Adaptive Data-Driven and Region-Aware Detection for Deceptive Advertising

We find a new trend, called deceptive advertising (deceptive ads), that some advertisers get paid through pay-perinstall schemes. Deceptive ads is the use of false or misleading statements in advertising, which attracts users to click but negatively affects stakeholders. Nevertheless, despite the popularity of deceptive ads, not much research attention has been devoted to the detection. Bad actors use regional advertising, and have a much shorter lifetime than other security threats (e.g., phishing). The fast flux-like behavior of deceptive ads even poses the difficulty in protecting against deceptive ads. Due to the above reasons, we introduce a detection system for data-driven and region-aware detection of deceptive ads with automated feature extraction for further text-mining and pattern recognition. Our proposed system has been deployed in our testbed for intensive analysis and has shown that such hybrid approach yields acceptable results based on our massive real dataset.