Advertiser and Publisher-centric Privacy Aware Online Behavioral Advertising

Online behavioral advertising (OBA) has become one of the most successful advertising models on the Internet. Nevertheless, all existing OBA systems are broker-centric in the billing phase, which means it is the broker who exclusively determines advertisers' expenses and publishers' revenues. Consequently, a malicious broker may cheat in their tallying of ad clicks to overcharge advertisers or underpay publishers. Furthermore, as the broker cannot justify the bills, malicious advertisers may deny actual clicks to ask for refunds, and malicious publishers may claim non-existing clicks to demand extra revenue shares. This paper solves these problems by reversing the priority between the broker and the advertisers and publishers. Specifically, when users click on ads, it makes corresponding advertisers and publishers forward click reports of clients to the broker after checking, anonymizing and signing them. The broker then settles accounts with advertisers and publishers fully based on these reports. To guarantee the interests of the broker after the priority reversal, we further propose effective mechanisms for detecting underreporting advertisers and over reporting publishers, respectively.