Deriving User Preferences of Mobile Apps from Their Management Activities

Abstract

App marketplaces host millions of mobile apps that are downloaded billions of times. Investigating how people manage mobile apps in their everyday lives creates a unique opportunity to understand the behavior and preferences of mobile device users, infer the quality of apps, and improve user experience. Existing literature provides very limited knowledge about app management activities, due to the lack of app usage data at scale. This article takes the initiative to analyze a very large app management log collected through a leading Android app marketplace. The dataset covers 5 months of detailed downloading, updating, and uninstallation activities, which involve 17 million anonymized users and 1 million apps. We present a surprising finding that the metrics commonly used to rank apps in app stores do not truly reflect the users' real attitudes. We then identify behavioral patterns from the app management activities that more accurately indicate user preferences of an app even when no explicit rating is available. A systematic statistical analysis is designed to evaluate machine learning models that are trained to predict user preferences using these behavioral patterns, which features an inverse probability weighting method to correct the selection biases in the training process.

Characterizing Smartphone Usage Patterns from Millions of Android Users

“…Many findings of the analysis lead to potential research questions or opportunities. In fact, some research topics such as optimizing an app store's performance, predicting an app's quality and popularity [31], [68], and prioritizing the fragmented Android devices for specific apps [51], have already been explored based on our dataset.…”

Graph-based Method for App Usage Prediction with Attributed Heterogeneous Network Embedding

“…Regarding the problem of App prediction about the LiveLab dataset, there are not many related experiments. Liu et al [35] mainly use data to study user application usage patterns, such as frequency of application usage, without focusing on App prediction. The main research point of the article [4] is the order about application startup.…”

Data-Driven Elicitation of Quality Requirements in Agile Companies

“…Implicit feedback can be considered as an alternative/complementary data source for requirements elicitation [15]. As an example, [19] exploits implicit feedback but does not aim at generating QRs but at discovering user preferences and usage patterns. The SUPERSEDE data-driven approach [20] combines both explicit and implicit end-user feedback with other sources like run-time monitors to detect and address different kinds of issues: bugs, new features, QoS violations.…”