CYBER BULLING DETECTION USING MACHINE LEARNING

Computer science engineering

Cyberbullying has emerged as a pervasive and concerning issue on social media platforms, impacting the mental health and well-being of individuals worldwide. To address this problem, this study proposes a cyberbullying detection system using the Support Vector Machine (SVM) algorithm. Leveraging the power of machine learning, the system aims to automatically identify and flag instances of cyberbullying in social media content. The development of the detection system begins with the collection and labelling of a comprehensive dataset containing examples of cyberbullying and non-cyberbullying posts or comments. After pre-processing the text data by removing irrelevant information, converting text to lowercase, and tokenizing it, meaningful features are extracted using the bag-of-words or TF-IDF techniques. These transformed feature vectors serve as inputs for training the SVM classifier, which seeks to find the optimal hyper plane for effectively distinguishing cyberbullying from non-cyberbullying content. The performance of the SVM model is evaluated using a separate testing dataset, with metrics such as accuracy, precision, recall, F1-score, and ROC-AUC analysed to assess its effectiveness in identifying cyberbullying instances. Model fine-tuning is conducted through experimentation with various SVM hyper parameters and cross-validation techniques to optimize performance.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.