The problem Tweet Sentiment Identifier (PEGASUS) solves

When someone is treated unfairly because of their ethnicity, culture, or recent activities, this is known as racism or trolling. It can include things like calling them names, withdrawing from them, and even denying them access to services or career prospects. The number of online trolls and bigots is rapidly increasing. This is because the user's anonymity has increased. They make comments and posts on social media accounts because they are confident that they will not be discovered. To prevent this and catch them, we designed a model that, when a user's username is entered, checks (and can display) his or her prior history, i.e. the types of postings he or has made. To prevent this and catch them, we established a model that reveals the person's prior history, such as what type of posts he has been posting, what type of posts he has been tagged in, and what he has commented on, and we can restrict that user from social media networks based on that information. Pegasus can also verify whether any entered comment is racist or not. Our model's specialty is that it can also track and identify comments made in HINDI or in any language. So any person can access the application and get information about their past comments. Thus it makes it more user-friendly and resourceful.

Challenges we ran into

1. Web- scraping of the Twitter Dataset (Request was not accepted by Twitter API)
2. Understanding the pecularities of FLASK (Routing, Post-Get requests etc.)
3. Linking the ML Model to the HTML Website (rendering CSS, JS and physical items etc.)