Mining the bit pipes: Discovering and leveraging users' behavior

This paper explores ways to enhance services provided to a user, based on the user's own behavior and context. The assumption is that the user continues to simply use data communication services, while the service provider (SP) mines the content exchanged over the communication pipes for information that can be used to further personalize the same service or other services to which the user has subscribed or opted-in. The content of the messages exchanged by the user may either directly or indirectly present mining opportunities. Messages may be exchanged between users leveraging communication technologies used in wireless networks (e.g. short message service (SMS)) or combinations of wireless and wireline networks (e.g. instant messaging (IM) over a next generation network (NGN) IP multimedia subsystem (IMS). Such messages may be intercepted or forwarded for content screening (CS) for subsequent content analysis (CA) (e.g. similar to deep packet inspection (DPI) techniques). The CA outcome represents information that can be applied to augment existing user's personal and context data, and be further used to offer new or enhanced services to the user (e.g. personalized advertising). In this paper, in addition to exploring the mining of data communication pipes, we are also describing how existing and/or emerging application enablement standards (e.g. OMA's converged IP messaging (CPM), categorization based content screening (CBCS), global permissions management (GPM), services user profile management (ServUserProf), etc.) can be combined and augmented with additional capabilities to create a well integrated ecosystem for personalized services. A specific example of a new or enhanced service that can feed on this ecosystem is also explored, using another emerging OMA standard, OMA mobile advertising (MobAd). A thorough architectural description of most of the open mobile alliance (OMA) specifications referred to in this paper, and how they relate to each other can be found in the work of Brenner and Unmehopa (2008).