DNN vs CNN for YouTube Shorts Abusive Word Classification
Abstract
Abusive words in YouTube Shorts are difficult to moderate because harmful expressions may appear in captions, comments, or embedded text within visual frames. This study compares the performance of a Deep Neural Network (DNN) and a Convolutional Neural Network (CNN) for abusive-word classification using text and image features from YouTube Shorts content. The dataset consisted of 1,013 image-text samples labeled into two binary classes: abusive and non-abusive. Textual features were derived from the text column representing OCR/caption information and transformed into 2,500-dimensional features, while image features were obtained from RGB images resized to 96 by 96 pixels. The dataset was divided into training, validation, and testing subsets consisting of 658, 202, and 153 samples, respectively. Both models were trained using dropout, L2 regularization, early stopping, and image augmentation. The results show that CNN outperformed DNN, achieving test accuracy of 0.8627 and macro F1-score of 0.8527, compared with DNN accuracy of 0.8301 and macro F1-score of 0.8195. These findings indicate that CNN provides more effective multimodal classification performance for detecting abusive words in YouTube Shorts content.
References
Baltrušaitis, T., Ahuja, C., & Morency, L. P. (2019). Multimodal machine learning: A survey and taxonomy. IEEE Transactions on Pattern Analysis and Machine Intelligence, 41(2), 423–443. https://doi.org/10.1109/TPAMI.2018.2798607
Bhandari, A., Shah, S. B., Thapa, S., Naseem, U., & Nasim, M. (2023). CrisisHateMM: Multimodal analysis of directed and undirected hate speech in text-embedded images from Russia-Ukraine conflict. In 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) (pp. 1994–2003). IEEE. https://doi.org/10.1109/CVPRW59228.2023.00193
Boishakhi, F. T., Shill, P. C., & Alam, M. G. R. (2021). Multi-modal hate speech detection using machine learning. In 2021 IEEE International Conference on Big Data (Big Data) (pp. 4496–4499). IEEE.
Chhabra, A., & Vishwakarma, D. K. (2023). A literature survey on multimodal and multilingual automatic hate speech identification. Multimedia Systems, 29, 1203–1230. https://doi.org/10.1007/s00530-023-01051-8
Das, M., Raj, R., Saha, P., Mathew, B., Gupta, M., & Mukherjee, A. (2023). HateMM: A multi-modal dataset for hate video classification. Proceedings of the International AAAI Conference on Web and Social Media, 17(1), 1014–1023. https://doi.org/10.1609/icwsm.v17i1.22209
El-Sayed, A., & Nasr, O. (2024). AAST-NLP at multimodal hate speech event detection 2024: A multimodal approach for classification of text-embedded images based on CLIP and BERT-based models. In Proceedings of the 7th Workshop on Challenges and Applications of Automated Extraction of Socio-political Events from Text (CASE 2024) (pp. 139–144). Association for Computational Linguistics. https://doi.org/10.18653/v1/2024.case-1.19
Ganguly, A., Bin Emran, A. N., Puspo, S. S. C., Raihan, M. N., Goswami, D., & Zampieri, M. (2024). MasonPerplexity at multimodal hate speech event detection 2024: Hate speech and target detection using transformer ensembles. In Proceedings of the 7th Workshop on Challenges and Applications of Automated Extraction of Socio-political Events from Text (CASE 2024) (pp. 125–131). Association for Computational Linguistics. https://doi.org/10.18653/v1/2024.case-1.17
Gomez, R., Gibert, J., Gomez, L., & Karatzas, D. (2020). Exploring hate speech detection in multimodal publications. In 2020 IEEE Winter Conference on Applications of Computer Vision (WACV) (pp. 1470–1478). IEEE. https://doi.org/10.1109/WACV45572.2020.9093414
Ibrohim, M. O., & Budi, I. (2018). A dataset and preliminaries study for abusive language detection in Indonesian social media. Procedia Computer Science, 135, 222–229. https://doi.org/10.1016/j.procs.2018.08.169
Ibrohim, M. O., & Budi, I. (2019). Multi-label hate speech and abusive language detection in Indonesian Twitter. In Proceedings of the Third Workshop on Abusive Language Online (pp. 46–57). Association for Computational Linguistics. https://doi.org/10.18653/v1/W19-3506
Ibrohim, M. O., & Budi, I. (2023). Hate speech and abusive language detection in Indonesian social media: Progress and challenges. Heliyon, 9(8), Article e18647. https://doi.org/10.1016/j.heliyon.2023.e18647
Janiesch, C., Zschech, P., & Heinrich, K. (2021). Machine learning and deep learning. Electronic Markets, 31, 685–695. https://doi.org/10.1007/s12525-021-00475-2
Kiela, D., Firooz, H., Mohan, A., Goswami, V., Singh, A., Ringshia, P., & Testuggine, D. (2020). The hateful memes challenge: Detecting hate speech in multimodal memes. In Advances in Neural Information Processing Systems, 33 (pp. 2611–2624).
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2012). ImageNet classification with deep convolutional neural networks. In Advances in Neural Information Processing Systems, 25.
Prabhu, R., & Seethalakshmi, V. (2025). A comprehensive framework for multi-modal hate speech detection in social media using deep learning. Scientific Reports, 15, Article 13020. https://doi.org/10.1038/s41598-025-94069-z
Salim, C. E. R., & Suhartono, D. (2021). Long short-term memory for hate speech and abusive language detection on Indonesian YouTube comment section. In 2021 the 11th International Workshop on Computer Science and Engineering (pp. 193–200). https://doi.org/10.18178/wcse.2021.06.029
Shang, L., Zhang, Y., Zha, Y., Chen, Y., Youn, C., & Wang, D. (2021). AOMD: An analogy-aware approach to offensive meme detection on social media. Information Processing & Management, 58(5), Article 102664. https://doi.org/10.1016/j.ipm.2021.102664
Suryawanshi, S., Chakravarthi, B. R., Arcan, M., & Buitelaar, P. (2020). Multimodal meme dataset (MultiOFF) for identifying offensive content in image and text. In Proceedings of the Second Workshop on Trolling, Aggression and Cyberbullying (pp. 32–41). European Language Resources Association.
Thapa, S., Shah, A., Jafri, F. A., Naseem, U., & Razzak, I. (2022). A multi-modal dataset for hate speech detection on social media: Case-study of Russia-Ukraine conflict. In Proceedings of the 5th Workshop on Challenges and Applications of Automated Extraction of Socio-political Events from Text (CASE) (pp. 1–6). Association for Computational Linguistics. https://doi.org/10.18653/v1/2022.case-1.1
Thapa, S., Jafri, F., Hürriyetoğlu, A., Vargas, F., Lee, R. K. W., & Naseem, U. (2023). Multimodal hate speech event detection—Shared task 4, CASE 2023. In Proceedings of the 6th Workshop on Challenges and Applications of Automated Extraction of Socio-political Events from Text (pp. 151–159). INCOMA Ltd.
Thapa, S., Rauniyar, K., Jafri, F., Veeramani, H., Jain, R., Jain, S., Vargas, F., Hürriyetoğlu, A., & Naseem, U. (2024). Extended multimodal hate speech event detection during Russia-Ukraine crisis—Shared task at CASE 2024. In Proceedings of the 7th Workshop on Challenges and Applications of Automated Extraction of Socio-political Events from Text (CASE 2024) (pp. 221–228). Association for Computational Linguistics. https://doi.org/10.18653/v1/2024.case-1.31
Zhang, C., Bengio, S., Hardt, M., Recht, B., & Vinyals, O. (2021). Understanding deep learning (still) requires rethinking generalization. Communications of the ACM, 64(3), 107–115. https://doi.org/10.1145/3446776
