Safeguarding Online Communications Using DistilRoBERTa for Detection of Terrorism and Offensive Chats
محتوى المقال
الملخص
People use social media for both good and distasteful purposes. When used with malicious intent, it raises
significant concerns as it involves the use of offensive language and hate speech that promote terrorism and
other negative behaviors. To create a safe, secure and pleasant environment, these communications must be
closely monitored to prevent severe problems, associated risks and other pertinent issues. With the help of
AI, specifically Large Language Models (LLM), we can quickly analyze text and speech to determine whether
the communications promote the dangers identified here above not to mention other toxic elements. For this
research, the LLM used is the DistilRoBERTa model from the Transformers library using Hugging Face. The
DistilRoBERTa model was trained on datasets consisting of terrorism-related conversations, offensive-related
conversations, and neutral conversations. These datasets were obtained from publicly available sources. The
results of the experimentation show that the model achieved 99% accuracy, precision, recall, F1 score, and
ROC curve. To improve the robustness of the model, it must be continuously fine-tuned to predict dynamic
communication behavior since real conversations are inaccessible due to restrictions. A drag-and-drop
interface is used to upload the files and get the categorical output, ensuring seamless and easy interaction.
الكلمات المفتاحية
Social Media، Offensive Language، Large Language Models، DistilRoBERTa model
التنزيلات
المراجع
A. Rajendran, V.S. Sahithi, C. Gupta, M. Yadav, S. Ahirrao, K. Kotecha, M. Gaikwad, A. Abraham, N. Ahmed, and S.M. Alhammad, "Detecting extremism on twitter during US capitol riot using deep learning techniques," *IEEE Access*, vol. 10, pp. 133052-133077, 2022.
S. Hussain and P. Mohideen, "Advanced Machine Learning Approach for Detection of Multilinguistic Terror Message to save human Lives," *Journal of Pharmaceutical Negative Results*, pp. 2528-2541, 2023.
O. Sharif, M.M. Hoque, A.S.M. Kayes, R. Nowrozy, and I.H. Sarker, "Detecting suspicious texts using machine learning techniques," *Applied Sciences*, vol. 10, no. 18, p. 6527, 2020.
A.B. Abhijith and P. Prithvi, "Automated Toxic Chat Synthesis, Reporting and Removing the Chat in Telegram Social Media Using Natural Language Processing Techniques," in *2024 Fourth International Conference on Advances in Electrical, Computing, Communication and Sustainable Technologies (ICAECT)*, Jan. 2024, pp. 1-7.
M. Gaikwad, S. Ahirrao, S. Phansalkar, and K. Kotecha, "Online extremism detection: A systematic literature review with emphasis on datasets, classification techniques, validation methods, and tools," *IEEE Access*, vol. 9, pp. 48364-48404, 2021.
A. Čepulionytė, J. Toldinas, and B. Lozinskis, "A multilayered preprocessing approach for recognition and classification of malicious social network messages," *Electronics*, vol. 12, no. 18, p. 3785, 2023.
S. Mussiraliyeva, B. Omarov, P. Yoo, and M. Bolatbek, "Applying machine learning techniques for religious extremism detection on online user contents," *Computers, Materials & Continua*, vol. 70, no. 1, pp. 915-934, 2022.
S. Pais et al., "Language-Independent Approaches to Detect Extremism and Collective Radicalisation Online," in *Proc. Conf.*, pp. 7–14, 2020.
F.E.T.H.I. Fkih, T.A.R.E.K. Moulahi, and A.B.D.U.L.A.T.I.F. Alabdulatif, "Machine learning model for offensive speech detection in online social networks slang content," *WSEAS Trans. Inf. Sci. Appl.*, vol. 20, pp. 7-15, 2023.
M. Fahim and S.S. Gokhale, "Identifying social media content supporting proud boys," in *2021 IEEE International Conference on Big Data (Big Data)*, Dec. 2021, pp. 2487-2495.
V. Shirsath, T. Wani, D. Bhakare, P. Lokhande, R. Chavda, and V. Shah, "ChatGuard: A Profanity Classification Approach for Safer Online Conversations," in *2023 1st DMIHER International Conference on Artificial Intelligence in Education and Industry 4.0 (IDICAIEI)*, Nov. 2023, vol. 1, pp. 1-5.
M. Gaikwad, S. Ahirrao, S. Phansalkar, K. Kotecha, and S. Rani, "Multi‐Ideology, Multiclass Online Extremism Dataset, and Its Evaluation Using Machine Learning," *Computational Intelligence and Neuroscience*, vol. 2023, no. 1, p. 4563145, 2023.
V. Nithyashree, B.N. Hiremath, L. Vanishree, A. Duvvuri, D.A. Madival, and G. Vidyashree, "Identification of Toxicity in Multimedia Messages for Controlling Cyberbullying on Social Media by Natural Language Processing," in *2022 International Conference on Distributed Computing, VLSI, Electrical Circuits and Robotics (DISCOVER)*, Oct. 2022, pp. 12-18.
A. Shevtsov, D. Antonakaki, I. Lamprou, P. Pratikakis, and S. Ioannidis, "BotArtist: Twitter bot detection Machine Learning model based on Twitter suspension," *arXiv preprint*, arXiv:2306.00037, 2023.
R. Alshalan and H. Al-Khalifa, "A deep learning approach for automatic hate speech detection in the Saudi twittersphere," *MDPI*, 2020. [Online]. Available: https://www.mdpi.com/2076-3417/10/23/8614. [Accessed: 20-Jun-2024].
W.A.A. NYANTUDRE, "Hate speech detection curated dataset," Kaggle, 2023. [Online]. Available: https://www.kaggle.com/datasets/waalbannyantudre/hate-speech-detection-curated-dataset. [Accessed: 13-Apr-2024].
F. Tribe, "How ISIS uses Twitter," Kaggle, 2019. [Online]. Available: https://www.kaggle.com/datasets/fifthtribe/how-isis-uses-twitter. [Accessed: 13-Apr-2024].
Marcoguerini, "Marcoguerini/Conan: A Repository with several curated datasets of counter-narratives to fight online hate speech," GitHub, 2022. [Online]. Available: https://github.com/marcoguerini/CONAN/tree/master. [Accessed: 13-Apr-2024].
Mpgii, "MPGIII/GAB-Hate: A research project involving trends of hate speech on Social Media Platform Gab," GitHub, 2020. [Online]. Available: https://github.com/mpgiii/gab-hate. [Accessed: 13-Apr-2024].
R. Kulkarni, "A million news headlines," Kaggle, 2022. [Online]. Available: https://www.kaggle.com/datasets/therohk/million-headlines. [Accessed: 13-Apr-2024].
Z. Usmani, "US mass shootings," Kaggle, 2022. [Online]. Available: https://www.kaggle.com/datasets/zusmani/us-mass-shootings-last-50years?select=US%2BMass%2BShootings%2BMay%2B24%2B2022.csv. [Accessed: 13-Apr-2024].
H. Hermessi, "Terrorism and jihadism speech detection," Kaggle, 2020. [Online]. Available: https://www.kaggle.com/datasets/haithemhermessi/terrorism-and-jihadist-speech-detection?select=train.csv. [Accessed: 13-Apr-2024].
تفاصيل المقال
هذا العمل مرخص بموجب Creative Commons Attribution-NonCommercial 4.0 International License.