Smartphones and applications have become widespread more and more. Thus, using the hardware and software of users’ mobile phones, we can get a large amount of personal data, in which a large part is about the user’s application usage patterns. By transforming and extracting these data, we can get user preferences, and provide personalized services and improve the experience for users. In a detailed way, studying application usage pattern benefits a variety of advantages such as precise bandwidth allocation, App launch acceleration, etc. However, the first thing to achieve the above advantages is to predict the next application accurately. In this paper, we propose AHNEAP, a novel network embedding based framework for predicting the next App to be used by characterizing the context information before one specific App being launched. AHNEAP transforms the historical App usage records in physical spaces to a large attributed heterogeneous network which contains three node types, three edges, and several attributes like App type, the day of the week. Then, the representation learning process is conducted. Finally, the App usage prediction problem was defined as a link prediction problem, realized by a simple neural network. Experiments on the LiveLab project dataset demonstrate the effectiveness of our framework which outperforms the three baseline methods for each tested user.

In this paper, we propose a multi-layer mobile application (app) scheduling method to extend the capability of low-end Android devices. With a quantitative analysis, we find that the increase of installed apps will negatively affect quality of experience (QoE) of the user, e.g., the action response time of a mobile device, by producing more periodically or irregularly background tasks. On the other hand, the user tends to install more apps than needed in case they could play a role someday, as indicated by the consumer app usage statistics. When the storage is running out, being forced to choose an app to uninstall due to space budget is a painful experience for the user. This contradiction is intensified for low-end devices due to limited resources. We try to reduce this dilemma by a multi-layer app scheduling (MAS) schema, along with a cloud service. For the first layer, we utilize the “freeze”feature of Android to prevent non-essential background activities. For the second layer, it is a network scheduler, which automatically schedules the available apps, together with their data, between local and cloud according to user's personal policy generated by big data analysis. By dynamically scheduling the apps among three states, QoE of a low-end Android device is improved. At the same time, with the help of an app state recovery mechanism, the user can directly access a large number of apps provided by the cloud with consistent app view. Experimental results on a low-end smartphone, i.e., Samsung Galaxy ON5, and a smart watch based on Newton2_Plus wearable development board show the benefits of the proposed MAS schema.