CAMF - Context-Aware Machine Learning Framework for Android

A.I. Wang and Q.K. Ahmad (Norway)

Keywords

Software Design, Mobile and Wireless Computing, Context-aware Computing, and Android platform

Abstract

Context-aware computing is a promising approach for utilizing the characteristics of mobile computing: communication, mobility and portability. By combining machine learning and context-aware computing, we can provide proactive services based on the users’ usage patterns of the mobile device combined with the environmental context of the user. Android has become a popular mobile platform, which have addressed context-awareness from day one through hardware and software support for sensor and context management. In this paper, we have evaluated the Android platform support for context awareness and identified some shortcomings. Further, we have designed a Context-Aware Machine learning Framework (CAMF) for the Android platform that addresses these shortcomings as well as incorporating machine learning. We also demonstrate the usefulness of the framework through the implementation of an application that monitors the applications a user is running on an Android device along with the environmental context the applications are running in. This information can be used to proactively launch Android applications when the context is appropriate. Finally, the paper evaluates the CAMF framework and our context-aware application AppL.

Important Links:



Go Back