Author

Diwas Sharma

Date of Award

2018

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Computer Science

Committee Chair

Ramazan Aygun

Committee Member

Huaming Zhang

Committee Member

Vineetha Menon

Subject(s)

Social media, Journalism, Computational linguistics--Methodology

Abstract

Social media platforms nowadays have a large number of fake or false news which have been misleading and negatively impacting viewers. In order to combat the problem, being able to differentiate important news stories which need to be verified from unimportant news stories which need not, would be a decent starting point. In this thesis, we introduce “significant news” and define it as news that affects a large number of people, changes the routines of daily life, and needs verification onthe information presented. This thesis then explores if it is possible to construct a classifier for detecting the significant news articles. A dataset containing 1548 significant and 595 non-significant articles was prepared by manually labelling the posts obtained from Twitter. Various classifiers including logistic regression, support vector machine, random forest, and neural network – were trained on the dataset. They each achieved an accuracy greater than 90 percent, with the neural network model achieving the highest accuracy of 93.654 percent. This indicates that it is in fact possible to build fairly accurate classifiers for detecting significant news. This thesis then describes a few possible future directions that could be explored for further improving the performance of significant news detection.

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.