weoverme / potentially_t_f

Creation of Project - locally and on github

Identifying Verifiable Statements in Politicians' Twitter Feeds

Product Name: Potentially T/F

Monash University Computer Science Project 2017

Author: Young Rae Cho

Motivation

In the current day and age, 'news' is circulated and reported at very high speeds, without regards to whether it is true or not, because whoever gets the news out first can start a trend which many people will look at (producing ad revenue).

In the most current US Presidential Elections in 2017/2016, one may say that the amount of fake news produced which affected the elections was the most out of all elections. This is a phenomenon which is extremely dangerous, as the Presidential seat may be given to the world's biggest liar or something along the lines of it. Fact-checking organizations, such as PolitiFact.com, exist in order to verify whether a statement made regarding political issues is speaking of the truth or false (with other classifications in between Truth and False).

Getting Started

As this application tried to use as few tools as possible, there are just as few dependencies. However, this application is intended for Windows use and has not been tested in Mac OSs.

The following must all be installed, in order of:

• Python 3.4.X - The language and compiler
• NumPy - Install with PyPI
• SciPy - Navigate to SciPy on http://www.lfd.uci.edu/~gohlke/pythonlibs/ and download scipy‑0.19.0‑cp36‑cp36m‑win_amd64.whl, then install this file using PyPI
• scikit-learn - Install with using PyPI, ONLY when the above are all installed.
• nltk v3.5 - See individual installation requirements
• tweepy - Install with PyPI

Currently there is no way to run the application without the in-editor compiler. This was developed in PyCharm, but you may use another editor of your choice.

1. Go to the Potentially T/F Github Repository to clone the application.
2. Open Application.
3. Run Application.

Made by using Dillinger.io