{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/transforma-at-semeval-2019-task-6-offensive","title":"Offensive Language Analysis using Deep Learning Architecture","arxiv_id":"1903.05280","date":"2019-03-12","proceeding":null,"authors":["Ryan Ong"],"abstract":"SemEval-2019 Task 6 (Zampieri et al., 2019b) requires us to identify and\ncategorise offensive language in social media. In this paper we will describe\nthe process we took to tackle this challenge. Our process is heavily inspired\nby Sosa (2017) where he proposed CNN-LSTM and LSTM-CNN models to conduct\ntwitter sentiment analysis. We decided to follow his approach as well as\nfurther his work by testing out different variations of RNN models with CNN.\nSpecifically, we have divided the challenge into two parts: data processing and\nsampling and choosing the optimal deep learning architecture. In preprocessing,\nwe experimented with two techniques, SMOTE and Class Weights to counter the\nimbalance between classes. Once we are happy with the quality of our input\ndata, we proceed to choosing the optimal deep learning architecture for this\ntask. Given the quality and quantity of data we have been given, we found that\nthe addition of CNN layer provides very little to no additional improvement to\nour model's performance and sometimes even lead to a decrease in our F1-score.\nIn the end, the deep learning architecture that gives us the highest macro\nF1-score is a simple BiLSTM-CNN.","url_abs":"http://arxiv.org/abs/1903.05280v3","url_pdf":"http://arxiv.org/pdf/1903.05280v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"transforma-at-semeval-2019-task-6-offensive","repo_url":"https://github.com/RyanOngAI/semeval-2019-task6","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":null}],"tasks":[{"task_slug":"abuse-detection","task_name":"Abuse Detection"},{"task_slug":"deep-learning","task_name":"Deep Learning"},{"task_slug":"sentiment-analysis","task_name":"Sentiment Analysis"},{"task_slug":"twitter-sentiment-analysis","task_name":"Twitter Sentiment Analysis"}],"methods":[{"method_slug":"smote","method_name":"SMOTE"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}