Classification and Analysis of Political Video Advertisements
Campaign advertising through social media platforms has been growing at a high rate, which creates a large volume of content on the Internet. To increase transparency in federal campaign advertising, Google LLC. created the Google Transparency Report (GTR). GTR provides website and searchable databases about federal election campaign ads aired on Google platforms (Google, YouTube, and partner properties). According to GTR, political advertisers have spent around $800M on election campaigns since May 2018. In this project, we made many improvements on CyAds, a platform for collecting video ads aired on YouTube and for automated content analysis. The CyAds system can 1) automatically classify a video ad into either a political category or a non-political one and 2) analyze predicted political ads into one of these types: promote, attack, or contrast. We analyzed the video ads reported by GTR during January 1, 2020 and November 3, 2020. Our analysis includes 1) extracting issues from political advertising of interest in political science research, 2) determining the polarity and subjectivity of a given ad, 3) comparing ad spending and impressions and issues mentioned by Donald J. Trump and Joe Biden during the 2020 presidential election.
Committee: Adisak Sukul (major professor), Wallapak Tavanapong (major professor), and David Peterson