Overall analysis of results
Machine Learning Output
Google Search Results

TrustyNews

Description

The purpose of TrustyNews is to have a way to systematically and accurately identify whether a given article is fake news or real news. The site uses three different techniques to accomplish this:

Sentiment and Emotional analysis powered by IBM Watson
Number of search results in Google News to scale credibility
Fool-proof collection of domains to verify validity of source

The site then reports a percentage reliability to the user by a weighted average of these three techniques.

TrustyNews is a website written in Python powered by Django, HTML5, CSS3 and Javascript.

Built With

Submitted to

DeltaHacks III

Created by

I worked on implementing the Machine Learning analysis using the Watson Python SDK, as well as working on some back end algorithms.

Haoran Zhang
I handled the custom google search API using python. The returning data was then passed on to Watson executed linguistic analysis. Further analysis was made against database containing previously proven fake news sites.

Leo Kim
I worked on front-end UI displaying data with HTML5, CSS3, and Javascript. I also worked on relating back-end server to front-end.

Jack Yan

Updates

Haoran Zhang started this project — Jan 29, 2017 10:42 AM EST

Leave feedback in the comments!

Log in or sign up for Devpost to join the conversation.