Research On Automatic Collaboration Framework For Android App

Massive Mobile Applications(App)provide a variety of convenient services for people's life.However,fragmentation of mobile Apps is reflected in that Apps usually focus on specific function and refuse interacting with each other.This fragmented situation leads to poor user interaction experience,employing several Apps like a flow for completing a specific task.In this paper,we propose a novel Automatic Collaboration Framework for Apps to realize automatic construction and execution of App flow.Firstly,We define an App abstract model,describing inner data of the App with <storage,semantic,syntax> three tuple structure,and explore static analysis technique to convert App into the model.With the abstract model,automatic extraction and injection of App internal data achieve automatic collaborative of multi Apps.Secondly,App flow recommendation and customization constructs flow.The proposed App recommendation method gives full consideration to the potential relationships of Apps and current status of the App flow apart from context features,and merges nowcasting practice of meteorology to relize automatic construction of App flow.In addition,customization provide a personalized and flexible tool for end-users to explicitly link installed Apps together to use.Finally,the recommendation method is verified by the collected user data,and taking travel scenario as an example,we implement a prototype on Android4.3 to show the effect of automatic collaboration.In summary,this paper has designed and implemented an Automatic Collaboration Framework aiming at mobile fragmentation,enabling automatic App flow for complex tasks,improve and enhancing the interactive mode and experience of mobile users.