{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,9]],"date-time":"2026-10-09T18:30:51Z","timestamp":1791570651109,"version":"4.3.4"},"reference-count":47,"publisher":"MDPI AG","issue":"6","license":[{"start":{"date-parts":[[2025,6,1]],"date-time":"2025-06-01T00:00:00Z","timestamp":1748736000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Mexican Government","award":["A1-S-47854"],"award-info":[{"award-number":["A1-S-47854"]}]},{"name":"Mexican Government","award":["20254236"],"award-info":[{"award-number":["20254236"]}]},{"name":"Mexican Government","award":["20253468"],"award-info":[{"award-number":["20253468"]}]},{"name":"Mexican Government","award":["20254341"],"award-info":[{"award-number":["20254341"]}]},{"name":"Secretar\u00eda de Investigaci\u00f3n y Posgrado of the Instituto Polit\u00e9cnico Nacional, Mexico","award":["A1-S-47854"],"award-info":[{"award-number":["A1-S-47854"]}]},{"name":"Secretar\u00eda de Investigaci\u00f3n y Posgrado of the Instituto Polit\u00e9cnico Nacional, Mexico","award":["20254236"],"award-info":[{"award-number":["20254236"]}]},{"name":"Secretar\u00eda de Investigaci\u00f3n y Posgrado of the Instituto Polit\u00e9cnico Nacional, Mexico","award":["20253468"],"award-info":[{"award-number":["20253468"]}]},{"name":"Secretar\u00eda de Investigaci\u00f3n y Posgrado of the Instituto Polit\u00e9cnico Nacional, Mexico","award":["20254341"],"award-info":[{"award-number":["20254341"]}]},{"name":"CONACYT","award":["A1-S-47854"],"award-info":[{"award-number":["A1-S-47854"]}]},{"name":"CONACYT","award":["20254236"],"award-info":[{"award-number":["20254236"]}]},{"name":"CONACYT","award":["20253468"],"award-info":[{"award-number":["20253468"]}]},{"name":"CONACYT","award":["20254341"],"award-info":[{"award-number":["20254341"]}]},{"name":"Microsoft","award":["A1-S-47854"],"award-info":[{"award-number":["A1-S-47854"]}]},{"name":"Microsoft","award":["20254236"],"award-info":[{"award-number":["20254236"]}]},{"name":"Microsoft","award":["20253468"],"award-info":[{"award-number":["20253468"]}]},{"name":"Microsoft","award":["20254341"],"award-info":[{"award-number":["20254341"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Algorithms"],"abstract":"<jats:p>Hate speech is a complex topic that transcends language, culture, and even social spheres. Recently, the spread of hate speech on social media sites like Facebook has added a new layer of complexity to the issue of online safety and content moderation. This study seeks to minimize this problem by developing an Arabic script-based tool for automatically detecting hate speech in Roman Urdu, an informal script used most commonly for South Asian digital communications. Roman Urdu is relatively complex as there are no standardized spellings, leading to syntactic variations, which increases the difficulty of hate speech detection. To tackle this problem, we adopt a holistic strategy using a combination of six machine learning (ML) and four Deep Learning (DL) models, a dataset from Facebook comments, which was preprocessed (tokenization, stopwords removal, etc.), and text vectorization (TF-IDF, word embeddings). The ML algorithms used in this study are LR, SVM, RF, NB, KNN, and GBM. We also use deep learning architectures like CNN, RNN, LSTM, and GRU to increase the accuracy of the classification further. It is proven by the experimental results that deep learning models outperform the traditional ML approaches by a significant margin, with CNN and LSTM achieving accuracies of 95.1% and 96.2%, respectively. As far as we are aware, this is the first work that investigates QLoRA for fine-tuning large models for the task of offensive language detection in Roman Urdu.<\/jats:p>","DOI":"10.3390\/a18060331","type":"journal-article","created":{"date-parts":[[2025,6,2]],"date-time":"2025-06-02T08:34:13Z","timestamp":1748853253000},"page":"331","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":16,"title":["Machine Learning- and Deep Learning-Based Multi-Model System for Hate Speech Detection on Facebook"],"prefix":"10.3390","volume":"18","author":[{"given":"Amna","family":"Naseeb","sequence":"first","affiliation":[{"name":"Instituto Polit\u00e9cnico Nacional (IPN), Centro de Investigaci\u00f3n en Computaci\u00f3n (CIC), Av. Juan de Dios Batiz, s\/n, Mexico City 07320, Mexico"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Muhammad","family":"Zain","sequence":"additional","affiliation":[{"name":"Instituto Polit\u00e9cnico Nacional (IPN), Centro de Investigaci\u00f3n en Computaci\u00f3n (CIC), Av. Juan de Dios Batiz, s\/n, Mexico City 07320, Mexico"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2877-8135","authenticated-orcid":false,"given":"Nisar","family":"Hussain","sequence":"additional","affiliation":[{"name":"Instituto Polit\u00e9cnico Nacional (IPN), Centro de Investigaci\u00f3n en Computaci\u00f3n (CIC), Av. Juan de Dios Batiz, s\/n, Mexico City 07320, Mexico"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7536-6969","authenticated-orcid":false,"given":"Amna","family":"Qasim","sequence":"additional","affiliation":[{"name":"Instituto Polit\u00e9cnico Nacional (IPN), Centro de Investigaci\u00f3n en Computaci\u00f3n (CIC), Av. Juan de Dios Batiz, s\/n, Mexico City 07320, Mexico"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fiaz","family":"Ahmad","sequence":"additional","affiliation":[{"name":"Instituto Polit\u00e9cnico Nacional (IPN), Centro de Investigaci\u00f3n en Computaci\u00f3n (CIC), Av. Juan de Dios Batiz, s\/n, Mexico City 07320, Mexico"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3901-3522","authenticated-orcid":false,"given":"Grigori","family":"Sidorov","sequence":"additional","affiliation":[{"name":"Instituto Polit\u00e9cnico Nacional (IPN), Centro de Investigaci\u00f3n en Computaci\u00f3n (CIC), Av. Juan de Dios Batiz, s\/n, Mexico City 07320, Mexico"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7845-9039","authenticated-orcid":false,"given":"Alexander","family":"Gelbukh","sequence":"additional","affiliation":[{"name":"Instituto Polit\u00e9cnico Nacional (IPN), Centro de Investigaci\u00f3n en Computaci\u00f3n (CIC), Av. Juan de Dios Batiz, s\/n, Mexico City 07320, Mexico"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2025,6,1]]},"reference":[{"key":"ref_1","first-page":"121507","article-title":"Enhancing multilingual hate speech detection: From language-specific insights to cross-linguistic integration","volume":"12","author":"Hashmi","year":"2024","journal-title":"IEEE Trans. Comput. Intell. AI Games"},{"key":"ref_2","doi-asserted-by":"crossref","unstructured":"Ashiq, W., Kanwal, S., Rafique, A., and Waqas, M. (2024). Roman Urdu hate speech detection using hybrid machine learning models and hyperparameter optimization. Sci. Rep., 14.","DOI":"10.1038\/s41598-024-79106-7"},{"key":"ref_3","doi-asserted-by":"crossref","unstructured":"Sujatha, R., Chatterjee, J.M., Pathy, B., and Hu, Y.C. (2025). Automatic emotion recognition using deep neural networks. Multimed. Tools Appl.","DOI":"10.1007\/s11042-024-20590-4"},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3414524","article-title":"Hate speech detection in Roman Urdu","volume":"20","author":"Khan","year":"2021","journal-title":"ACM Trans. Asian Low-Resour. Lang. Inf. Process."},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Aziz, S., Sarfraz, M.S., Usman, M., Aftab, M.U., and Rauf, H.T. (2023). Geo-spatial mapping of hate speech prediction in Roman Urdu. Mathematics, 11.","DOI":"10.3390\/math11040969"},{"key":"ref_6","first-page":"10","article-title":"Detection of hate speech in social networks: A survey on multilingual corpus","volume":"Volume 10","year":"2019","journal-title":"Proceedings of the 6th International Conference on Computer Science and Information Technology"},{"key":"ref_7","unstructured":"Madukwe, J.K. (2025). The Detection of Online Textual Hate Speech. [Ph.D. Thesis, Open Access Te Herenga Waka\u2014Victoria University of Wellington]."},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Hussain, N., Qasim, A., Mehak, G., Kolesnikova, O., Gelbukh, A., and Sidorov, G. (2025). Hybrid machine learning and Deep Learning Approaches for Insult Detection in Roman Urdu Text. AI, 6.","DOI":"10.3390\/ai6020033"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3711710","article-title":"Hate Speech Detection Research in South Asian Languages: A Survey of Tasks, Datasets and Methods","volume":"24","author":"Sharma","year":"2025","journal-title":"ACM Trans. Asian Low-Resour. Lang. Inf. Process."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"114762","DOI":"10.1016\/j.eswa.2021.114762","article-title":"A probabilistic clustering model for hate speech classification in Twitter","volume":"173","author":"Ayo","year":"2021","journal-title":"Expert Syst. Appl."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"681","DOI":"10.13053\/cys-28-2-4130","article-title":"Comparing Pre-Trained Language Model for Arabic Hate Speech Detection","volume":"28","author":"Daouadi","year":"2024","journal-title":"Computaci\u00f3n y Sistemas"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3616111","article-title":"ENML: Multi-label ensemble learning for Urdu text classification","volume":"22","author":"Mehmood","year":"2023","journal-title":"ACM Trans. Asian Low-Resour. Lang. Inf. Process."},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Hussain, N., Qasim, A., Akhtar, Z.U.D., Qasim, A., Mehak, G., del Socorro Espindola Ulibarri, L., and Gelbukh, A. (2023). Stock Market Performance Analytics Using XGBoost. Advances in Computational Intelligence, Proceedings of the Mexican International Conference on Artificial Intelligence, Yucat\u00e1n, Mexico, 13\u201318 November 2023, Springer Nature.","DOI":"10.1007\/978-3-031-47765-2_1"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"e2372","DOI":"10.7717\/peerj-cs.2372","article-title":"Understanding hate speech: The HateInsights dataset and model interpretability","volume":"10","author":"Arshad","year":"2024","journal-title":"PeerJ Comput. Sci."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"27225","DOI":"10.1109\/ACCESS.2024.3367281","article-title":"Enhancing Hate Speech Detection in the Digital Age: A Novel Model Fusion Approach Leveraging a Comprehensive Dataset","volume":"12","author":"Sharif","year":"2024","journal-title":"IEEE Access"},{"key":"ref_16","first-page":"1","article-title":"Hate speech detection in Asian languages: A survey","volume":"Volume 1","author":"Dhanya","year":"2021","journal-title":"Proceedings of the 2021 International Conference on Communication, Control and Information Sciences (ICCISc)"},{"key":"ref_17","first-page":"1031","article-title":"Offensive Language and Hate Speech Detection Using Transformers and Ensemble Learning Approaches","volume":"28","author":"Aklouche","year":"2024","journal-title":"Comput. Sist."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"100194","DOI":"10.1016\/j.osnem.2021.100194","article-title":"Selecting and combining complementary feature representations and classifiers for hate speech detection","volume":"28","author":"Cruz","year":"2022","journal-title":"Online Soc. Netw. Media"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"106363","DOI":"10.1109\/ACCESS.2021.3100435","article-title":"Detecting white supremacist hate speech using domain-specific word embedding with deep learning and BERT","volume":"9","author":"Alatawi","year":"2021","journal-title":"IEEE Access"},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"109465","DOI":"10.1109\/ACCESS.2021.3101977","article-title":"Un-compromised credibility: Social media based multi-class hate speech classification for text","volume":"9","author":"Qureshi","year":"2021","journal-title":"IEEE Access"},{"key":"ref_21","first-page":"575","article-title":"Offensive Language Detection on Social Media using machine learning","volume":"15","author":"Abdrakhmanov","year":"2024","journal-title":"Int. J. Adv. Comput. Sci. Appl."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"2324","DOI":"10.1016\/j.procs.2020.04.252","article-title":"DHOT-repository and classification of offensive tweets in the Hindi language","volume":"171","author":"Jha","year":"2020","journal-title":"Procedia Comput. Sci."},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Nasir, S., Seerat, A., and Wasim, M. (2024, January 19\u201320). Hate speech detection in Roman Urdu using machine learning techniques. Proceedings of the 2024 5th International Conference on Advancements in Computational Sciences (ICACS), Lahore, Pakistan.","DOI":"10.1109\/ICACS60934.2024.10473250"},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"713","DOI":"10.1007\/s10579-023-09642-7","article-title":"UHated: Hate speech detection in Urdu language using transfer learning","volume":"57","author":"Arshad","year":"2023","journal-title":"Lang. Resour. Eval."},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Hussain, N., Qasim, A., Mehak, G., Kolesnikova, O., Gelbukh, A., and Sidorov, G. (2025). ORUD-Detect: A Comprehensive Approach to Offensive Language Detection in Roman Urdu Using Hybrid machine learning\u2013Deep Learning Models with Embedding Techniques. Information, 16.","DOI":"10.3390\/info16020139"},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Jadon, P., Bhatia, D., and Mishra, D.K. (2024, January 13\u201314). Social Media Text Classification for Hate Speech Detection Using Different Feature Selection Techniques. Proceedings of the 2024 IEEE 4th Int. Conf. on ICT in Business Industry & Government (ICTBIG), Indore, India.","DOI":"10.1109\/ICTBIG64922.2024.10911696"},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"121133","DOI":"10.1109\/ACCESS.2022.3216375","article-title":"Context-aware deep learning model for detection of Roman Urdu hate speech on social media platform","volume":"10","author":"Bilal","year":"2022","journal-title":"IEEE Access"},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"159660","DOI":"10.1109\/ACCESS.2024.3487433","article-title":"Improving Hate Speech Detection Using Double-Layers Hybrid CNN-RNN Model on Imbalanced Dataset","volume":"12","author":"Riyadi","year":"2024","journal-title":"IEEE Access"},{"key":"ref_29","first-page":"396","article-title":"Hate speech detection in social networks using machine learning and deep learning methods","volume":"14","author":"Toktarova","year":"2023","journal-title":"Int. J. Adv. Comput. Sci. Appl."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"8","DOI":"10.1140\/epjds\/s13688-022-00319-9","article-title":"Tackling racial bias in automated online hate detection: Towards fair and accurate detection of hateful users with geometric deep learning","volume":"11","author":"Ahmed","year":"2022","journal-title":"EPJ Data Sci."},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"Abdellaoui, I., Ibrahimi, A., El Bouni, M.A., Mourhir, A., Driouech, S., and Aghzal, M. (2024). Investigating Offensive Language Detection in a Low-Resource Setting with a Robustness Perspective. Big Data Cogn. Comput., 8.","DOI":"10.3390\/bdcc8120170"},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Meque, A.G.M., Hussain, N., Sidorov, G., and Gelbukh, A. (2023). Machine Learning-Based Guilt Detection in Text. Sci. Rep., 13.","DOI":"10.1038\/s41598-023-38171-0"},{"key":"ref_33","doi-asserted-by":"crossref","unstructured":"Aluru, S.S., Mathew, B., Saha, P., and Mukherjee, A. (2021). A deep dive into multilingual hate speech classification. Machine Learning and Knowledge Discovery in Databases. Applied Data Science and Demo Track: European Conference, ECML PKDD 2020, Ghent, Belgium, September 14\u201318, 2020, Proceedings, Part V, Springer International Publishing.","DOI":"10.1007\/978-3-030-67670-4_26"},{"key":"ref_34","doi-asserted-by":"crossref","unstructured":"Rizwan, H., Shakeel, M.H., and Karim, A. (2020, January 16\u201320). Hate-speech and offensive language detection in Roman Urdu. Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), Online.","DOI":"10.18653\/v1\/2020.emnlp-main.197"},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"1621","DOI":"10.1080\/08839514.2021.1988443","article-title":"Semi-supervised self-training of hate and offensive speech from social media","volume":"35","author":"Alsafari","year":"2021","journal-title":"Appl. Artif. Intell."},{"key":"ref_36","unstructured":"Mikolov, T., Chen, K., Corrado, G., and Dean, J. (2013). Efficient Estimation of Word Representations in Vector Space. arXiv."},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"215","DOI":"10.1111\/j.2517-6161.1958.tb00292.x","article-title":"The regression analysis of binary sequences","volume":"20","author":"Cox","year":"1958","journal-title":"J. R. Stat. Soc. Ser. B Stat. Methodol."},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"273","DOI":"10.1007\/BF00994018","article-title":"Support-vector networks","volume":"20","author":"Cortes","year":"1995","journal-title":"Mach. Learn."},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"5","DOI":"10.1023\/A:1010933404324","article-title":"Random forests","volume":"45","author":"Breiman","year":"2001","journal-title":"Mach. Learn."},{"key":"ref_40","doi-asserted-by":"crossref","unstructured":"Lewis, D.D. (1998). Naive (Bayes) at forty: The independence assumption in information retrieval. Machine Learning: ECML-98, Proceedings of the European Conference on Machine Learning, Chemnitz, Germany, 21\u201323 April 1998, Springer.","DOI":"10.1007\/BFb0026666"},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"21","DOI":"10.1109\/TIT.1967.1053964","article-title":"Nearest neighbor pattern classification","volume":"13","author":"Cover","year":"1967","journal-title":"IEEE Trans. Inf. Theory"},{"key":"ref_42","doi-asserted-by":"crossref","unstructured":"Friedman, J.H. (2001). Greedy function approximation: A gradient boosting machine. Ann. Stat., 1189\u20131232.","DOI":"10.1214\/aos\/1013203451"},{"key":"ref_43","unstructured":"Zhang, X., Zhao, J., and LeCun, Y. (2015). Character-level convolutional networks for text classification. arXiv."},{"key":"ref_44","doi-asserted-by":"crossref","first-page":"179","DOI":"10.1207\/s15516709cog1402_1","article-title":"Finding structure in time","volume":"14","author":"Elman","year":"1990","journal-title":"Cogn. Sci."},{"key":"ref_45","doi-asserted-by":"crossref","unstructured":"Qasim, A., Mehak, G., Hussain, N., Gelbukh, A., and Sidorov, G. (2025). Detection of Depression Severity in Social Media Text Using Transformer-Based Models. Information, 16.","DOI":"10.3390\/info16020114"},{"key":"ref_46","doi-asserted-by":"crossref","unstructured":"Cho, K., Van Merri\u00ebnboer, B., Gulcehre, C., Bahdanau, D., Bougares, F., Schwenk, H., and Bengio, Y. (2014). Learning phrase representations using RNN encoder-decoder for statistical machine translation. arXiv.","DOI":"10.3115\/v1\/D14-1179"},{"key":"ref_47","unstructured":"Powers, D.M. (2020). Evaluation: From precision, recall and F-measure to ROC, informedness, markedness and correlation. arXiv."}],"container-title":["Algorithms"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1999-4893\/18\/6\/331\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,9]],"date-time":"2025-10-09T17:45:07Z","timestamp":1760031907000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1999-4893\/18\/6\/331"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,6,1]]},"references-count":47,"journal-issue":{"issue":"6","published-online":{"date-parts":[[2025,6]]}},"alternative-id":["a18060331"],"URL":"https:\/\/doi.org\/10.3390\/a18060331","relation":{},"ISSN":["1999-4893"],"issn-type":[{"value":"1999-4893","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,6,1]]}}}