QBERTox: A Quantum-Enhanced Explainable Model for Cyberbullying Detection in a Code-Mixed Language

Year 2025, Volume: 1 Issue: 2, 220 - 238, 28.07.2025

Akriti Singh , Ashok Yadav , Vrijendra Singh

https://doi.org/10.26650/d3ai.1729000

Abstract

The rise of social media has amplified online bullying, particularly in code-mixed languages such as Hinglish, where detecting harmful content remains challenging due to linguistic complexity and nuanced expressions. We propose QBERTox, a novel model for classifying Hinglish social media text as bullying or non-bullying, integrating a quantum-inspired layer with explainable AI techniques. Built on a fine tuned BERT architecture, QBERTox incorporates cyberbully-specific features, including toxicity scores from the Detoxify model and sentiment analysis, to capture semantic nuances. The quantum-inspired layer, implemented as a variational quantum circuit with 8 qubits, enhances feature entanglement for improved detection of complex linguistic patterns, outperforming classical BERT by 2.3% in the F1-score on a Hinglish dataset. Our dataset, comprising 6,432 annotated Hinglish tweets (46% non-bullying, 54% bullying). QBERTox achieves 85% accuracy and 0.85 F1-score and surpasses baselines. Explainability is ensured through LIME and SHAP, providing interpretable feature importance for bullying predictions. QBERTox offers a scalable, trustworthy solution for combating cyberbullying in multilingual contexts, with guidelines for platform integration and moderator training.

CCS Concepts: • Do Not Use This Code → Generate the Correct Terms for Your Paper; Generate the Correct Terms for Your Paper; Generate the Correct Terms for Your Paper; Generate the Correct Terms for Your Paper.

Keywords

Cyberbullying , Deep Learning , Explainability , Detoxify , Quantum

References

• Himakshi Baruah, PragaYa Dashora, and Arti Parmar. 2017. Impact of cYberbullYing on psYchological health of adolescents. International Journal of Humanities and Social Sciences (IJHSS) 6, 4 (2017), 137-144. google scholar
• Siriporn Santre. 2023. CYberbullYing in adolescents: A literature review. International Journal of Adolescent Medicine and Health 35, 1 (2023), 1-7. google scholar
• Varun Jain, Vishant Kumar, Vivek Pal, and Dinesh Kumar Vishwakarma. 2021. Detection of cYberbullYing on social media using machine learning. In Proceedings of the 2021 5th International Conference on Computing Methodologies and Communication (ICCMC). IEEE, 1091-1096 (https://doi.org/l0.1109/ICCMC51019.2021.9418254) google scholar
• Maral Dadvar and Kai Eckert. 2018. CYberbullYing detection in social networks using deep learning based models; a reproducibil-itY studY. arXiv preprint arXiv:1812.08046. google scholar
• Celestine Iwendi, Gautam Srivastava, Suleman Khan, and Praveen Kumar ReddY Maddikunta. 2023. CYberbullYing detection solutions based on deep learning architectures. Multimedia SYstems 29, 3 (2023), 1839-1852. google scholar
• Munif Alotaibi, Bandar Alotaibi, and Abdul Razaque. 2021. A multichannel deep learning framework for cYberbullYing detection on social media. Electronics 10, 21 (2021), 2664. google scholar
• Arafat Rahman, Afia Farjana, Hafizur Rahman ChowdhurY, et al. 2024. Multilingual cYberbullYing classification for social platforms. In Proceedings of the 2024 6th International Conference on Electrical Engineering and Information & Communication TechnologY (ICEEICT). IEEE, 904-909. google scholar
• Saurav Kumar, MrinmoY Mondal, Tanuja Dutta, and Thoudam Doren Singh. 2024. CYberbullYing detection in Hinglish comments from social media using machine learning techniques. Multimedia Tools and Applications, 1-22. google scholar
• Krishanu MaitY, Raghav Jain, Prince Jha, Sriparna Saha, and Pushpak BhattacharYYa. 2023. GenEx: A commonsense-aware unified generative framework for explainable cYberbullYing detection. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing (EMNLP). 16632-16645. google scholar
• Mohammad Zia Ur Rehman, SomYa Mehta, Kuldeep Singh, Kunal Kaushik, and Nagendra Kumar. 2023. User-aware multilingual abusive content detection in social media. Information Processing & Management 60, 5 (2023), 103450. google scholar
• Prashant Kapil, Gitanjali Kumari, Asif Ekbal, Santanu Pal, Arindam Chatterjee, and BN Vinutha. 2023. Hhld: Hateful posts identi-fication in Hindi language leveraging multi-task learning. IEEE Access (2023). google scholar
• Hind Saleh, Areej Alhothali, and Kawthar Moria. 2023. Detection of hate speech using BERT and hate speech word embedding with deep model. Applied Artificial Intelligence 37, 1 (2023), 2166719. google scholar
• Seffi Cohen, Dan Presil, Or Katz, Ofir Arbili, Shvat Messica, and Lior Rokach. 2023. Enhancing social network hate detection using back translation and GPT-3 augmentations during training and test-time. Information Fusion 99 (2023), 101887. google scholar
• Purbani Kar and Swapan Debbarma. 2023. Sentimental analYsis & hate speech detection on English and German text collected from social media platforms using optimal feature extraction and hYbrid diagonal gated recurrent neural network. Engineering Applications of Artificial Intelligence 126 (2023), 107143. google scholar
• Shivang Agarwal, Ankur Sonawane, and C Ravindranath ChowdarY. 2023. Accelerating automatic hate speech detection using parallelized ensemble learning models. Expert SYstems with Applications 230 (2023), 120564. google scholar
• Krishanu MaitY, Shaubhik BhattacharYa, Sriparna Saha, and Manjeevan Seera. 2023. A deep learning framework for the detection of MalaY hate speech. IEEE Access (2023). google scholar
• Muhammad Shahid Khan, Muhammad Shahid Iqbal Malik, and Aamer Nadeem. 2024. Detection of violence incitation expressions in Urdu tweets using convolutional neural network. Expert SYstems with Applications 245 (2024), 123174. google scholar
• Sunil SaumYa, Abhinav Kumar, and JYoti Prakash Singh. 2024. Filtering offensive language from multilingual social media contents: A deep learning approach. Engineering Applications of Artificial Intelligence 133 (2024), 108159. google scholar
• Esshaan Mahajan, Hemaank Mahajan, and SanjaY Kumar. 2024. EnsmulhatecYb: Multilingual hate speech and cYberbullY detec-tion in online social media. Expert SYstems with Applications 236 (2024), 121228. google scholar
• Arpan Nandi, Kamal Sarkar, Arjun Mallick, and Arkadeep De. 2024. Combining multiple pre-trained models for hate speech detection in Bengali, Marathi, and Hindi. Multimedia Tools and Applications, 1-25. google scholar
• Cendra DevaYana Putra and Hei-Chia Wang. 2024. Advanced BERT-CNN for hate speech detection. Procedia Computer Science 234 (2024), 239-246. google scholar
• Simrat Kaur, Sarbjeet Singh, and Sakshi Kaushal. 2024. Deep learning-based approaches for abusive content detection and classification for multi-class Online user-generated data. International Journal of Cognitive Computing in Engineering 5 (2024), 104-122. google scholar
• Atif Khan, Abrar Ahmed, Salman Jan, Muhammad Bilal, and Megat F Zuhairi. 2024. Abusive language detection in Urdu text: Leveraging deep learning and attention mechanism. IEEE Access (2024). google scholar
• Yasmine M İbrahim, Reem Essameldin, and Saad M Saad. 2024. Social media forensics: An adaptive CYberbullYİng-related hate speech detection approach based on neural networks with uncertaintY. IEEE Access (2024). google scholar
• Deepanshu VijaY, AditYa Bohra, VinaY Singh, SYed Sarfaraz Akhtar, and Manish Shrivastava. 2018. A dataset for detecting ironY in Hindi-English code-mixed social media text. EMSASW@ESWC 2111 (2018), 38-46. google scholar
• Aditya Bohra, Deepanshu Vijay, VinaY Singh, SYed Sarfaraz Akhtar, and Manish Shrivastava. 2018. A dataset of Hindi-English code-mixed social media text for hate speech detection. İn Proceedings of the Second Workshop on Computational Modeling of People’s Opinions, PersonalitY, and Emotions in Social Media. 36-41. google scholar
• Maliha Jahan, Istiak Ahamed, Md RaYanuzzaman Bishwas, and Swakkhar Shatabda. 2019. Abusive comments detection in Bangla-English code-mixed and transliterated text. İn Proceedings of the 2019 2nd International Conference on Innovation in Engineering and TechnologY (ICIET). IEEE, 1-6. google scholar
• Ninad Mehendale, Karan Shah, Chaitanya Phadtare, and Keval Rajpara. 2022. Cyber bullying detection for Hindi-English language using machine learning. SSRN 4116143, (2022). google scholar
• Ankush Khandelwal, Sahil Swami, Syed Sarfaraz Akhtar, and Manish Shrivastava. 2018. Gender prediction in English-Hindi code-mixed social media content: Corpus and baseline system. Computacion y Sistemas 22, 4 (2018), 1241-1247. google scholar
• K Sreelakshmi, B Premjith, and KP Soman. 2020. Detection of hate speech text in Hindi-English code-mixed data. Procedia Computer Science 171 (2020), 737-744. google scholar
• Sanket Sonu. 2022. IdentifYing emotions for code mixed Hindi-English tweets. PhD thesis. National College of Ireland, Dublin. google scholar
• Tulika Ranjan, Anish Singh, Rina Kumari, Sujata Swain, Anjan BandYopadhYaY, and AjaYa Kumar Parida. 2023. Multilingual code-mixed sentiment analYsis in hate speech. Scalable Computing: Practice and Experience 24, 4 (2023), 873-882. google scholar
• Kirti Kumari and JYoti Prakash Singh. 2022. Machine learning approach for hate speech and offensive content identification in English and Indo-ArYan code-mixed languages. In FIRE (Working Notes). 575-583. google scholar
• Sweta Agrawal and Amit Awekar. 2018. Deep learning for detecting cYberbullYing across multiple social media platforms. In European Conference on Information Retrieval (ECIR). Springer, 141-153. google scholar
• SatYajit Kamble and AditYa Joshi. 2018. Hate speech detection from code-mixed Hindi-English tweets using deep learning models. arXiv preprint arXiv:1811.05145. google scholar
• T Tulasi Sasidhar, B Premjith, and KP Soman. 2020. Emotion detection in Hinglish (Hindi+English) code-mixed social media text. Procedia Computer Science 171 (2020), 1346-1352. google scholar
• Kirti Kumari, JYoti Prakash Singh, Yogesh Kumar Dwivedi, and Nripendra Pratap Rana. 2021. Bilingual cYber-aggression detection on social media using LSTM autoencoder. Soft Computing 25 (2021), 8999-9012. google scholar
• Tulika Saha, Apoorva UpadhYaYa, Sriparna Saha, and Pushpak BhattacharYYa. 2021. A multitask multimodal ensemble model for sentiment- and emotion-aided tweet act classification. IEEE Transactions on Computational Social SYstems 9, 2 (2021), 508-517. google scholar
• Md Saroar Jahan, Mourad Oussalah, Jhuma Kabir Mim, and Mominul Islam. 2021. Offensive language identification using Hindi-English code-mixed tweets, and code-mixed data augmentation. In Working Notes of FIRE 2021 - Forum for Information Retrieval Evaluation. RWTH Aachen UniversitY, Gandhinagar, India, Dec 13-17, 2021. google scholar
• Anshul Wadhawan and Akshita Aggarwal. 2021. Towards emotion recognition in Hindi-English code-mixed data: A transformer based approach. arXiv preprint arXiv:2102.09943. google scholar
• Krishanu MaitY and Sriparna Saha. 2021. BERT-capsule model for cYberbullYing detection in code-mixed Indian languages. In Proceedings of the International Conference on Applications of Natural Language to Information SYstems. Springer, 147-155. google scholar
• Krishanu MaitY and Sriparna Saha. 2021. A multi-task model for sentiment aided cYberbullYing detection in code-mixed Indian languages. In Neural Information Processing: 28th International Conference (ICONIP 2021). Springer, 440-451. google scholar
• SaYanta Paul and Sriparna Saha. 2022. CYberBERT: BERT for cYberbullYing identification. Multimedia SYstems 28, 6 (2022), 18971904. google scholar
• Krishanu MaitY, Sriparna Saha, and Pushpak BhattacharYYa. 2022. CYberbullYİng detection in code-mixed languages: Dataset and techniques. İn 2022 26th International Conference on Pattern Recognition (ICPR). IEEE, 1692-1698. google scholar
• Abhishek Chopra, Deepak Kumar Sharma, Aashna Jha, and Uttam Ghosh. 2023. A framework for online hate speech detection on code-mixed Hindi-English text and Hindi text in Devanagari. ACM Transactions on Asian and Low-Resource Language Information Processing 22, 5 (2023), 1-21. google scholar
• Hiren Madhu, ShreY Satapara, Sandip Modha, Thomas Mandl, and Prasenjit Majumder. 2023. Detecting offensive speech in conversational code-mixed dialogue on social media: A contextual dataset and benchmark experiments. Expert SYstems with Applications 215 (2023), 119342. google scholar
• Shubham Das and TanYa Singh. 2023. Sentiment recognition of Hinglish code mixed data using deep learning models-based approach. In 2023 13th International Conference on Cloud Computing, Data Science & Engineering (Confluence). IEEE, 265-269. google scholar
• Soumitra Ghosh, Amit PriYankar, Asif Ekbal, and Pushpak BhattacharYYa. 2023. Multitasking of sentiment detection and emotion recognition in code-mixed Hinglish data. Knowledge-Based SYstems 260 (2023), 110182. google scholar
• Krishanu MaitY, Prince Jha, Raghav Jain, Sriparna Saha, and Pushpak BhattacharYYa. 2023. “Explain ThYself BullY”: Senti-ment-aided CYberbullYİng detection with explanation. In International Conference on Document AnalYsis and Recognition (ICDAR). Springer, 132-148. google scholar
• Krishanu MaitY, Raghav Jain, Prince Jha, and Sriparna Saha. 2023. Explainable cYberbullYing detection in Hinglish: A generative approach. IEEE Transactions on Computational Social SYstems (2023). google scholar
• Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. BERT: Pre-training of deep bidirectional transformers for language understanding. In Proceedings of NAACL-HLT 2019. Minneapolis, MN, USA, 1-12. google scholar