Artificial Intelligence Based Defamation Detection System

Objective: The aim of this project is to automatically detect offensive comments by analysing text from social media platforms or other text sources. This can be used to determine whether community rules have been violated, filter out harmful content and improve the user experience. Artificial Intelligence Model: This system uses the BERT model called "dbmdz/bert-base-turkish-uncased", which can specifically analyse texts in the Turkish language. BERT (Bidirectional Encoder Representations from Transformers) is a revolutionary artificial intelligence model for language modelling and shows high performance in many tasks such as text classification. How does it work? Data Preparation: The dataset used for training consists of texts ("tweet") and tags ("subtask_a") that indicate whether these texts contain insults or not. Model Training: The model is trained on the examples in the dataset. In this process, the model learns the content of the texts and uses this information to detect offensive comments. Prediction: Once trained, the model can make predictions on new text samples. These predictions indicate whether the text is insulting or not (POSITIVE = NOT INSULTING, NEGATIVE = INSULTING). Usage Scenarios: Social Media Moderation: This system can automatically detect harmful comments on social media platforms. Community Moderation: It can be used as an automated moderation tool for forums or other community platforms. Digital Marketing: With this system, brands can analyse customer feedback and detect negative comments. Citation Information @inproceedings{coltekin202020lrec, author = {\c{C}"{o}ltekin, \c{C}a\u{g}r{\i}}, year = {202020}, title = {A Corpus of Turkish Offensive Language on Social Media}, booktitle = {Proceedings of The 12th Language Resources and Evaluation Conference}, pages = {6174--6184}, address = {Marseille, France}, url = {https://www.aclweb.org/anthology/2020.lrec-1.758}, }