Title

Measuring the Bias of the Media's Many Voices

Presenter Information

Paul Williams

Document Type

Oral Presentation

Location

SURC 137B

Start Date

21-5-2015

End Date

21-5-2015

Keywords

Media, Bias, Metric

Abstract

Breaking news is often delivered by various sources of media, but the wording used can elicit a specific response from the readers, creating bias. We have created a suite of tools for conducting media bias research. Our beta version uses an open source web spider toolkit called Scrapy to obtain the media website’s text. This web spider is implemented by custom built back-end Python and Bash scripts. These scripts generate an XML file containing the text gathered from the media websites. Our web tool calculates metrics for performing media bias analysis by use of a large library of adjectives that are rated as either positive or negative. Finally, the tool displays those metrics for the user on a web graphical user interface. Using this first version of our tool, we are able to demonstrate a ranking of the bias of the text of two media sources.

Faculty Mentor(s)

Filip Jagodzinski

Department/Program

Computer Science

Additional Mentoring Department

Computer Science

This document is currently not available here.

Share

COinS
 
May 21st, 10:00 AM May 21st, 10:20 AM

Measuring the Bias of the Media's Many Voices

SURC 137B

Breaking news is often delivered by various sources of media, but the wording used can elicit a specific response from the readers, creating bias. We have created a suite of tools for conducting media bias research. Our beta version uses an open source web spider toolkit called Scrapy to obtain the media website’s text. This web spider is implemented by custom built back-end Python and Bash scripts. These scripts generate an XML file containing the text gathered from the media websites. Our web tool calculates metrics for performing media bias analysis by use of a large library of adjectives that are rated as either positive or negative. Finally, the tool displays those metrics for the user on a web graphical user interface. Using this first version of our tool, we are able to demonstrate a ranking of the bias of the text of two media sources.