Are apps abusing your smartphone battery?

Recommendation and feedback systems that increase usage experience of smart mobile devices utilize data collection and analysis. In her doctoral thesis for the University of Helsinki Ella Peltonen presents approaches to generate actionable feedback, such as energy consumption and application recommendations. They can be utilized in the mobile devices themselves to benefit the user or when understanding crowds of smartphone users.

Mobile devices, especially smartphones, are an essential part of everyday life. They are used worldwide and across all the demographic groups - they have functions including, but not limited, to communications, game playing, social interactions, maps and navigation, leisure, work, and education. And, with a large on-device sensor base, mobile devices also provide a rich source of data.

“Understanding how these devices are used helps us to increase the knowledge of people's everyday habits, needs, and rituals”, says doctoral candidate Ella Peltonen from the University of Helsinki.

Data collection and analysis can be utilized in different recommendation and feedback systems that further increase usage experience of these smart devices. Crowdsensed computing describes a paradigm where multiple autonomous devices are used together to collect large-scale data. In the case of smartphones, this kind of data can include running and installed applications, different system settings, such as network connection and screen brightness, and various subsystem variables, such as processor and memory usage.

Large-scale autonomously crowdsensed mobile analytics

Ella Peltonen’s doctoral thesis provides an approach to a large-scale crowdsensed mobile analytics. Her data comes thru Carat, a free app with more than 850 000 mobile users that tells you what is using up the battery of your mobile device and whether that is normal. Her work describes procedures for cleaning and preprocessing mobile data collected from real-life conditions, such as current system settings and running applications. It shows how interdependencies between different data items are important to consider when analyzing the smartphone system state as a whole. For example, battery temperature can be affected by various features, such as processor usage, ambient temperature, or their combination.

Interdependencies between different data items are important to know

Peltonen’s research study also provides us with distributed machine learning and statistical analysis methods suitable for analyzing large-scale mobile data. The algorithms, such as the decision tree-based classification and recommendation system, and information analysis methods presented are implemented in the distributed cloud-computing environment Apache Spark. Spark is a popular open-source platform designed for effective distributed data analysis.

Her work also provides approaches to generate actionable feedback, such as energy consumption and application recommendations, which can be utilized in the mobile devices themselves or when understanding large crowds of smartphone users.

“The application areas especially covered in my work are smartphone energy consumption analysis in the case of system settings and subsystem variables, trend-based application recommendation system, and analysis of demographic, geographic, and cultural factors in smartphone usage", she says. "For example, adults living with children have different needs for smartphone apps than single people. On the other hand, students seem to be a quite homogenous group all over the world.”

Management tools but also user questionnaires

To understand smartphone usage as a whole, different procedures are needed for cleaning missing and misleading values and preprocessing information from various sets of variables.

“Analyzing large-scale data sets - rising in size to terabytes - requires understanding of different Big Data management tools, distributed computing environments, and efficient algorithms to perform suitable data analysis and machine learning tasks”, Ella Peltonen explains.

In addition to the autonomous data collection, user questionnaires can be used to provide a wider view to the user community. – Together, these procedures and methodologies aim to provide actionable feedback, such as recommendations and visualizations, for the benefit of smartphone users, researchers, and application development.

Ella Peltonen is defending her doctoral thesis Crowdsensed Mobile Data Analytics on February the 26th at the University of Helsinki.