SlideShare une entreprise Scribd logo
1  sur  17
Télécharger pour lire hors ligne
Didem Gündoğdu
16 September 2016
Emergency Event Detection
Using Mobile Phone Data
Symposium on Big Data and Human
Development, Oxford, September 2016
2010 Post-Election Crisis in Cote d’Ivoire
3
600.000 Displaced people
3.000 Civilian dead
More than
Prob.Stmt.
Research Question
• Where is the anomalous event?
• What time?
• What type of event?
• Social
• Emergency
3 / 15
ConclusionEvaluationMethodologyProb.Def.Background
}Event Detection
( Mobile phone
usage activity )
Prob.Stmt.
How?
• Data -> Mobile Phone Dataset
• Data for Development (D4D) - Ivory Coast (Whole
country)
• Data -> Validation
• United Nations Security Reports and newspapers
• Methodology
• Markov modulated Poisson Process
4 / 15
EvaluationMethodologyProb.Def.Background Conclusion
Prob.Stmt.
Call Detail Records (CDR)
• Collected for billing issues by mobile phone operators
5 / 15
EvaluationMethodologyProb.Def.Background
TimeStamp
Originating
Cell Tower
Terminating
Cell Tower
Number of
VoiceCall
Duration
(sec) Voice
2012-04-28 23:00:00 1236 786 2 96
2012-04-28 23:00:00 1236 804 1 539
2012-04-28 23:00:00 1236 867 3 1778
Conclusion
Prob.Stmt.
• Backward analysis, knowing an anomaly and exploit. [1]
• Aggregated daily anomalies; coarse. [2]
• Track individual change in behaviour; computational cost. [2, 3, 4]
• Supervised learning methods; not adaptable. [5, 6]
6 / 15
EvaluationMethodologyProb.Def.Background
[1] L. Gao, C. Song, Z. Gao, A.-L. Barabási, J. P. Bagrow, and D. Wang. Quantifying information flow during emergencies. Scientific
reports, 4, 2014.
[2] Dobra, N. E. Williams, and N. Eagle. Spatiotemporal detection of unusual human population behavior using mobile phone data. PLoS
ONE, 2015.
[3] L. Akoglu and C. Faloutsos. Event detection in time series of mobile communication graphs. In Army Science Conference, 2010.
[4] V. A. Traag, A. Browet, F. Calabrese, and F. Morlot. Social event detection in massive mobile phone data using probabilistic location
inference. In IEEE Third Int. Conf. on Social Computing, 2011.
[5] M. Faulkner, M. Olson, R. Chandy, J. Krause, K. M. Chandy, and A. Krause. The next big one: Detecting earthquakes and other rare
events from community-based sensors. In 10th International Conference on Information Processing in Sensor Networks (IPSN), 2011.
[6] P. Paraskevopoulos, T. Dinh, Z. Dashdorj, T. Palpanas, and L. Serafini. Identification and characterization of human behavior patterns
from mobile phone data. In International Conference the Analysis of Mobile Phone Datasets (NetMob 2013), Special Session on the Data
for Development (D4D) Challenge, 2013.
Current Works
Conclusion
Prob.Stmt.
Novelty
• Hourly prediction of the anomalous events in spatial data
• Detecting the signature of the event type from the
dissemination velocity and direction
7 / 15
EvaluationMethodologyProb.Def.Background Conclusion
Problem Definition : Spatial Behavioural Understanding from Time Series
8 / 15
EvaluationMethodologyProb.Def.Background
• 970 Antennae
• 7 x 24 time slice
• 7 Weeks
Conclusion
Example: Weekly data from a cell tower
9 / 15
EvaluationMethodologyProb.Def.Background Conclusion
MMPP Model for detecting time varying events
10 / 15
Taken from: Adaptive Event Detection
with Time–Varying Poisson Processes, Ihler et al.
EvaluationMethodologyProb.Def.Background Conclusion
Ground Truth Data from United Nations, News…
Date
Incident
Locations
Subprefec
Subprefec
Name
Antennae
4. Jan.2012 Peite Guiglio 237 Guiglou 521 524
4. Jan.2012
Béoumi near
Bouaké 29 BEOUMI 1119 186
5.Jan.2012 Dobia 150 ISSIA 555 556
6.Jan.2012
Toa Zeo near
Duékoué 165 Duékoué 426 884
11 / 15
EvaluationMethodologyProb.Def.Background Conclusion
Preliminary Results Summary
12 / 15
Baseline MMPP
Emergency
Events
8/19 15/19
Non-Emergency
Events
7/11 8/11
EvaluationMethodologyProb.Def.Background Conclusion
• Gundogdu, D., Incel, O. D., Salah, A. A., & Lepri, B. (2016).
Countrywide arrhythmia: emergency event detection using mobile
phone data. EPJ Data Science, 5(1), 25.
Prob.Stmt.
Lessons Learned
• Understand the data. ( Visualise, have background
information for the analysed period for that country e.g.
there was a civil war in CIV ).
• Data pre-processing is important.
• Missing and/or not reliable periods (e.g. 37 days
western part of CIV very low call volume + 5 days
deleted for keeping weekly periodicity ).
• Evaluating the model: Obtaining ground truth for events in
country scale.
13 / 15
EvaluationMethodologyProb.Def.Background Conclusion
Prob.Stmt.
Future Work
14 / 15
}Event Propagation
• Where is it spreading?
• What type of event?
( Mobility & Activity)
EvaluationMethodologyProb.Def.Background Conclusion
Prob.Stmt.
Conclusions
• Early detection of security incidents can be predicted
through mobile phone data.
• Temporal dissemination of the events can be predicted.
• Governments, international organisations can benefit to
create secure cities for the human well being.
• Another implication can be the verification of
misinformation dissemination in social networks.
15 / 15
EvaluationMethodologyProb.Def.Background Conclusion
–Didem Gündoğdu
gundogdu@fbk.eu
“Thank You.”

Contenu connexe

Similaire à Event detection using mobile phone data

Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metad...
Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metad...Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metad...
Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metad...
Andrey Bogomolov
 
The emergent opportunity of Big Data for Social Good - Nuria Oliver @ PAPIs C...
The emergent opportunity of Big Data for Social Good - Nuria Oliver @ PAPIs C...The emergent opportunity of Big Data for Social Good - Nuria Oliver @ PAPIs C...
The emergent opportunity of Big Data for Social Good - Nuria Oliver @ PAPIs C...
PAPIs.io
 
Cloud and Crowd research at ITC
Cloud and Crowd research at ITCCloud and Crowd research at ITC
Cloud and Crowd research at ITC
Frank Ostermann
 

Similaire à Event detection using mobile phone data (20)

Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metad...
Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metad...Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metad...
Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metad...
 
Mobile Data Analytics
Mobile Data AnalyticsMobile Data Analytics
Mobile Data Analytics
 
Big Data & Analytics for Government - Case Studies
Big Data & Analytics for Government - Case StudiesBig Data & Analytics for Government - Case Studies
Big Data & Analytics for Government - Case Studies
 
Presentation Template
Presentation TemplatePresentation Template
Presentation Template
 
Mlhil ljr.web.285
Mlhil ljr.web.285Mlhil ljr.web.285
Mlhil ljr.web.285
 
The emergent opportunity of Big Data for Social Good - Nuria Oliver @ PAPIs C...
The emergent opportunity of Big Data for Social Good - Nuria Oliver @ PAPIs C...The emergent opportunity of Big Data for Social Good - Nuria Oliver @ PAPIs C...
The emergent opportunity of Big Data for Social Good - Nuria Oliver @ PAPIs C...
 
International Visual Methods Conference Sept OU 2011
International Visual Methods Conference Sept OU 2011International Visual Methods Conference Sept OU 2011
International Visual Methods Conference Sept OU 2011
 
ICCM 2013 Panel 1: What's so Big about Big Data?
ICCM 2013 Panel 1: What's so Big about Big Data?ICCM 2013 Panel 1: What's so Big about Big Data?
ICCM 2013 Panel 1: What's so Big about Big Data?
 
Participatory Data Gathering for Public Sector Reuse: Lessons Learned from T...
Participatory Data Gathering for Public Sector Reuse:  Lessons Learned from T...Participatory Data Gathering for Public Sector Reuse:  Lessons Learned from T...
Participatory Data Gathering for Public Sector Reuse: Lessons Learned from T...
 
DPS-Tartu-ResearchAgenda2023-huberflores.pdf
DPS-Tartu-ResearchAgenda2023-huberflores.pdfDPS-Tartu-ResearchAgenda2023-huberflores.pdf
DPS-Tartu-ResearchAgenda2023-huberflores.pdf
 
Role of Data Accessibility During Pandemic
Role of Data Accessibility During PandemicRole of Data Accessibility During Pandemic
Role of Data Accessibility During Pandemic
 
Big data for development
Big data for development Big data for development
Big data for development
 
Opportunities in technology and connected health for population science
Opportunities in technology and connected health for population science Opportunities in technology and connected health for population science
Opportunities in technology and connected health for population science
 
Setting up of a Kenya National Domestic Violence Call Center (K-NDVCC): Proto...
Setting up of a Kenya National Domestic Violence Call Center (K-NDVCC): Proto...Setting up of a Kenya National Domestic Violence Call Center (K-NDVCC): Proto...
Setting up of a Kenya National Domestic Violence Call Center (K-NDVCC): Proto...
 
IRJET- Awareness and Knowledge about Android Smartphones Security among Ghana...
IRJET- Awareness and Knowledge about Android Smartphones Security among Ghana...IRJET- Awareness and Knowledge about Android Smartphones Security among Ghana...
IRJET- Awareness and Knowledge about Android Smartphones Security among Ghana...
 
Introduction to privacy feedback research @ DesRes2016
Introduction to privacy feedback research @ DesRes2016Introduction to privacy feedback research @ DesRes2016
Introduction to privacy feedback research @ DesRes2016
 
Science, evidence and data in government presentation
Science, evidence and data in government presentationScience, evidence and data in government presentation
Science, evidence and data in government presentation
 
Cloud and Crowd research at ITC
Cloud and Crowd research at ITCCloud and Crowd research at ITC
Cloud and Crowd research at ITC
 
Csec 650 individual assignment i
Csec 650 individual assignment iCsec 650 individual assignment i
Csec 650 individual assignment i
 
An Exposition Of The Nature Of Volunteered Geographical Information And Its S...
An Exposition Of The Nature Of Volunteered Geographical Information And Its S...An Exposition Of The Nature Of Volunteered Geographical Information And Its S...
An Exposition Of The Nature Of Volunteered Geographical Information And Its S...
 

Dernier

EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptxEIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
Earley Information Science
 

Dernier (20)

Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
 
08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men
 
Understanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdfUnderstanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdf
 
How to convert PDF to text with Nanonets
How to convert PDF to text with NanonetsHow to convert PDF to text with Nanonets
How to convert PDF to text with Nanonets
 
How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerHow to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected Worker
 
Boost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityBoost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivity
 
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
 
presentation ICT roal in 21st century education
presentation ICT roal in 21st century educationpresentation ICT roal in 21st century education
presentation ICT roal in 21st century education
 
Presentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreterPresentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreter
 
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationFrom Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
 
Strategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a FresherStrategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a Fresher
 
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf
 
A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)
 
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptxEIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
 
Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024
 
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
 
Driving Behavioral Change for Information Management through Data-Driven Gree...
Driving Behavioral Change for Information Management through Data-Driven Gree...Driving Behavioral Change for Information Management through Data-Driven Gree...
Driving Behavioral Change for Information Management through Data-Driven Gree...
 
Exploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone ProcessorsExploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone Processors
 
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdfThe Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
 
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemkeProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
 

Event detection using mobile phone data

  • 1. Didem Gündoğdu 16 September 2016 Emergency Event Detection Using Mobile Phone Data Symposium on Big Data and Human Development, Oxford, September 2016
  • 2. 2010 Post-Election Crisis in Cote d’Ivoire
  • 3. 3 600.000 Displaced people 3.000 Civilian dead More than
  • 4. Prob.Stmt. Research Question • Where is the anomalous event? • What time? • What type of event? • Social • Emergency 3 / 15 ConclusionEvaluationMethodologyProb.Def.Background }Event Detection ( Mobile phone usage activity )
  • 5. Prob.Stmt. How? • Data -> Mobile Phone Dataset • Data for Development (D4D) - Ivory Coast (Whole country) • Data -> Validation • United Nations Security Reports and newspapers • Methodology • Markov modulated Poisson Process 4 / 15 EvaluationMethodologyProb.Def.Background Conclusion
  • 6. Prob.Stmt. Call Detail Records (CDR) • Collected for billing issues by mobile phone operators 5 / 15 EvaluationMethodologyProb.Def.Background TimeStamp Originating Cell Tower Terminating Cell Tower Number of VoiceCall Duration (sec) Voice 2012-04-28 23:00:00 1236 786 2 96 2012-04-28 23:00:00 1236 804 1 539 2012-04-28 23:00:00 1236 867 3 1778 Conclusion
  • 7. Prob.Stmt. • Backward analysis, knowing an anomaly and exploit. [1] • Aggregated daily anomalies; coarse. [2] • Track individual change in behaviour; computational cost. [2, 3, 4] • Supervised learning methods; not adaptable. [5, 6] 6 / 15 EvaluationMethodologyProb.Def.Background [1] L. Gao, C. Song, Z. Gao, A.-L. Barabási, J. P. Bagrow, and D. Wang. Quantifying information flow during emergencies. Scientific reports, 4, 2014. [2] Dobra, N. E. Williams, and N. Eagle. Spatiotemporal detection of unusual human population behavior using mobile phone data. PLoS ONE, 2015. [3] L. Akoglu and C. Faloutsos. Event detection in time series of mobile communication graphs. In Army Science Conference, 2010. [4] V. A. Traag, A. Browet, F. Calabrese, and F. Morlot. Social event detection in massive mobile phone data using probabilistic location inference. In IEEE Third Int. Conf. on Social Computing, 2011. [5] M. Faulkner, M. Olson, R. Chandy, J. Krause, K. M. Chandy, and A. Krause. The next big one: Detecting earthquakes and other rare events from community-based sensors. In 10th International Conference on Information Processing in Sensor Networks (IPSN), 2011. [6] P. Paraskevopoulos, T. Dinh, Z. Dashdorj, T. Palpanas, and L. Serafini. Identification and characterization of human behavior patterns from mobile phone data. In International Conference the Analysis of Mobile Phone Datasets (NetMob 2013), Special Session on the Data for Development (D4D) Challenge, 2013. Current Works Conclusion
  • 8. Prob.Stmt. Novelty • Hourly prediction of the anomalous events in spatial data • Detecting the signature of the event type from the dissemination velocity and direction 7 / 15 EvaluationMethodologyProb.Def.Background Conclusion
  • 9. Problem Definition : Spatial Behavioural Understanding from Time Series 8 / 15 EvaluationMethodologyProb.Def.Background • 970 Antennae • 7 x 24 time slice • 7 Weeks Conclusion
  • 10. Example: Weekly data from a cell tower 9 / 15 EvaluationMethodologyProb.Def.Background Conclusion
  • 11. MMPP Model for detecting time varying events 10 / 15 Taken from: Adaptive Event Detection with Time–Varying Poisson Processes, Ihler et al. EvaluationMethodologyProb.Def.Background Conclusion
  • 12. Ground Truth Data from United Nations, News… Date Incident Locations Subprefec Subprefec Name Antennae 4. Jan.2012 Peite Guiglio 237 Guiglou 521 524 4. Jan.2012 Béoumi near Bouaké 29 BEOUMI 1119 186 5.Jan.2012 Dobia 150 ISSIA 555 556 6.Jan.2012 Toa Zeo near Duékoué 165 Duékoué 426 884 11 / 15 EvaluationMethodologyProb.Def.Background Conclusion
  • 13. Preliminary Results Summary 12 / 15 Baseline MMPP Emergency Events 8/19 15/19 Non-Emergency Events 7/11 8/11 EvaluationMethodologyProb.Def.Background Conclusion • Gundogdu, D., Incel, O. D., Salah, A. A., & Lepri, B. (2016). Countrywide arrhythmia: emergency event detection using mobile phone data. EPJ Data Science, 5(1), 25.
  • 14. Prob.Stmt. Lessons Learned • Understand the data. ( Visualise, have background information for the analysed period for that country e.g. there was a civil war in CIV ). • Data pre-processing is important. • Missing and/or not reliable periods (e.g. 37 days western part of CIV very low call volume + 5 days deleted for keeping weekly periodicity ). • Evaluating the model: Obtaining ground truth for events in country scale. 13 / 15 EvaluationMethodologyProb.Def.Background Conclusion
  • 15. Prob.Stmt. Future Work 14 / 15 }Event Propagation • Where is it spreading? • What type of event? ( Mobility & Activity) EvaluationMethodologyProb.Def.Background Conclusion
  • 16. Prob.Stmt. Conclusions • Early detection of security incidents can be predicted through mobile phone data. • Temporal dissemination of the events can be predicted. • Governments, international organisations can benefit to create secure cities for the human well being. • Another implication can be the verification of misinformation dissemination in social networks. 15 / 15 EvaluationMethodologyProb.Def.Background Conclusion