What it does
Analyzes student's Facebook posts in order to tell if they are likely to commit suicide, commit violent crimes, or otherwise deviate from normal student behavior.
How I built it
We built it using many different Python libraries, which we put together on Jupyter notebook. Libraries include keras, nltk (Natural Language ToolKit), Pandas, and Numpy
Challenges I ran into
We were able to find a labeled dataset of sentences with their respective sentiment, but we were unable to get good results using a deep neural network. Instead we used the natural language toolkit in order to analyze a phrase's sentiment.
What's next for Facebook Text Sentiment Analysis
We will connect the offline portion (sentiment analysis) with the online portion (educators viewing student's levels of happiness, depression, sadness, etc.)