Inspiration
Twitter Microblogs and how any possible misinformation can affect the emotional intensity of people reacting to the tweets
What it does
Detects if the news is fake or not and how intense the emotional level is
How we built it
Created an instance in Z platform , initiated a collaborative Notebook, referred online, processed the data, and customized the program/algorithm to fit our data
Challenges we ran into
low Server computation , large dataset , deep learning model takes more time to fit
Accomplishments that we're proud of
90+% accuracy in the decision tree model and 50% accuracy in the LSTM model
What we learned
Understanding of Fake news , RNN , decision tree, Fuzzy logic
What's next for Misinformation and Emotion Intensity Detection on Microblogs
making the model to fit in multiple languages, more accurate and compile them into one extension
Log in or sign up for Devpost to join the conversation.