ANALYSIS OF NLP METHODS TO IDENTIFY OFFENSIVELANGUAGE

Abstract

This research focuses on the application of Natural Language
Processing (NLP) techniques to detect offensive language in textual data aimed at improving content moderation on digital communication platforms. Using a dataset, the study evaluates the effectiveness of advanced NLP models and algorithms in detecting explicit and implicit forms of offensive language. The core of the analysis centers around transformer-based models, in particular
BERT (Bidirectional Encoder Representations from Transformers). The study addresses the challenges of offensive expression detection, highlighting both the successes and challenges faced in accurately classifying text as offensive or not.
This research contributes to ongoing efforts to create a safer and more inclusive digital environment by offering insight into the potential of NLP technologies to address the widespread problem of profanity on the Internet.