Machine Learning Capstone to differentiate Authentic vs Fake comments
Proposal: Create a machine learning model to identify authentic vs fake comments for FCC Net Nuetrality.
Problem: The FCC Net Neutrality public comment section has been spammed by bots with fake pro-repeal comments.
Client: The internet using public and the FCC.
Data: A SQL data base of public comments with their text and the associated zip codes.
Approach: Develop a model on a sample of the 161GB of data expanding upon on the work of Jeff Keo’s post at hackermoon.com, to create a simple classifier to identify which texts are authentic.
Deliverables: A report with a machine learning model that can distinguish between