A Personalized Integration Framework For Mobile Applications And Its Implementation On The Android Platform

The explosive growth of mobile applications has brought new opportunities and challenges to both developers and users. From the perspective of developers, it is now possible to build new mobile applications rapidly based on massive existing ones. If so, it is no doubt that this mechanism would shorten the development cycle, reduce the risk and ensure product quality. However, such high-level integration is largely ignored by most existing developing tools whose focuses are on the API level. Meanwhile, massive mobile applications make it urgent to solve the problem of selecting mobile applications which meet the individual requirements. In the scenario of reusing existing mobile applications sequentially, the function sequence will be accomplished by application sequences which are more suitable for users if it is possible to recommend an ideal application sequence. This mechanism will greatly enhance the user experience and reduce side effects causing by application overload problem. Currently, recommending application sequences for users still remains as an open problem while recommending a single personalized application has been widely explored.In order to solve the above problems, we propose a personalized integration framework for mobile applications, and the main contributions are summarized as follows:●We propose a personalized integration framework for mobile applications. Developers only need to edit the script file that describes the function sequences. This framework offers the engine to parse the script file and provide run-time support. Meanwhile, this framework offers a personalized recommendation module to recommend application sequences which meet users’ requirements, so that it could solve the application overload problem. ●We propose the programming language IPEL (Intent Process Execution Language) and implement the execution engine. IPEL is used to solve the problem of descripting the function sequences and facilitate the developers to build applications more naturally. IPEL adopts the loose coupling structure, so developers do not have to specify the specific application in the script file. The IPEL-grammar-parsing engine would bind specific application components at run time dynamically. This mechanism makes the personalized recommendation of mobile application sequences becomes possible.●We propose the Collaborative Hidden Markov Model and an algorithm to recommend application sequences which meet users’ requirements. Different users always prefer to choose different application sequences to accomplish the same function sequence due to their different preferences. We improve the traditional Hidden Markov Model and propose the Collaborative HMM. Also, we give the algorithm to recommend application sequence based on the Collaborative HMM. The function sequence will be accomplished by application sequences which are more suitable for users.●We implement the framework based on the Android platform due to its huge market and openness. This framework is constructed of Client/Server model. The client offers the SDK which is easily used for developers. The server is deployed on the Sina App Engine, which generates the recommendation result of application sequence in real time and return them to users.