Mobile devices have become an increasingly ubiquitous part of our everyday life. We use mobile services to perform a broad range of tasks (e.g., booking travel or conducting remote office work), leading to often lengthy interactions with several distinct apps and services. Existing mobile systems handle mostly simple user needs, where a single app is taken as the unit of interaction. To understand users’ expectations and to provide context-aware services, it is important to model users’ interactions with their performed task in mind. To provide a comprehensive picture of common mobile tasks, we first conduct a small-scale user study to understand annotated mobile tasks in-depth, while we demonstrate that by using a set of features (temporal, similarity, and log sequence), we can identify if a pair of app usage belong to the same task effectively. Secondly, the proposed best task detection model is applied to a large-scale data set of commercial mobile app usage logs to infer characteristics of complex (multi-app) mobile tasks in the wild. By applying an unsupervised learning framework, we discover common mobile task types that span multiple apps based on various extracted characteristics. We observe that users generally perform 17 common tasks with 47 sub-tasks, ranging from “social media browsing” to “dining out” and “family entertainments”. Finally, we demonstrate that we can predict the next complex mobile task that users are likely to perform by leveraging features from the historically inferred mobile tasks and user contexts. Our work facilitates an in-depth understanding of mobile tasks at scale, enabling applications for promoting task-aware services.

Demographic attributes are a key factor in marketing products and services, which enable a business owner to find the ideal customer. Users' app usage behaviors could reveal rich clues regarding their personal attributes since they always determine what apps to use depending on their personal needs and interests. Prior studies [1, 2] have tried to predict users' gender and age through their app usage behavior. However, most of the existing methods for users' demographic prediction are straightforward, simply using popular used apps or app usage frequency as features, without considering the internal semantic relationship of apps usage.

User engagement is crucial to the long-term success of a mobile app. Several metrics, such as dwell time, have been used for measuring user engagement. However, how to effectively predict user engagement in the context of mobile apps is still an open research question. For example, do the mobile usage contexts (e.g., time of day) in which users access mobile apps impact their dwell time? Answers to such questions could help mobile operating system and publishers to optimize advertising and service placement. In this article, we first conduct an empirical study for assessing how user characteristics, temporal features, and the short/long-term contexts contribute to gains in predicting users’ app dwell time on the population level. The comprehensive analysis is conducted on large app usage logs collected through a mobile advertising company. The dataset covers more than 12K anonymous users and 1.3 million log events. Based on the analysis, we further investigate a novel mobile app engagement prediction problem—can we predict simultaneously what app the user will use next and how long he/she will stay on that app? We propose several strategies for this joint prediction problem and demonstrate that our model can improve the performance significantly when compared with the state-of-the-art baselines. Our work can help mobile system developers in designing a better and more engagement-aware mobile app user experience.

Mobile devices have become an increasingly ubiquitous part of our everyday life. We use mobile services to perform a broad range of tasks (e.g. booking travel or office work), leading to often lengthy interactions within distinct apps and services. Existing mobile systems handle mostly simple user needs, where a single app is taken as the unit of interaction. To understand users' expectations and to provide context-aware services, it is important to model users' interactions in the task space. In this work, we first propose and evaluate a method for the automated segmentation of users' app usage logs into task units. We focus on two problems: (i) given a sequential pair of app usage logs, identify if there exists a task boundary, and (ii) given any pair of two app usage logs, identify if they belong to the same task. We model these as classification problems that use features from three aspects of app usage patterns: temporal, similarity, and log sequence. Our classifiers improve on traditional timeout segmentation, achieving over 89% performance for both problems. Secondly, we use our best task classifier on a large-scale data set of commercial mobile app usage logs to identify common tasks. We observe that users' performed common tasks ranging from regular information checking to entertainment and booking dinner. Our proposed task identification approach provides the means to evaluate mobile services and applications with respect to task completion.

Smartphones utilize context signals, such as time and location, to predict users' app usage tailored to individual users. To be effective, such personalization relies on access to sufficient information about each user's behavioral habits. For new users, the behavior information may be sparse or non-existent. To handle these cases, app category usage prediction approaches can employ signals from users who are similar along one or more dimensions, i.e., those in the same cohort. In this paper, we describe a characterization and evaluation of the use of such cohort modeling to enhance app category usage prediction. We experiment with pre-defined cohorts from three taxonomies - demographics, psychographics, and behavioral patterns - independently and in combination. We also evaluate various approaches to assign users into the corresponding cohorts. We show, through extensive experiments with large-scale mobile app usage logs from a mobile advertising company, that leveraging cohort behavior can yield significant prediction performance gains than when using the personalized signals at the individual prediction level. In addition, compared to the personalized model, the cohort-based approach can significantly alleviate the cold-start problem, achieving strong predictive performance even with limited amount of user interactions.