Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.
Date of Award
Bachelor of Science (BS)
Department of Computer Science
Ramon A. Mata-Toledo
The purpose of this thesis is to assist in automating the detection of Fake News by identifying which features are more useful for different classifiers. The effectiveness of different extracted features for Fake News detection are going to be examined. When classifying text with machine learning algorithms features have to be extracted from the articles for the classifiers to be trained on. In this thesis, several different features are extracted: word counts, ngram counts, term frequency-inverse document frequency, sentiment analysis, lemmatization, and named entity recognition to train the classifiers. Two classifiers are used, a Random Forest classifier and a Naïve Bayes classifier. Training on different features combined with different machine learning algorithms yields different accuracies. By testing the different features on different classifiers, it can be determined which features are the best for Fake News detection. Classifying news articles as either Fake News or as not Fake News is explored using three datasets, which in total contains over 40,000 articles. One of the datasets is used to partly to train the classifiers and partly to test the classifiers. The remaining two datasets are used purely for testing the classifiers. All the code used in conjunction with thesis can be found in Appendix B.
Shoemaker, Eliza, "Using data science to detect fake news" (2019). Senior Honors Projects, 2010-current. 714.