{"type":"ScholarlyArticle","title":"Using Machine Learning to Enhance the Detection of Obfuscated Abusive Words in Swahili: A Focus on Child Safety","authors":[{"name":"Phyllis Nabangi","slug":"phyllis-nabangi","given":"Phyllis","family":"Nabangi","affiliation":"Information and Communication Technology , Carnegie Mellon University Africa"},{"name":"Abdul-Jalil Zakaria","slug":"abdul-jalil-zakaria","given":"Abdul-Jalil","family":"Zakaria","affiliation":"Information and Communication Technology , Carnegie Mellon University Africa"},{"name":"Jema David Ndibwile","slug":"jema-david-ndibwile","given":"Jema David","family":"Ndibwile","affiliation":"Information and Communication Technology , Carnegie Mellon University Africa"}],"abstract":"The rise of digital technology has dramatically increased the potential for cyberbullying and online abuse, necessitating enhanced measures for detection and prevention, especially among children. This study focuses on detecting abusive obfuscated language in Swahili, a low-resource language that poses unique challenges due to its limited linguistic resources and technological support. Swahili is chosen due to its popularity and being the most widely spoken language in Africa, with over 16 million native speakers and upwards of 100 million speakers in total, spanning regions in East Africa and some parts of the Middle East. We employed machine learning models including Support Vector Machines (SVM), Logistic Regression, and Decision Trees, optimized through rigorous parameter tuning and techniques like Synthetic Minority Over-sampling Technique (SMOTE) to handle data imbalance. Our analysis revealed that, while these models perform well in high-dimensional textual data, our dataset’s small size and imbalance limit our findings’ generalizability. The precision, recall, and F1 scores were thoroughly analyzed, highlighting the nuanced performance of each model in detecting obfuscated language. This research contributes to the broader discourse on ensuring safer online environments for children, advocating for expanded datasets and advanced machine-learning techniques to improve the effectiveness of cyberbullying detection systems. Future work will focus on enhancing data robustness, exploring transfer learning, and integrating multimodal data to create more comprehensive and culturally sensitive detection mechanisms.","doi":"10.24963/ijcai.aai4g.2024/6","url":"https://ijcai.org/proceedings_aai4g/2024/6","pdf":"https://ijcai.org/proceedings_aai4g/2024/0006.pdf","markdown":"https://ijcai.org/proceedings_aai4g/2024/6.md","bibtex":"https://ijcai.org/proceedings_aai4g/2024/bibtex/6","volume":{"slug":"aai4g-2024","title":"Proceedings of the Second IJCAI AI for Good Symposium in Africa hosted by Deep Learning Indaba","short_title":"AI for Good Africa 2024","url":"https://ijcai.org/proceedings_aai4g/2024/","year":2024,"location":"Dakar, Senegal","dates":"1-7 September 2024","editors":"Avishkar Bhoopchand, Girmaw Abebe Tadesse, Sibusisiwe Makhanya, Frank Dignum, Georgina Curto","isbn":"978-1-956792-07-2","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.aai4g.2024"},"section":"Indaba","track":"Indaba","number":6,"pages":{"first":35,"last":41},"keywords":["Indaba: Indaba"],"videos":[],"references":[{"position":1,"raw":"[Achieng' et al., 2022] Rachel Achieng', Emmah Wakoli, and Michelle Rodrot. Africa's ed-tech platforms: Pro- tecting children's right to privacy. Journal of Intellec- tual Property and Information Technology Law (JIPIT), 2, 2022.","title":"Rachel Achieng', Emmah Wakoli, and Michelle Rodrot. Africa's ed-tech platforms: Protecting children's right to privacy","authors":"' Achieng","year":"2022","venue":"Journal of Intellectual Property and Information Technology Law (JIPIT)","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Akhter et al., 2022] Muhammad Pervez Akhter, Zheng Jiangbin, Irfan Raza Naqvi, Mohammed Abdel Majeed, and Tehseen Zia. Abusive language detection from social media comments using conventional machine learning and deep learning approaches. volume 28, 2022.","title":"Abusive language detection from social media comments using conventional machine learning and deep learning approaches","authors":"Muhammad Pervez Akhter, Zheng Jiangbin, Irfan Raza Naqvi, Mohammed, Abdel Majeed, Tehseen Zia","year":"2022","venue":null,"doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Alliance, 2023a] WeProtect Global Alliance. Analysis of the sexual threats children face online, 2023.","title":"WeProtect Global Alliance. Analysis of the sexual threats children face online","authors":null,"year":"2023","venue":null,"doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Alliance, 2023b] WeProtect Global Alliance. Global threat assessment 2023, 2023.","title":"WeProtect Global Alliance. Global threat assessment","authors":null,"year":"2023","venue":null,"doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Conneau et al., 2020] Alexis Conneau, Kartikay Khandel- wal, Naman Goyal, Vishrav Chaudhary, Guillaume Wen- zek, Francisco Guzmán, Edouard Grave, Myle Ott, Luke Zettlemoyer, and Veselin Stoyanov. Unsupervised cross- lingual representation learning at scale. arXiv preprint arXiv:1911.02116, 2020.","title":"Unsupervised crosslingual representation learning at scale","authors":"Alexis Conneau, Kartikay Khandelwal, Naman Goyal, Vishrav Chaudhary, Guillaume Wenzek, Francisco Guzmán, Edouard Grave, Myle Ott, Luke Zettlemoyer, Veselin Stoyanov","year":"2020","venue":null,"doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Davidson et al., 2017] Thomas Davidson, Dana Warmsley, Michael Macy, and Ingmar Weber. Automated hate speech detection and the problem of offensive language. Proceed- ings of the 11th International AAAI Conference on Web and Social Media (ICWSM), pages 512-515, 2017.","title":"Automated hate speech detection and the problem of offensive language","authors":"Thomas Davidson, Dana Warmsley, Michael Macy, Ingmar Weber","year":"2017","venue":"Proceedings of the 11th International AAAI Conference on Web and Social Media (ICWSM)","doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Eronen et al., 2023] Juuso Eronen, Michal Ptaszynski, and Fumito Masui. Zero-shot cross-lingual transfer language selection using linguistic similarity. Information Process- ing & Management, 60(3):103250, 2023.","title":"Zero-shot cross-lingual transfer language selection using linguistic similarity","authors":"Eronen","year":"2023","venue":"Information Processing & Management","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Ethem F. Can and Can, 2018] Aysu Ezen-Can Ethem F. Can and Fazli Can. Multilingual sentiment analysis: An rnn- based framework for limited data. In Proceedings of ACM SIGIR 2018 Workshop on Learning from Limited or Noisy Data, 2018.","title":"Multilingual sentiment analysis: An rnnbased framework for limited data","authors":"F Ethem, Can, F Can ; Aysu Ezen-Can Ethem, Fazli Can, Can","year":"2018","venue":"Proceedings of ACM SIGIR 2018 Workshop on Learning from Limited or Noisy Data","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Gunawan et al., 2018] Fergyanto Gunawan, Livia Ashianti, and Nobumasa Sekishita. A simple classifier for de- tecting online child grooming conversation. Telkomnika (Telecommunication Computing Electronics and Control), 16:1239-1248, 06 2018.","title":"Fergyanto Gunawan, Livia Ashianti, and Nobumasa Sekishita. A simple classifier for detecting online child grooming conversation","authors":"Gunawan","year":"2018","venue":"Telkomnika (Telecommunication Computing Electronics and Control)","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Isaza et al., 2022] Gustavo Isaza, Fabián Muñoz, Luis Castillo, and Felipe Buitrago. Classifying cybergrooming for child online protection using hybrid machine learning model. Neurocomputing, 484:250-259, 2022.","title":"Classifying cybergrooming for child online protection using hybrid machine learning model","authors":"Gustavo Isaza, Fabián Muñoz, Luis Castillo, Felipe Buitrago","year":"2022","venue":"Neurocomputing","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Jonathan et al., 2021] Francis Jonathan, Dong Yang, Glyn Gowing, and Songjie Wei. Machine learning framework for detecting offensive swahili messages in social net- works with apache spark implementation. pages 293- 297. Institute of Electrical and Electronics Engineers Inc., 2021.","title":"Machine learning framework for detecting offensive swahili messages in social networks with apache spark implementation","authors":"Francis Jonathan, Dong Yang, Glyn Gowing, Songjie Wei","year":"2021","venue":null,"doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Khairy et al., 2023] Marwa Khairy, Tarek M. Mahmoud, Ahmed Omar, and Tarek Abd El-Hafeez. Comparative performance of ensemble machine learning for arabic cy- berbullying and offensive language detection. Language Resources and Evaluation, 2023.","title":"Comparative performance of ensemble machine learning for arabic cyberbullying and offensive language detection","authors":"Marwa Khairy, Tarek M Mahmoud, Ahmed Omar, Tarek Abd El-Hafeez","year":"2023","venue":"Language Resources and Evaluation","doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Mahmud et al., 2023] Tanjim Mahmud, Michal Ptaszynski, Juuso Eronen, and Fumito Masui. Cyberbullying detec- tion for low-resource languages and dialects: Review of the state of the art. Information Processing & Manage- ment, 60(5):103454, 2023.","title":"Cyberbullying detection for low-resource languages and dialects: Review of the state of the art","authors":"Mahmud","year":"2023","venue":"Information Processing & Management","doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Mambina et al., 2022] I. S. Mambina, J. D. Ndibwile, and K. F. Michael. Classifying swahili smishing attacks for mobile money users: A machine-learning approach. IEEE Access, 10:83061-83074, 2022.","title":"Classifying swahili smishing attacks for mobile money users: A machine-learning approach","authors":"I S Mambina, J D Ndibwile, K F Michael","year":"2022","venue":"IEEE Access","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Mekala et al., 2023] Sreenivas Mekala, Y Nithin, B Raghav, and T Kumar. Cyberbullying detection using machine learning. 12:729-733, 06 2023.","title":"Cyberbullying detection using machine learning","authors":"Sreenivas Mekala, B Nithin, Raghav, Kumar","year":"2023","venue":null,"doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Nekoto et al., 2020] Wilhelmina Nekoto, Vukosi Marivate, Sarah Matseleng, Timi Fasubaa, Tolulope Fagbohungbe, Temidayo Fagbohungbe, Solomon Oluwarotimi Akinola, Jade Abbott, Hussein Suleman, and Iroro Orife. Par- ticipatory research for low-resourced machine transla- tion: A case study in african languages. arXiv preprint arXiv:2008.12272, 2020.","title":"Participatory research for low-resourced machine translation: A case study in african languages","authors":"Wilhelmina Nekoto, Vukosi Marivate, Sarah Matseleng, Timi Fasubaa, Tolulope Fagbohungbe, Temidayo Fagbohungbe, Solomon Oluwarotimi Akinola, Jade Abbott, Hussein Suleman, Iroro Orife","year":"2020","venue":null,"doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Njovangwa and Justo, 2021] Geofrey Njovangwa and God- frey Justo. Automated detection of bilingual obfuscated abusive words on social media forums: A case of swahili and english texts. Tanzania Journal of Science, 47:1352- 1361, 10 2021.","title":"Geofrey Njovangwa and Godfrey Justo. Automated detection of bilingual obfuscated abusive words on social media forums: A case of swahili and english texts","authors":"Justo Njovangwa","year":"2021","venue":"Tanzania Journal of Science","doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Pooyandeh et al., 2022] Mitra Pooyandeh, Ki-Jin Han, and Insoo Sohn. Cybersecurity in the ai-based metaverse: A survey. In Applied Sciences, volume 12, page 12993, 2022.","title":"Cybersecurity in the ai-based metaverse: A survey","authors":"Mitra Pooyandeh, Ki-Jin Han, Insoo Sohn","year":"2022","venue":"Applied Sciences","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Ptaszynski and Masui, 2018] Michal E Ptaszynski and Fu- mito Masui. Automatic cyberbullying detection: Emerg- ing research and opportunities: Emerging research and op- portunities. 2018.","title":"Automatic cyberbullying detection: Emerging research and opportunities: Emerging research and opportunities","authors":"Michal E Ptaszynski, Fumito Masui","year":"2018","venue":null,"doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Qaiser and Ali, 2018] Shahzad Qaiser and Ramsha Ali. Text mining: use of TF-IDF to examine the relevance of words to documents. International Journal of Computer Applications, 181(1):25-29, 2018.","title":"Text mining: use of TF-IDF to examine the relevance of words to documents","authors":"Ali ; Qaiser, Shahzad Qaiser, Ramsha Ali","year":"2018","venue":"International Journal of Computer Applications","doi":null,"url":null,"internal_url":null},{"position":21,"raw":"Radisek, ] Vince Radisek. What is extended reality? every- thing you need to know.","title":"What is extended reality? everything you need to know","authors":"] Radisek, Vince Radisek","year":null,"venue":null,"doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Ruder et al., 2019] Sebastian Ruder, Matthew E Peters, Swabha Swayamdipta, and Thomas Wolf. Neural transfer learning for natural language processing. arXiv preprint arXiv:1903.05987, 2019.","title":"Neural transfer learning for natural language processing","authors":"Sebastian Ruder, Matthew E Peters, Swabha Swayamdipta, Thomas Wolf","year":"2019","venue":null,"doi":null,"url":null,"internal_url":null},{"position":23,"raw":"[Sanchez et al., 2019] L. Sanchez, Cinthya Grajeda, Ibrahim Baggili, and Cory Hall. A practitioner survey exploring the value of forensic tools, ai, filtering, & safer presenta- tion for investigating child sexual abuse material (csam). Digital Investigation, 29, 2019.","title":"A practitioner survey exploring the value of forensic tools, ai, filtering, & safer presentation for investigating child sexual abuse material (csam)","authors":"L Sanchez, Cinthya Grajeda, Ibrahim Baggili, Cory Hall","year":"2019","venue":"Digital Investigation","doi":null,"url":null,"internal_url":null},{"position":24,"raw":"[scikit learn, 2024] scikit learn. sklearn.metrics, 2024. [UNICEF, 2018] UNICEF. More than 175,000 children go online for the first time every day, tapping into great op- portunities, but facing grave risks, 2018. [Unicef, 2022] Unicef. Protecting children online, 2022.","title":"scikit learn. sklearn.metrics","authors":null,"year":"2018","venue":null,"doi":null,"url":null,"internal_url":null},{"position":25,"raw":"[WeProtect Global Alliance, 2023] PA Consulting WePro- tect Global Alliance. Global threat assessment 2023 as- sessing the scale and scope of child sexual exploitation and abuse online, to transform the response ai-generated image, 2023.","title":"PA Consulting WeProtect Global Alliance. Global threat assessment 2023 assessing the scale and scope of child sexual exploitation and abuse online, to transform the response ai-generated image","authors":null,"year":"2023","venue":null,"doi":null,"url":null,"internal_url":null},{"position":26,"raw":"[Zaccagnino et al., 2021] Rocco Zaccagnino, Carmine Capo, Alfonso Guarino, Nicola Lettieri, and Delfina Malandrino. Techno-regulation and intelligent safeguards: Analysis of touch gestures for online child protection. Multimedia Tools and Applications, 80, 2021.","title":"Techno-regulation and intelligent safeguards: Analysis of touch gestures for online child protection","authors":"Rocco Zaccagnino, Carmine Capo, Alfonso Guarino, Nicola Lettieri, Delfina Malandrino","year":"2021","venue":"Multimedia Tools and Applications","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}