Is It Real-ly Fake News?
Nathan Jin, Qi Jia Sun
Burnaby Mountain Secondary
Floor Location : M 040 F

With the constant increase in number of political news sites on the internet, individuals today are being exposed to more information than ever before. Of course, some of this information is not necessarily accurate. This study looks to determine if there is a relationship between the presence of certain words in a political article and the corresponding validity of said article. To test for the existence of such a relationship, data was compiled on the distribution of words in political articles that have been verified to be “fake” as well as the distribution of words in credible political news sources (the control group). To assist in compiling data, code was written in Java that could access the URLs of each political news article and compile the frequencies of all the words that appear in each article and consolidate all of it into one list. The top 100 most common words in the English Language (as set by the Oxford English Dictionary) will be removed from the data list to shift the focus towards words that can dramatically impact the implications of a document. The word distributions of the “fake” news articles will then be compared with that of the control group, and then the chi-square testing method will be implemented to verify if the differences in the data are statistically significant.

Currently, the study is still in progress. However, data trends are projected to have significant differences from each other. It is predicted that the “fake” news sites will generally use simpler wording and may also employ extreme and as a result biased phrasing to elicit a desired response from the intended audience. The “real” or credible news sources will generally be more professional in diction and impartial in word choice. As the process of interpreting data is still ongoing, a more specific answer will be made available at a later date.

The study examines the relationship between word usage and the credibility of a political news article. As previously elaborated, a relationship is predicted to exist between word usage and reliability of a political news article, but further analysis is necessary to completely establish such a conclusion.