Practical prediction and prefetch for faster access to applications on mobile phones

Abstract

Mobile phones have evolved from communication devices to indispensable accessories with access to real-time content. The increasing reliance on dynamic content comes at the cost of increased latency to pull the content from the Internet before the user can start using it. While prior work has explored parts of this problem, they ignore the bandwidth costs of prefetching, incur significant training overhead, need several sensors to be turned on, and do not consider practical systems issues that arise from the limited background processing capability supported by mobile operating systems. In this paper, we make app prefetch practical on mobile phones. Our contributions are twofold. First, we design an app prediction algorithm, APPM, that requires no prior training, adapts to usage dynamics, predicts not only which app will be used next but also when it will be used, and provides high accuracy without requiring additional sensor context. Second, we perform parallel prefetch on screen unlock, a mechanism that leverages the benefits of prediction while operating within the constraints of mobile operating systems. Our experiments are conducted on long-term traces, live deployments on the Android Play Market, and user studies, and show that we outperform prior approaches to predicting app usage, while also providing practical ways to prefetch application content on mobile phones.

Personal Tracking of Screen Time on Digital Devices

“…In other cases, particularly where an app is to be released through an app store and no incentive is given for installing the app, an interface has been provided (and indeed the interface serves as the incentive to install and retain the app [7]). Böhmer et al [4] and Parate et al [37] created recommender widgets for users to launch apps, Ferreira et al [16] and Athukorala et al [2] created battery tracking software for end users, and Wagner et al [49] created a personal analytics style interface offering a range of information about the use of one's device.…”

Predicting Replacement of Smartphones with Mobile App Usage

“…Firstly, the studies on the usage analysis focus on understanding when and where apps are used in mobile phones [1]. The contextual usage patterns can then be leveraged for apps prediction (or recommendation) [8,10], which usually guides the development of adaptive user interfaces [5]. Moreover, a number of mobile recommender systems and target advertising engines have been designed based on the app usage analysis [9,13].…”

“…There are mainly two aspects of this interesting problem that the existing research has been done. On the one hand, the usage prediction and classification of mobile apps themselves [8,10,14], which can help users to search and launch apps efficiently. On the other hand, to exploit the app usage for developing other business intelligence services, such as recommendation, churn controlling, target advertising [13].…”

Smartphone application launch with smarter scheduling

“…Yan et al [11] and Parate et al [10] propose systems predicting application launch to reduce the launch delay. However, mispredictions of the proposed approaches will lead to significant memory and energy overhead.…”