Cyberbullying and Fake Account Detection on Social Media Using Machine Learning and NLP
B.Amarnath Reddy · INTERANTIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT · 2024
Now a days the use of social media has grown exponentially with the simultaneous growth of the internet throughout the world and it is mostly attracted by the youth. However, the enhancement of social connectivity by using this social media platform may also lead to a negative impact on society mainly cybercrime, cyberbullying, abuse, etc. This cyberbullying is a major problem in this society which leads to frequent physical and mental stress, particularly for teenagers, transgenders, and women. Cyberbullying is done by using comments on social media platforms commenting vulgar words and also due to the increase of user’s various malicious entities like increasing fake accounts increased a lot. Detection of cyberbullying and fake accounts on such large platforms is very difficult and may sometimes lead to false detection. So many incidents have recently occurred throughout the world like suicides, mental illness, etc. So our goal is to identify bullying comments and fake accounts on social media platforms like Facebook, Twitter, etc. By merging natural language processing and machine learning algorithms like xgboost(extreme gradient boosting) to detect fake accounts and logistic regression for cyberbullying. Key Words: Cyberbullying, Natural language processing, Machine learning, xgboost(extreme gradient boosting), logistic regression, Random forest, etc.