Next place predictions based on user mobility traces

Bhaskar Prabhala, Thomas La Porta

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

Abstract

Collecting user's current location(s) and place-to-place transitions, predicting future destinations, equipping users with location sensitive information, and handling relevant communication requests are core ingredients of new generation of service provider applications on mobile devices. Periodic place-to-place transitions are inherent in human movements. Next place predictions are the atomic units in constructing end-to-end user mobility trajectories based on historical trace data. We make next place predictions by recognizing and utilizing periodicity in user mobility traces. We start with a baseline of the user's current place, start time, and end time to predict the next place. We demonstrate the efficiency of our algorithms through aggregated average prediction accuracies across all users over a large set of diverse participants. We improve these predictions through existing semantic information in the trace data sets, deduced place semantics, and other temporal considerations from the trace data.

Original languageEnglish (US)
Title of host publication2015 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2015
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages93-94
Number of pages2
ISBN (Electronic)9781467371315
DOIs
StatePublished - Aug 4 2015
EventIEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2015 - Hong Kong, Hong Kong
Duration: Apr 26 2015May 1 2015

Publication series

NameProceedings - IEEE INFOCOM
Volume2015-August
ISSN (Print)0743-166X

Other

OtherIEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2015
CountryHong Kong
CityHong Kong
Period4/26/155/1/15

Fingerprint

Semantics
Mobile devices
Trajectories
Communication

All Science Journal Classification (ASJC) codes

  • Computer Science(all)
  • Electrical and Electronic Engineering

Cite this

Prabhala, B., & La Porta, T. (2015). Next place predictions based on user mobility traces. In 2015 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2015 (pp. 93-94). [7179359] (Proceedings - IEEE INFOCOM; Vol. 2015-August). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/INFCOMW.2015.7179359
Prabhala, Bhaskar ; La Porta, Thomas. / Next place predictions based on user mobility traces. 2015 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2015. Institute of Electrical and Electronics Engineers Inc., 2015. pp. 93-94 (Proceedings - IEEE INFOCOM).
@inproceedings{a1b89670f7494e3eac18a11eaacbe46a,
title = "Next place predictions based on user mobility traces",
abstract = "Collecting user's current location(s) and place-to-place transitions, predicting future destinations, equipping users with location sensitive information, and handling relevant communication requests are core ingredients of new generation of service provider applications on mobile devices. Periodic place-to-place transitions are inherent in human movements. Next place predictions are the atomic units in constructing end-to-end user mobility trajectories based on historical trace data. We make next place predictions by recognizing and utilizing periodicity in user mobility traces. We start with a baseline of the user's current place, start time, and end time to predict the next place. We demonstrate the efficiency of our algorithms through aggregated average prediction accuracies across all users over a large set of diverse participants. We improve these predictions through existing semantic information in the trace data sets, deduced place semantics, and other temporal considerations from the trace data.",
author = "Bhaskar Prabhala and {La Porta}, Thomas",
year = "2015",
month = "8",
day = "4",
doi = "10.1109/INFCOMW.2015.7179359",
language = "English (US)",
series = "Proceedings - IEEE INFOCOM",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "93--94",
booktitle = "2015 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2015",
address = "United States",

}

Prabhala, B & La Porta, T 2015, Next place predictions based on user mobility traces. in 2015 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2015., 7179359, Proceedings - IEEE INFOCOM, vol. 2015-August, Institute of Electrical and Electronics Engineers Inc., pp. 93-94, IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2015, Hong Kong, Hong Kong, 4/26/15. https://doi.org/10.1109/INFCOMW.2015.7179359

Next place predictions based on user mobility traces. / Prabhala, Bhaskar; La Porta, Thomas.

2015 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2015. Institute of Electrical and Electronics Engineers Inc., 2015. p. 93-94 7179359 (Proceedings - IEEE INFOCOM; Vol. 2015-August).

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

TY - GEN

T1 - Next place predictions based on user mobility traces

AU - Prabhala, Bhaskar

AU - La Porta, Thomas

PY - 2015/8/4

Y1 - 2015/8/4

N2 - Collecting user's current location(s) and place-to-place transitions, predicting future destinations, equipping users with location sensitive information, and handling relevant communication requests are core ingredients of new generation of service provider applications on mobile devices. Periodic place-to-place transitions are inherent in human movements. Next place predictions are the atomic units in constructing end-to-end user mobility trajectories based on historical trace data. We make next place predictions by recognizing and utilizing periodicity in user mobility traces. We start with a baseline of the user's current place, start time, and end time to predict the next place. We demonstrate the efficiency of our algorithms through aggregated average prediction accuracies across all users over a large set of diverse participants. We improve these predictions through existing semantic information in the trace data sets, deduced place semantics, and other temporal considerations from the trace data.

AB - Collecting user's current location(s) and place-to-place transitions, predicting future destinations, equipping users with location sensitive information, and handling relevant communication requests are core ingredients of new generation of service provider applications on mobile devices. Periodic place-to-place transitions are inherent in human movements. Next place predictions are the atomic units in constructing end-to-end user mobility trajectories based on historical trace data. We make next place predictions by recognizing and utilizing periodicity in user mobility traces. We start with a baseline of the user's current place, start time, and end time to predict the next place. We demonstrate the efficiency of our algorithms through aggregated average prediction accuracies across all users over a large set of diverse participants. We improve these predictions through existing semantic information in the trace data sets, deduced place semantics, and other temporal considerations from the trace data.

UR - http://www.scopus.com/inward/record.url?scp=84943254715&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84943254715&partnerID=8YFLogxK

U2 - 10.1109/INFCOMW.2015.7179359

DO - 10.1109/INFCOMW.2015.7179359

M3 - Conference contribution

AN - SCOPUS:84943254715

T3 - Proceedings - IEEE INFOCOM

SP - 93

EP - 94

BT - 2015 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2015

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Prabhala B, La Porta T. Next place predictions based on user mobility traces. In 2015 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2015. Institute of Electrical and Electronics Engineers Inc. 2015. p. 93-94. 7179359. (Proceedings - IEEE INFOCOM). https://doi.org/10.1109/INFCOMW.2015.7179359