Using Information in Access Logs for Large Scale Identity Linkage

With the world becoming more connected than ever and the growth of the number of connected devices, each individual user accesses services from a range of devices, including personal desktops and laptop computers, tablets, mobile devices, vehicles, and entertainment systems. One fundamental problem is to identify the user using the fragmentation of identity and consumer profiles across these connected devices. In this paper, we discuss Adobe's Identity Graph that provides a comprehensive solution to the challenge posed by fragmentation of identities. In particular, we propose an approach to use all the features in the logs using both online data traffic and offline data logs. We use probability based correlation methods to link identities across different devices with high accuracy. We validate the effectiveness of our proposed approach with massive large data covering more than 1.9 billion connected devices resulting in more than 200 million user identities. Our evaluation results show that using information in access logs can be effective in linking identities and achieving a practical and scalable solution.