Conference Paper2023Cited by 3

Aambela at BLP-2023 Task 1: Focus on [UNK] tokens: Analyzing Violence Inciting Bengali Text with Adding Dataset Specific New Word Tokens

Md Fahim

Proceedings of the 1st International Workshop on Bangla Language Processing (BLP-2023)

Association for' Computational Linguistics, pp. 201-207

  • Peace, Justice and Strong Institutions — Relevance: 53%
DOI: 10.18653/v1/2023.banglalp-1.24Open AccessURL

Abstract

The BLP-2023 Task 1 aims to develop a Natural Language Inference system tailored for detecting and analyzing threats from Bangla YouTube comments. Bangla language models like BanglaBERT have demonstrated remarkable performance in various Bangla natural language processing tasks across different domains. We utilized BanglaBERT for the violence detection task, employing three different classification heads. As BanglaBERT’s vocabulary lacks certain crucial words, our model incorporates some of them as new special tokens, based on their frequency in the dataset, and their embeddings are learned during training. The model achieved the 2nd position on the leaderboard, boasting an impressive macro-F1 Score of 76.04% on the official test set. With the addition of new tokens, we achieved a 76.90% macro-F1 score, surpassing the top score (76.044%) on the test set.

Keywords

  • violence detection
  • BanglaBERT
  • YouTube comments
  • Bangla text
  • natural language inference
  • threat detection

CCDS Authors

MD Fahim
Md Fahim

MD Fahim

  • Research Assistant, CCDS

Published at

ACL, EMNLP, ECAI, WACV, LREC, ICPR, BEA, BLP, Information Sciences, Smart Agricultural Technology

Research Interests

Machine Learning, Natural Language Processing

References

  1. 1.Jacob Devlin, Ming-Wei Chang, Kenton Lee, Kristina Toutanova. (2019). , 4171–4186[10.18653/v1/n19-1423]
  2. 2.Hanqing Zhang, Haolin Song, Shaoyu Li, Ming Zhou, Dawei Song. (2023). A Survey of Controllable Text Generation Using Transformer-based Pre-trained Language Models. ACM Computing Surveys, 56(3), 1–37[10.1145/3617680]
  3. 3.Abhik Bhattacharjee, Tahmid Hasan, Wasi Ahmad, Kazi Samin Mubasshir, Md Saiful Islam, Anindya Iqbal, M. Sohel Rahman, Rifat Shahriyar. (2022). BanglaBERT: Language Model Pretraining and Benchmarks for Low-Resource Language Understanding Evaluation in Bangla. Findings of the Association for Computational Linguistics: NAACL 2022, 1318–1327[10.18653/v1/2022.findings-naacl.98]
  4. 4.M. F. Mridha, Md. Anwar Hussen Wadud, Md. Abdul Hamid, Muhammad Mostafa Monowar, M. Abdullah-Al-Wadud, Atif Alamri. (2021). L-Boost: Identifying Offensive Texts From Social Media Post in Bengali. IEEE Access, 9, 164681–164699[10.1109/access.2021.3134154]
  5. 5.David Yenicelik, Florian Schmidt, Yannic Kilcher. (2020). How does BERT capture semantics? A closer look at polysemous words. , 156–162[10.18653/v1/2020.blackboxnlp-1.15]
  6. 6.Sourav Saha, Jahedul Alam Junaed, Maryam Saleki, Arnab Sen Sharma, Mohammad Rashidujjaman Rifat, Mohamed Rahouti, Syed Ishtiaque Ahmed, Nabeel Mohammed, Mohammad Ruhul Amin. (2023). Vio-Lens: A Novel Dataset of Annotated Social Network Posts Leading to Different Forms of Communal Violence and its Evaluation. , 72–84[10.18653/v1/2023.banglalp-1.9]
  7. 7.Anmol Nayak, Hariprasad Timmapathini, Karthikeyan Ponnalagu, Vijendran Gopalan Venkoparao. (2020). Domain adaptation challenges of BERT in tokenization and sub-word representations of Out-of-Vocabulary words. , 1–5[10.18653/v1/2020.insights-1.1]
  8. 8.Sourav Saha, Jahedul Alam Junaed, Maryam Saleki, Mohamed Rahouti, Nabeel Mohammed, Mohammad Ruhul Amin. (2023). BLP-2023 Task 1: Violence Inciting Text Detection (VITD). , 255–265[10.18653/v1/2023.banglalp-1.33]

Cited By

  1. 1.Sourav Saha, Jahedul Alam Junaed, Maryam Saleki, Mohamed Rahouti, Nabeel Mohammed, Mohammad Ruhul Amin. (2023). BLP-2023 Task 1: Violence Inciting Text Detection (VITD). Workshop on Bangla Language Processing (BLP), 255–265[10.18653/v1/2023.banglalp-1.33]
  2. 2.Prosenjit Chakraborty, Md Mynoddin, Md Saharab Hossain, S. M. Golam Sarwar Apu, U Mong Sing Marma Biplob. (2026). Stop the spark detecting violence instigating Bangla text using ViolenceBanglaBERT. Discover Artificial Intelligence, 6(1)[10.1007/s44163-026-01696-0]
  3. 3.MD. Tahidul Islam, Nakib Aman, Md. Abu Johab, Md. Sadikul Islam, Md. Roton Ahmed. (2025). Distinguishing Between Saint(Sadhu) and Common(Cholit): A Bangla BERT Approach to Bengali Language Classification. International Conference on Computer and Information Technology (ICCIT), 2157–2162[10.1109/iccit68739.2025.11490255]
← All publications