Morphing content in mobile applications

Abstract

Smart phones are quickly becoming an integral part of our everyday lives. However, the mobile industry is still young, and the full potential of mobile phones has yet to be tapped. In this thesis, I present the design of a new mobile "super-application" called MobileHelp that aims to push the boundaries of how smart phones can make people's lives easier. MobileHelp uses Bayesian inference to determine a user's current purpose. Then, it suggests applications the user may want to use, offers deals and discounts for relevant nearby businesses, shows information about nearby friends and their statuses, and present search results for relevant queries. The power of MobileHelp is that it does all this without actively querying the user for information. It can use past information to make an accurate guess at the user's current purpose, which, if wrong, can be corrected by the user and learned from. I discuss how such a system is conceptually designed and then go into the details of how it could be implemented on the Android platform. The purpose of this thesis is to lay out the framework for a context-aware mobile application that can be implemented as a first-stage demonstration for France Telecom/Orange.