Fake or true 2023

Inspiration

We are a team of two, and this is our first hackathon experience. It seemed challenging and also exciting to do such a project in two days, and we really enjoy solving problems and taking on challenges. Also, we like data visualisation, and this was a great opportunity to assess our abilities.

What it does

We tried to analyse and dive deeply into the dataset to see what features fake news has that true news does not have, and we reached amazing results. Also, we deployed a passive-aggressive machine learning model that classifies texts given to it with 95% accuracy.

How we built it

At first, we analysed the data and understood the concept of it deeply, and then we started to make a strong model to recognise the difference between two kinds of news.

Challenges we ran into

Time management was really important in this project, as we only had two days. On the other hand, the dataset was huge, and we had to use GPA to run our codes, so we chose Colab to use its GPU, but it tended to crash most of the time, and running the code took so much time.

Accomplishments that we're proud of

1- Making a webpage illustrating the plots and figures interactively 2- training the machine learning model with good accuracy (95%). 3- deploying the machine learning model on the server and making a user-friendly UI for users to test it easily and unlimitedly. 4- Accomplishing the task of doing more than eight analyses and explaining the interpretations about them in detail.

What we learned

This project helped us to investigate and learn the most useful lessons about natural language processing, text analysis, and visualization in a limited time, which is a big achievement.