coast-core 1.0.0

Core functionality of COAST

COAST_CORE is a tool designed for aiding the credibility assessment of online articles. It is a collection of modules that are useful for assessing various aspects of credibility.

• Free software: MIT license
• Documentation: https://coast-core.readthedocs.io.

Features

COAST_CORE is made up of several modules for:

• N-Gram extraction
• Citation detection and classification
• Clarity of writing assessment
• Code detection
• Event detection
• Keyword detection
• Named entity detection

Prerequisites

The tool is built in Python 3 and tested in versions 3.5 and 3.6.

There are two methods of named entity detection included as part of COAST_CORE. For running the Stanford named entity detection, you will need Java installed.

Installation

To install COAST_CORE, run this command in your terminal:

$ pip install coast_core

This is the preferred method to install COAST_CORE, as it will always install the most recent stable release.

If you don’t have pip installed, this Python installation guide can guide you through the process.

To install from source, visit our documentation.

History

Date	Status
April 2016	Research on credibility begins, some initial scripts are written as part of various studies
First half of 2017	Adrien Aucher joins UC as an intern and works with Ashley Williams on the first version of this tool. It is only used internally at this point.
April 2018	Yann Le Norment joins UC as an intern and works on a first public release.
May 2018	First public release!
August 2018	Version 0.1.2 released; fixing bugs and making things stable.