‘Surprise’ Detection in Human Pattern-of-Life Behaviour

Yazan K. Qarout, David Lowe

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

Abstract

In smart city data analytics, the observation of human behavioural characteristics from automated sensor analysis is a useful tool to help identify common sub-clusters and anomalous activity relative to the population’s pattern-of-life dynamic. Avoiding the use of video and mobile phone active applications, because of ethical and privacy concerns, we are more interested in what information can be extracted from lower sensing modalities, such as basic tracking data as obtained from time course GPS signals for example. However, trajectory data has its own technical problems; primarily sparsity, non-uniform sampling, usually unlabelled and sporadic sensor data location errors. In this paper we analyse basic track data obtained from consenting subjects using an approach which combines both a direct time series modelling framework, and a probabilistic framework; the latter derived from an acknowledgement that the residual errors in modelling are never Gaussian distributed for real world data due to our uncertainties in data, models and observations. We construct a hybrid dissimilarity representation of the problem to visualise anomalous individuals without needing extreme value statistics (i.e.. behaving differently to the common consensus trajectories) which also reveals preliminary insights into population sub-clusters. In disaster management situations, or pre-disaster planning, this information can help in quickly identifying dangerous behaviour and predicting dangerous behavioural sub-grouping in population movement prompting fast and effective action.
Original languageEnglish
Title of host publication2018 5th International Conference on Information and Communication Technologies for Disaster Management (ICT-DM)
PublisherIEEE
Pages1-8
ISBN (Electronic)978-1-5386-6638-8
ISBN (Print)978-1-5386-6639-5
DOIs
Publication statusPublished - 7 Feb 2019
Event2018 5th International Conference on Information and Communication Technologies for Disaster Management (ICT-DM) - Sendai, Japan
Duration: 4 Dec 20187 Dec 2018

Conference

Conference2018 5th International Conference on Information and Communication Technologies for Disaster Management (ICT-DM)
Period4/12/187/12/18

Fingerprint

Disasters
Trajectories
Sensors
Mobile phones
Data structures
Global positioning system
Time series
Statistics
Sampling
Planning
disaster
trajectory
sensor
disaster management
detection
Surprise
population development
grouping
modeling
time series

Keywords

  • Pattern-of-Life
  • data visualisation
  • residual analysis
  • trajectory analysis

Cite this

Qarout, Y. K., & Lowe, D. (2019). ‘Surprise’ Detection in Human Pattern-of-Life Behaviour. In 2018 5th International Conference on Information and Communication Technologies for Disaster Management (ICT-DM) (pp. 1-8). [8636382] IEEE. https://doi.org/10.1109/ICT-DM.2018.8636382
Qarout, Yazan K. ; Lowe, David. / ‘Surprise’ Detection in Human Pattern-of-Life Behaviour. 2018 5th International Conference on Information and Communication Technologies for Disaster Management (ICT-DM). IEEE, 2019. pp. 1-8
@inproceedings{6aa1036f9a1242149b38a898a8352804,
title = "‘Surprise’ Detection in Human Pattern-of-Life Behaviour",
abstract = "In smart city data analytics, the observation of human behavioural characteristics from automated sensor analysis is a useful tool to help identify common sub-clusters and anomalous activity relative to the population’s pattern-of-life dynamic. Avoiding the use of video and mobile phone active applications, because of ethical and privacy concerns, we are more interested in what information can be extracted from lower sensing modalities, such as basic tracking data as obtained from time course GPS signals for example. However, trajectory data has its own technical problems; primarily sparsity, non-uniform sampling, usually unlabelled and sporadic sensor data location errors. In this paper we analyse basic track data obtained from consenting subjects using an approach which combines both a direct time series modelling framework, and a probabilistic framework; the latter derived from an acknowledgement that the residual errors in modelling are never Gaussian distributed for real world data due to our uncertainties in data, models and observations. We construct a hybrid dissimilarity representation of the problem to visualise anomalous individuals without needing extreme value statistics (i.e.. behaving differently to the common consensus trajectories) which also reveals preliminary insights into population sub-clusters. In disaster management situations, or pre-disaster planning, this information can help in quickly identifying dangerous behaviour and predicting dangerous behavioural sub-grouping in population movement prompting fast and effective action.",
keywords = "Pattern-of-Life, data visualisation, residual analysis, trajectory analysis",
author = "Qarout, {Yazan K.} and David Lowe",
year = "2019",
month = "2",
day = "7",
doi = "10.1109/ICT-DM.2018.8636382",
language = "English",
isbn = "978-1-5386-6639-5",
pages = "1--8",
booktitle = "2018 5th International Conference on Information and Communication Technologies for Disaster Management (ICT-DM)",
publisher = "IEEE",
address = "United States",

}

Qarout, YK & Lowe, D 2019, ‘Surprise’ Detection in Human Pattern-of-Life Behaviour. in 2018 5th International Conference on Information and Communication Technologies for Disaster Management (ICT-DM)., 8636382, IEEE, pp. 1-8, 2018 5th International Conference on Information and Communication Technologies for Disaster Management (ICT-DM), 4/12/18. https://doi.org/10.1109/ICT-DM.2018.8636382

‘Surprise’ Detection in Human Pattern-of-Life Behaviour. / Qarout, Yazan K.; Lowe, David.

2018 5th International Conference on Information and Communication Technologies for Disaster Management (ICT-DM). IEEE, 2019. p. 1-8 8636382.

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

TY - GEN

T1 - ‘Surprise’ Detection in Human Pattern-of-Life Behaviour

AU - Qarout, Yazan K.

AU - Lowe, David

PY - 2019/2/7

Y1 - 2019/2/7

N2 - In smart city data analytics, the observation of human behavioural characteristics from automated sensor analysis is a useful tool to help identify common sub-clusters and anomalous activity relative to the population’s pattern-of-life dynamic. Avoiding the use of video and mobile phone active applications, because of ethical and privacy concerns, we are more interested in what information can be extracted from lower sensing modalities, such as basic tracking data as obtained from time course GPS signals for example. However, trajectory data has its own technical problems; primarily sparsity, non-uniform sampling, usually unlabelled and sporadic sensor data location errors. In this paper we analyse basic track data obtained from consenting subjects using an approach which combines both a direct time series modelling framework, and a probabilistic framework; the latter derived from an acknowledgement that the residual errors in modelling are never Gaussian distributed for real world data due to our uncertainties in data, models and observations. We construct a hybrid dissimilarity representation of the problem to visualise anomalous individuals without needing extreme value statistics (i.e.. behaving differently to the common consensus trajectories) which also reveals preliminary insights into population sub-clusters. In disaster management situations, or pre-disaster planning, this information can help in quickly identifying dangerous behaviour and predicting dangerous behavioural sub-grouping in population movement prompting fast and effective action.

AB - In smart city data analytics, the observation of human behavioural characteristics from automated sensor analysis is a useful tool to help identify common sub-clusters and anomalous activity relative to the population’s pattern-of-life dynamic. Avoiding the use of video and mobile phone active applications, because of ethical and privacy concerns, we are more interested in what information can be extracted from lower sensing modalities, such as basic tracking data as obtained from time course GPS signals for example. However, trajectory data has its own technical problems; primarily sparsity, non-uniform sampling, usually unlabelled and sporadic sensor data location errors. In this paper we analyse basic track data obtained from consenting subjects using an approach which combines both a direct time series modelling framework, and a probabilistic framework; the latter derived from an acknowledgement that the residual errors in modelling are never Gaussian distributed for real world data due to our uncertainties in data, models and observations. We construct a hybrid dissimilarity representation of the problem to visualise anomalous individuals without needing extreme value statistics (i.e.. behaving differently to the common consensus trajectories) which also reveals preliminary insights into population sub-clusters. In disaster management situations, or pre-disaster planning, this information can help in quickly identifying dangerous behaviour and predicting dangerous behavioural sub-grouping in population movement prompting fast and effective action.

KW - Pattern-of-Life

KW - data visualisation

KW - residual analysis

KW - trajectory analysis

UR - https://ieeexplore.ieee.org/document/8636382/

UR - http://www.scopus.com/inward/record.url?scp=85062834690&partnerID=8YFLogxK

U2 - 10.1109/ICT-DM.2018.8636382

DO - 10.1109/ICT-DM.2018.8636382

M3 - Conference contribution

SN - 978-1-5386-6639-5

SP - 1

EP - 8

BT - 2018 5th International Conference on Information and Communication Technologies for Disaster Management (ICT-DM)

PB - IEEE

ER -

Qarout YK, Lowe D. ‘Surprise’ Detection in Human Pattern-of-Life Behaviour. In 2018 5th International Conference on Information and Communication Technologies for Disaster Management (ICT-DM). IEEE. 2019. p. 1-8. 8636382 https://doi.org/10.1109/ICT-DM.2018.8636382