Research Article Klasifikasi Polaritas Meme Berbasis Deep Learning dan Clustering dengan Penerapan Deteksi Teks Sarkasme
Kata Kunci:
Natural Language Processing, Deep Learning, Sarcasm Detection, Sentiment AnalysisAbstrak
Meme merupakan konten internet populer yang cepat menyebar di media sosial dan sering digunakan untuk mengekspresikan ide, kritik, atau ketertarikan. Namun, interpretasi meme dapat bervariasi sehingga menimbulkan tantangan dalam analisis sentimen, karena satu meme bisa dipandang positif atau negatif oleh individu berbeda. Untuk mengatasi hal tersebut, diperlukan sistem otomatis yang mampu memprediksi polaritas sentimen secara konsisten. Meme bersifat multimodal karena menggabungkan komponen visual dan teks, sehingga cocok untuk penelitian analisis sentimen berbasis multimodal. Penelitian ini mengusulkan model deep learning gabungan BERT dan DenseNet121, yang mengintegrasikan teks, gambar, serta fitur cluster berbasis face encoding. Untuk meningkatkan pemahaman konteks teks, BERT digunakan dengan pelatihan deteksi sarkasme. Dataset yang digunakan adalah SemEval 2020 Task 8: Memotion Analysis, yang menyediakan anotasi lengkap tentang sentimen dan sarkasme pada meme. Hasil penelitian menunjukkan bahwa model gabungan dengan deteksi sarkasme mencapai Macro-F1 sebesar 0.3047 dan akurasi 0.3738, melampaui baseline resmi (0.2176) dengan peningkatan sekitar 40%. Model ini juga lebih efektif dalam mendeteksi sentimen positif dan netral serta mengurangi kesalahan negatif palsu akibat sarkasme. Hal ini membuktikan bahwa integrasi deteksi sarkasme memperkuat performa klasifikasi sentimen pada meme.
Kata kunci— klasifikasi sentimen, meme, deep learning, clustering, deteksi sarkasme, memotion
Referensi
S. Tabatabaei dan E. A. Ivanova, “The role of memes on emotional contagion”, Elementary Education Online, vol. 20, no. 5, hlm. 6028–6036, 2023. [Daring]. Tersedia: https://ilkogretim- online.org/index.php/pub/article/view/4690
S. Kholmatov, “Multimodal sentiment analysis: A study on emotion understanding and classification by integrating text and images,” Beijing Institute of Technology, 2023. [Daring]. Tersedia: https://doi.org/10.5281/zenodo.13909963
C. Sharma, D. Bhageria, W. Scott et al., “SemEval-2020 Task 8: Memotion analysis – The visuo-lingual metaphor!” dalam Proceedings of the 12th International Semantic Evaluation Conference (SemEval-2020), 2020. [Daring]. Tersedia: https://aclanthology.org/2020.semeval-1.99.pdf
Y. Guo, J. Huang, Y. Dong et al., “Guoym at SemEval-2020 Task 8: Ensemble-based classification of visuo-lingual metaphor in memes,” dalam Proceedings of the Fourteenth Workshop on Semantic Evaluation, Barcelona (online): International Committee for Computational Linguistics, 2020, hlm. 1120–1125. [Daring]. Tersedia: https://aclanthology.org/2020.semeval-1.148
X. Guo, J. Ma, dan A. Zubiaga, “Cluster-based deep ensemble learning for emotion classification in internet memes,” arXiv preprint arXiv:2302.08343, 2023. [Daring]. Tersedia: https://doi.org/10.48550/arXiv.2302.08343
W. Chen, F. Lin, G. Li, dan B. Liu, “A survey of automatic sarcasm detection: Fundamental theories, formulation, datasets, detection methods, and opportunities,” Neurocomputing, vol. 578, hlm. 127428, 2024. [Daring]. Tersedia: https://doi.org/10.1016/j.neucom.2024.127428
A. Bhat dan A. Chauhan, “Multimodal sarcasm detection: A survey,” dalam 2022 IEEE Delhi Section Conference (DELCON), 2022, hlm. 1–7.
A. Baruah, K. Das, F. Barbhuiya, dan K. Dey, “Context-aware sarcasm detection using BERT,” dalam Proceedings of the Second Workshop on Figurative Language Processing, B. B. Klebanov et al., Eds. Online: Association for Computational Linguistics, Jul. 2020, hlm. 83–87. [Daring]. Tersedia: https://aclanthology.org/2020.figlang-1.12/
P. Parameswaran, A. Trotman, V. Liesaputra, dan D. Eyers, “BERT’s the word: Sarcasm target detection using BERT,” dalam Proceedings of the 19th Annual Workshop of the Australasian Language Technology Association, A. Rahimi et al., Eds. Online: ALTA, Des. 2021, hlm. 185–191. [Daring]. Tersedia: https://aclanthology.org/2021.alta-1.21/
G. Huang, Z. Liu, L. Van Der Maaten, dan K. Weinberger, “Densely connected convolutional networks,” dalam Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017, hlm. 4700–4708.
S. Pouyanfar et al., “A survey on deep learning: Algorithms, techniques, and applications,” ACM Computing Surveys, vol. 51, no. 5, Art. 92, hlm. 1–36, Sep. 2019. [Daring]. Tersedia: https://doi.org/10.1145/3234150
J. Devlin, M. Chang, K. Lee, dan K. Toutanova, “BERT: Pre-training of deep bidirectional transformers for language understanding,” arXiv preprint, arXiv:1810.04805, 2018. [Daring]. Tersedia: https://doi.org/10.48550/arXiv.1810.04805
F. Schroff, D. Kalenichenko, dan J. Philbin, “FaceNet: A unified embedding for face recognition and clustering,” dalam Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Boston, MA, USA, 2015, hlm. 815–823.
D. Sandberg, "inception_resnet_v1.py," GitHub repository, 2018. [Daring]. Tersedia: https://github.com/davidsandberg/facenet/blob/master/src/models/ince ption_resnet_v1.py
D. Müllner, “Modern hierarchical, agglomerative clustering algorithms,” 2011. [Daring]. Tersedia: https://arxiv.org/abs/1109.2378
S. I. Serengil and A. Ozpinar, "LightFace: A Hybrid Deep Face Recognition Framework," dalam Proceedings of the 2020 Innovations in Intelligent Systems and Applications Conference (ASYU), Istanbul, Turki, 2020, hlm. 23–27, doi: 10.1109/ASYU50717.2020.9259802.
T.-Y. Lin, P. Goyal, R. Girshick, K. He, dan P. Dollár, “Focal Loss untuk Deteksi Objek Padat,” dalam Proceedings of the 2017 IEEE International Conference on Computer Vision (ICCV), Venice, Italia, 2017, hlm. 2999– 3007. [Daring]. Tersedia: https://doi.org/10.1109/ICCV.2017.324
S. Barak dan T. Mokfi, “Evaluation and selection of clustering methods using a hybrid group MCDM,” Expert Systems with Applications, vol. 138, hlm. 112817, 2019.
##submission.downloads##
Diterbitkan
Versi
- 2026-03-12 (2)
- 2026-03-12 (1)
