatic sarcasm detection: A survey, ACM Computing Surveys (CSUR) 50 (2017) 1–22.

[9] S. Poria, E. Cambria, D. Hazarika, N. Majumder, A. Zadeh, L.-P. Morency, Context-dependent sentiment analysis in user-generated videos, in: Proceedings of the 55th annual meeting of the association for computational linguistics (volume 1: Long papers), 2017, pp. 873–883.

[10] S. K. Bharti, K. S. Babu, S. K. Jena, Parsing-based sarcasm sentiment recognition in twitter data, in: Proceedings of the 2015 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining 2015, 2015, pp. 1373–1380.