A project focused on hate speech detection across Twitter and Instagram using machine learning and deep learning techniques. Combines supervised learning on Twitter datasets and unsupervised learning techniques applied to scraped Instagram comments for a generalized, robust model.

Successfully established a multiclass text classification model by fine-tuning pretrained DistilBERT transformer model to classify several distinct types of mental health statuses such as anxiety, stress, personality disorder, etc. with an accuracy of 77%.