Association Rule Discovery from Collaborative Mobile Data

Show full item record

Title: Association Rule Discovery from Collaborative Mobile Data
Author: Hamberg, Jiri
Contributor: University of Helsinki, Faculty of Science, Department of Computer Science
Publisher: Helsingin yliopisto
Date: 2018
Language: eng
Thesis level: master's thesis
Discipline: Computer science
Abstract: Sophisticated mobile devices have rapidly become essential tools for various daily activities of billions of people worldwide. Subsequently, the demand for longer battery lives is constantly increasing. The Carat project is advancing the understanding of mobile energy consumption by using collaborative mobile data to estimate and model energy consumption of mobile devices. This thesis presents a method for estimating mobile application energy consumption from mobile device system settings and context factors using association rules. These settings and factors include CPU usage, device travel distance, battery temperature, battery voltage, screen brightness, used mobile networking technology, network type, WiFi signal strength, and WiFi connection speed. The association rules are mined using Apache Spark cluster-computing framework from collaborative mobile data collected by the Carat project. Additionally, this thesis presents a prototype of a web based API for discovering these association rules. The web service integrates Apache Spark based analysis engine with a user friendly front-end allowing an aggregated view of the dataset to be accessible without revealing data of individual participants of the Carat project. This thesis shows that association rules can be used effectively in modelling mobile device energy consumption. Example rules are presented and the performance of the implementation is evaluated experimentally.

Files in this item

Total number of downloads: Loading...

Files Size Format View
jiri_hamberg_thesis.pdf 1.551Mb PDF View/Open

This item appears in the following Collection(s)

Show full item record