Inspiration
In this data-intensive world, it is extremely tough to find labeled data. Hence people have no option but to label data manually which is an extremely time taking process. With datafighters, we intend to gamify this process and make it a fun activity to label data
Discord : Alpha Shivam#4260 punisher0319#6355
What it does
For the sake of this hackathon, we have extracted tweets related to coronavirus and we make the users play a game wherein they have to label the tweets as real or fake and we compare it to the ground value to calculate a score for the user. We then show the user how he fared against the machine learning algorithm for the same task.
How we built it
We used a google app engine to make an API for the ML model and compared the accuracy of the model on a dataset to the user's intuitive accuracy for the same dataset using a Flask backend.
Challenges we ran into
We had a tough time integrating the ML model with our web app. We had to use a google bucket and app engine to generate an API for the same We also used Mongodb for storing the results
Accomplishments that we're proud of
We were able to put up a great UI with a full working functionality
What we learned
We learned how to work together as a team, especially virtually which indeed was a challenge
What's next for Datafighters
We aim to generalize this model for any dataset determined by the user to revolutionize the way data is labelled.
Log in or sign up for Devpost to join the conversation.