| Abstrakt: |
Cyber bullying, which is a critical global issue that affects both individual victims and societies in a large number. Cyber bullying refers to the use of electronic communication to bully a person, typically by sending messages of an intimidating or threatening nature. Harmful bullying behaviour can include posting rumours, threats, sexual remarks, a victims' personal information, or pejorative labels. Many attempts have been introduced in the literature to intervene in, prevent, or mitigate cyber bullying; however, because these attempts rely on the victims' interactions, they are practical. Therefore, detection of cyber bullying without the involvement of the victims is necessary. Machine Learning Concepts are used for the implementation of the project which aims to classify the input into bullying or non-bullying statements. This paper detects cyber bullying in tweets using ML Classification Algorithms. Each of these algorithms are evaluated using accuracy, precision, recall, and F1 score as the performance metrics to determine the classifiers' recognition rates applied to the dataset. Training and predicting pipeline is implemented to contrast performance of various classification algorithms and determine the best suited model. It also shows that only the non-bullying tweets can be posted by the user, and any bullying content is discarded and hence not posted. [ABSTRACT FROM AUTHOR] |
