Where am I? Using mobile sensor data to predict a user’s semantic place with a random forest algorithm

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

Abstract

We use mobile sensor data to predict a mobile phone user’s semantic place, e.g. at home, at work, in a restaurant etc. Such information can be used to feed context-aware systems, that adapt for instance mobile phone settings like energy saving, connection to Internet, volume of ringtones etc. We consider the task of semantic place prediction as classification problem. In this paper we exploit five feature groups: (i) daily patterns, (ii) weekly patterns, (iii) WLAN information, (iv) battery charging state and (v) accelerometer data. We compare the performance of a Random Forest algorithm and two Support Vector Machines, one with an RBF kernel and one with a Pearson VII function based kernel, on a labelled dataset, and analyse the separate performances of the feature groups as well as promising combinations of feature groups. The winning combination of feature groups achieves an accuracy of 0.871 using a Random Forest algorithm on daily patterns and accelerometer data. A detailed analysis reveals that daily patterns are the most discriminative feature group for the given semantic place labels. Combining daily patterns with WLAN information, battery charging state or accelerometer data further improves the performance. The classifiers using these selected combinations perform better than the classifiers using all feature groups. This is especially encouraging for mobile computing, as fewer features mean that less computational power is required for classification.

LanguageEnglish
Title of host publicationMobile and Ubiquitous Systems
Subtitle of host publicationComputing, Networking, and Services - 9th International Conference, MobiQuitous 2012, Revised Selected Papers
EditorsHongbo Jiang, Kan Zheng, Mo Li
PublisherSpringer Verlag Heidelberg
Pages64-75
Number of pages12
ISBN (Print)9783642402371
StatusPublished - 1 Jan 2013
Event9th International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services, MobiQuitous 2012 - Beijing, China
Duration: 12 Dec 201214 Dec 2012

Publication series

NameLecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering
Volume120
ISSN (Print)1867-8211

Conference

Conference9th International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services, MobiQuitous 2012
CountryChina
CityBeijing
Period12/12/1214/12/12

Fingerprint

Accelerometers
Charging (batteries)
Semantics
Wireless local area networks (WLAN)
Mobile phones
Sensors
Classifiers
Mobile computing
Support vector machines
Labels
Energy conservation
Internet

ASJC Scopus subject areas

  • Computer Networks and Communications

Fields of Expertise

  • Information, Communication & Computing

Cite this

Lex, E., Pimas, O., Simon, J., & Pammer-Schindler, V. (2013). Where am I? Using mobile sensor data to predict a user’s semantic place with a random forest algorithm. In H. Jiang, K. Zheng, & M. Li (Eds.), Mobile and Ubiquitous Systems: Computing, Networking, and Services - 9th International Conference, MobiQuitous 2012, Revised Selected Papers (pp. 64-75). (Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering; Vol. 120). Springer Verlag Heidelberg.

Where am I? Using mobile sensor data to predict a user’s semantic place with a random forest algorithm. / Lex, Elisabeth; Pimas, Oliver; Simon, Jörg; Pammer-Schindler, Viktoria.

Mobile and Ubiquitous Systems: Computing, Networking, and Services - 9th International Conference, MobiQuitous 2012, Revised Selected Papers. ed. / Hongbo Jiang; Kan Zheng; Mo Li. Springer Verlag Heidelberg, 2013. p. 64-75 (Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering; Vol. 120).

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

Lex, E, Pimas, O, Simon, J & Pammer-Schindler, V 2013, Where am I? Using mobile sensor data to predict a user’s semantic place with a random forest algorithm. in H Jiang, K Zheng & M Li (eds), Mobile and Ubiquitous Systems: Computing, Networking, and Services - 9th International Conference, MobiQuitous 2012, Revised Selected Papers. Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering, vol. 120, Springer Verlag Heidelberg, pp. 64-75, 9th International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services, MobiQuitous 2012, Beijing, China, 12/12/12.
Lex E, Pimas O, Simon J, Pammer-Schindler V. Where am I? Using mobile sensor data to predict a user’s semantic place with a random forest algorithm. In Jiang H, Zheng K, Li M, editors, Mobile and Ubiquitous Systems: Computing, Networking, and Services - 9th International Conference, MobiQuitous 2012, Revised Selected Papers. Springer Verlag Heidelberg. 2013. p. 64-75. (Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering).
Lex, Elisabeth ; Pimas, Oliver ; Simon, Jörg ; Pammer-Schindler, Viktoria. / Where am I? Using mobile sensor data to predict a user’s semantic place with a random forest algorithm. Mobile and Ubiquitous Systems: Computing, Networking, and Services - 9th International Conference, MobiQuitous 2012, Revised Selected Papers. editor / Hongbo Jiang ; Kan Zheng ; Mo Li. Springer Verlag Heidelberg, 2013. pp. 64-75 (Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering).
@inproceedings{026bec7fa8f54122ae4f424fc40a9d37,
title = "Where am I? Using mobile sensor data to predict a user’s semantic place with a random forest algorithm",
abstract = "We use mobile sensor data to predict a mobile phone user’s semantic place, e.g. at home, at work, in a restaurant etc. Such information can be used to feed context-aware systems, that adapt for instance mobile phone settings like energy saving, connection to Internet, volume of ringtones etc. We consider the task of semantic place prediction as classification problem. In this paper we exploit five feature groups: (i) daily patterns, (ii) weekly patterns, (iii) WLAN information, (iv) battery charging state and (v) accelerometer data. We compare the performance of a Random Forest algorithm and two Support Vector Machines, one with an RBF kernel and one with a Pearson VII function based kernel, on a labelled dataset, and analyse the separate performances of the feature groups as well as promising combinations of feature groups. The winning combination of feature groups achieves an accuracy of 0.871 using a Random Forest algorithm on daily patterns and accelerometer data. A detailed analysis reveals that daily patterns are the most discriminative feature group for the given semantic place labels. Combining daily patterns with WLAN information, battery charging state or accelerometer data further improves the performance. The classifiers using these selected combinations perform better than the classifiers using all feature groups. This is especially encouraging for mobile computing, as fewer features mean that less computational power is required for classification.",
author = "Elisabeth Lex and Oliver Pimas and J{\"o}rg Simon and Viktoria Pammer-Schindler",
year = "2013",
month = "1",
day = "1",
language = "English",
isbn = "9783642402371",
series = "Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering",
publisher = "Springer Verlag Heidelberg",
pages = "64--75",
editor = "Hongbo Jiang and Kan Zheng and Mo Li",
booktitle = "Mobile and Ubiquitous Systems",

}

TY - GEN

T1 - Where am I? Using mobile sensor data to predict a user’s semantic place with a random forest algorithm

AU - Lex,Elisabeth

AU - Pimas,Oliver

AU - Simon,Jörg

AU - Pammer-Schindler,Viktoria

PY - 2013/1/1

Y1 - 2013/1/1

N2 - We use mobile sensor data to predict a mobile phone user’s semantic place, e.g. at home, at work, in a restaurant etc. Such information can be used to feed context-aware systems, that adapt for instance mobile phone settings like energy saving, connection to Internet, volume of ringtones etc. We consider the task of semantic place prediction as classification problem. In this paper we exploit five feature groups: (i) daily patterns, (ii) weekly patterns, (iii) WLAN information, (iv) battery charging state and (v) accelerometer data. We compare the performance of a Random Forest algorithm and two Support Vector Machines, one with an RBF kernel and one with a Pearson VII function based kernel, on a labelled dataset, and analyse the separate performances of the feature groups as well as promising combinations of feature groups. The winning combination of feature groups achieves an accuracy of 0.871 using a Random Forest algorithm on daily patterns and accelerometer data. A detailed analysis reveals that daily patterns are the most discriminative feature group for the given semantic place labels. Combining daily patterns with WLAN information, battery charging state or accelerometer data further improves the performance. The classifiers using these selected combinations perform better than the classifiers using all feature groups. This is especially encouraging for mobile computing, as fewer features mean that less computational power is required for classification.

AB - We use mobile sensor data to predict a mobile phone user’s semantic place, e.g. at home, at work, in a restaurant etc. Such information can be used to feed context-aware systems, that adapt for instance mobile phone settings like energy saving, connection to Internet, volume of ringtones etc. We consider the task of semantic place prediction as classification problem. In this paper we exploit five feature groups: (i) daily patterns, (ii) weekly patterns, (iii) WLAN information, (iv) battery charging state and (v) accelerometer data. We compare the performance of a Random Forest algorithm and two Support Vector Machines, one with an RBF kernel and one with a Pearson VII function based kernel, on a labelled dataset, and analyse the separate performances of the feature groups as well as promising combinations of feature groups. The winning combination of feature groups achieves an accuracy of 0.871 using a Random Forest algorithm on daily patterns and accelerometer data. A detailed analysis reveals that daily patterns are the most discriminative feature group for the given semantic place labels. Combining daily patterns with WLAN information, battery charging state or accelerometer data further improves the performance. The classifiers using these selected combinations perform better than the classifiers using all feature groups. This is especially encouraging for mobile computing, as fewer features mean that less computational power is required for classification.

UR - http://www.scopus.com/inward/record.url?scp=84945984099&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9783642402371

T3 - Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering

SP - 64

EP - 75

BT - Mobile and Ubiquitous Systems

PB - Springer Verlag Heidelberg

ER -