Ad amīcus: Analysis of click behaviour in social relationships extracted from URLs

Social networks have great reach and generate a lot of traffic, especially on mobile devices. Advertising on these networks, though highly rewarding, has many unique challenges. We present a method that deduces social network relationships from just the URLs of the pages on which the ads are sought to be displayed. We show that there is a high correlation between clicking behaviour and relationships in such an inferred social network. The results have potentially huge implications for third party advertising networks like InMobi that serve ads on mobile social networking websites.