Problem Description

Sentiment analysis refers to the analysis of textual content to determine the user’s emotion, sentiment, opinion, and feelings. Over the years, social media has become one of the most important aspects of one’s life, both personally and professionally.

People are increasingly inclined toward social media to express their thoughts and opinions. However, too much of anything is an issue. There have been numerous occasions in which the anonymity offered by social media has proved extremely harmful, be it cyberbullying or the spread of hate speech.

Our Solution

For this sentiment analysis, we have trained a sequential deep learning model using publicly available datasets. We have built an application that takes input from the user, which could be any social media post, preprocesses it, runs our deep learning sequential model on it, and predicts whether the text is hate speech or cyberbullying.

For ease of representation and interaction, we have built a Flask app that deploys our model and provides an interactive interface. This tool can be used for automatic detection and flagging of hate speech and cyberbullying across any social media platform.