MLStar

Machine learning in energy profile estimation of android apps

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Improving the energy efficiency of smartphones is critical for increasing the utility that they provide to the users. With most mobile operating systems, users are responsible for managing their phone’s battery efficiency by utilizing the various settings provided by the operating system, as well as selecting energy-efficient apps. However, current app marketplaces do not provide users with information about app energy efficiency, which makes it challenging for the user to make informed decision when selecting an app. This paper presents a novel machine learning approach to estimate app energy efficiency by utilizing textual information available in the Google Play store such as an app’s description, user reviews, as well as system permissions. Our detailed analysis of the resulting system shows that hardware permissions, app description, and user reviews correlate well with energy efficiency ratings. We evaluate five models that represent popular classes of machine learning algorithms in their ability to predict energy efficiency ratings. Finally, we compare our approach to gold truth ratings obtained by the actual energy profiling of the app, demonstrating that the proposed system is able to estimate an app’s energy efficiency within less than 1 point on the 1 – 5 scale provided by the profiler, without requiring any kind of profiling.

Original languageEnglish (US)
Title of host publicationProceedings of the 15th EAI International Conference on Mobile and Ubiquitous Systems
Subtitle of host publicationComputing, Networking and Services, Mobiquitous 2018
PublisherAssociation for Computing Machinery
Pages216-225
Number of pages10
ISBN (Electronic)9781450360937
DOIs
StatePublished - Nov 5 2018
Event15th EAI International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services, Mobiquitous 2018 - New York, United States
Duration: Nov 5 2018Nov 7 2018

Other

Other15th EAI International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services, Mobiquitous 2018
CountryUnited States
CityNew York
Period11/5/1811/7/18

Fingerprint

Application programs
Learning systems
Energy efficiency
Computer systems
Android (operating system)
Smartphones
Learning algorithms
Computer hardware
Gold

Keywords

  • Energy
  • Machine learning
  • Mobile

ASJC Scopus subject areas

  • Human-Computer Interaction
  • Computer Networks and Communications
  • Computer Vision and Pattern Recognition
  • Software

Cite this

Gaska, B., Gniady, C., & Surdeanu, M. (2018). MLStar: Machine learning in energy profile estimation of android apps. In Proceedings of the 15th EAI International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services, Mobiquitous 2018 (pp. 216-225). Association for Computing Machinery. https://doi.org/10.1145/3286978.3287011

MLStar : Machine learning in energy profile estimation of android apps. / Gaska, Benjamin; Gniady, Christopher; Surdeanu, Mihai.

Proceedings of the 15th EAI International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services, Mobiquitous 2018. Association for Computing Machinery, 2018. p. 216-225.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Gaska, B, Gniady, C & Surdeanu, M 2018, MLStar: Machine learning in energy profile estimation of android apps. in Proceedings of the 15th EAI International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services, Mobiquitous 2018. Association for Computing Machinery, pp. 216-225, 15th EAI International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services, Mobiquitous 2018, New York, United States, 11/5/18. https://doi.org/10.1145/3286978.3287011
Gaska B, Gniady C, Surdeanu M. MLStar: Machine learning in energy profile estimation of android apps. In Proceedings of the 15th EAI International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services, Mobiquitous 2018. Association for Computing Machinery. 2018. p. 216-225 https://doi.org/10.1145/3286978.3287011
Gaska, Benjamin ; Gniady, Christopher ; Surdeanu, Mihai. / MLStar : Machine learning in energy profile estimation of android apps. Proceedings of the 15th EAI International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services, Mobiquitous 2018. Association for Computing Machinery, 2018. pp. 216-225
@inproceedings{20b16287541345918d3bf0988ba48b53,
title = "MLStar: Machine learning in energy profile estimation of android apps",
abstract = "Improving the energy efficiency of smartphones is critical for increasing the utility that they provide to the users. With most mobile operating systems, users are responsible for managing their phone’s battery efficiency by utilizing the various settings provided by the operating system, as well as selecting energy-efficient apps. However, current app marketplaces do not provide users with information about app energy efficiency, which makes it challenging for the user to make informed decision when selecting an app. This paper presents a novel machine learning approach to estimate app energy efficiency by utilizing textual information available in the Google Play store such as an app’s description, user reviews, as well as system permissions. Our detailed analysis of the resulting system shows that hardware permissions, app description, and user reviews correlate well with energy efficiency ratings. We evaluate five models that represent popular classes of machine learning algorithms in their ability to predict energy efficiency ratings. Finally, we compare our approach to gold truth ratings obtained by the actual energy profiling of the app, demonstrating that the proposed system is able to estimate an app’s energy efficiency within less than 1 point on the 1 – 5 scale provided by the profiler, without requiring any kind of profiling.",
keywords = "Energy, Machine learning, Mobile",
author = "Benjamin Gaska and Christopher Gniady and Mihai Surdeanu",
year = "2018",
month = "11",
day = "5",
doi = "10.1145/3286978.3287011",
language = "English (US)",
pages = "216--225",
booktitle = "Proceedings of the 15th EAI International Conference on Mobile and Ubiquitous Systems",
publisher = "Association for Computing Machinery",

}

TY - GEN

T1 - MLStar

T2 - Machine learning in energy profile estimation of android apps

AU - Gaska, Benjamin

AU - Gniady, Christopher

AU - Surdeanu, Mihai

PY - 2018/11/5

Y1 - 2018/11/5

N2 - Improving the energy efficiency of smartphones is critical for increasing the utility that they provide to the users. With most mobile operating systems, users are responsible for managing their phone’s battery efficiency by utilizing the various settings provided by the operating system, as well as selecting energy-efficient apps. However, current app marketplaces do not provide users with information about app energy efficiency, which makes it challenging for the user to make informed decision when selecting an app. This paper presents a novel machine learning approach to estimate app energy efficiency by utilizing textual information available in the Google Play store such as an app’s description, user reviews, as well as system permissions. Our detailed analysis of the resulting system shows that hardware permissions, app description, and user reviews correlate well with energy efficiency ratings. We evaluate five models that represent popular classes of machine learning algorithms in their ability to predict energy efficiency ratings. Finally, we compare our approach to gold truth ratings obtained by the actual energy profiling of the app, demonstrating that the proposed system is able to estimate an app’s energy efficiency within less than 1 point on the 1 – 5 scale provided by the profiler, without requiring any kind of profiling.

AB - Improving the energy efficiency of smartphones is critical for increasing the utility that they provide to the users. With most mobile operating systems, users are responsible for managing their phone’s battery efficiency by utilizing the various settings provided by the operating system, as well as selecting energy-efficient apps. However, current app marketplaces do not provide users with information about app energy efficiency, which makes it challenging for the user to make informed decision when selecting an app. This paper presents a novel machine learning approach to estimate app energy efficiency by utilizing textual information available in the Google Play store such as an app’s description, user reviews, as well as system permissions. Our detailed analysis of the resulting system shows that hardware permissions, app description, and user reviews correlate well with energy efficiency ratings. We evaluate five models that represent popular classes of machine learning algorithms in their ability to predict energy efficiency ratings. Finally, we compare our approach to gold truth ratings obtained by the actual energy profiling of the app, demonstrating that the proposed system is able to estimate an app’s energy efficiency within less than 1 point on the 1 – 5 scale provided by the profiler, without requiring any kind of profiling.

KW - Energy

KW - Machine learning

KW - Mobile

UR - http://www.scopus.com/inward/record.url?scp=85060004462&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85060004462&partnerID=8YFLogxK

U2 - 10.1145/3286978.3287011

DO - 10.1145/3286978.3287011

M3 - Conference contribution

SP - 216

EP - 225

BT - Proceedings of the 15th EAI International Conference on Mobile and Ubiquitous Systems

PB - Association for Computing Machinery

ER -