Using Sentiment Analysis for Better Placement of Contextual Advertisements

Abstract

Online advertising is one of the most lucrative forms of advertising, with digital ad-spending expected to reach $83 billion in 2017, making it one of the most important channels of advertising media. Contextual Advertising is a type of online display advertising that takes cues from the content of the triggering page and displays advertisements that are relevant to the current context, increasing the probability of an impression conversion, while at the same time avoiding being too annoying or disinteresting to the user. However, on several occasions, the context may have a negative connotation, and displaying advertisements that are relevant to it might prove to be detrimental to the advertiser. We refer to such a scenario as an unfortunate placement.In this thesis, we propose APNEA (Ad Positive NEgative Analysis), a light-weight system that extracts the sentiment from the triggering page and associates it with the relevant advertising brands. APNEA uses a sentiment-oriented approach to rank the advertisers based on their associated sentiments such that positively correlated brands are ranked higher than brands that are neutral or negatively correlated, while maintaining the relative order of relevance, thereby avoiding an unfortunate placement.Experiments show that APNEA helps avoid unfortunate placements while maintaining ad-relevance. It outperforms several baselines in terms of accuracy on human-annotated test data while having a lower run-time, which is crucial for real-time bidding systems.