SmartActions: Context-Aware Mobile Phone Shortcuts

Abstract

Abstract. Mobile phones are often regarded as difficult to use due to their size restrictions. To improve on this, in this paper we described our approach using unsupervised learning to automate common tasks on a mobile phone, thereby requiring less key presses, by means of context-dependent quick-access shortcuts presented in the homescreen of the phone. We also briefly reviewed some of our user study findings, and raised the issue of possible privacy concerns with our implementation.

Where and what: Using smartphones to predict next locations and applications in daily life

“…In the context of Human Computer Interaction, utilizing information from a user's behavior to modify mobile interfaces has been investigated in the past. Vetek et al [26] for example, proposed a method to dynamically improve the rec-ommendation of shortcuts on the home-screen of a smartphone using user context. These authors used an unsupervised clustering based method that uses location data, in the form of GSM Cell IDs to learn user contexts to improve recommendations of these shortcuts.…”

A study on icon arrangement by smartphone users

“…They evaluated different approaches to provide users with shortcuts to common tasks on their mobile devices. Vetek et al [26] presented SmartActions: a contextaware menu that automatically creates context-aware shortcuts to phone functionality based on unsupervised learning.…”

“…This is of particular interest for mobile menus, since the tasks of mobile users [1] and the apps they use [5,8] are perpetually changing, and the design of context-aware menus is a topic of current research (cf. [4,13,22,26]). However, in contrast to pre-designed menus, smartphone launch-ers are highly customized by the very users themselves.…”

Human interaction discovery in smartphone proximity networks

“…Prediction is a very important task in context-aware mobile applications [6,37,42]. Our main goal is to validate the learning capability of the proposed model by computing the likelihood on unseen data.…”