predictive modeling of human behavior: supervised learning from telecom metadata [doctoral thesis...
TRANSCRIPT
![Page 1: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/1.jpg)
April 12, 2017, Trento, Italy.
SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY IN COMPUTER SCIENCE
Predictive Modelingof Human Behavior:Supervised Learning from Telecom Metadata
by Andrey Bogomolov 1
Advisor: Prof. Fabio Pianesi, Fondazione Bruno Kessler, EIT Digital
Industrial Co-Advisor: Michele Caraviello, Telecom Italia SpA
1University of Trento
Via Sommarive, 9, 38123, Trento, Italy
1 / 56
![Page 2: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/2.jpg)
Outline I
1 Introduction
– Motivation - The Problem of Human Behavior Computation
– Research Goals
– Contribution and Structure
– Main Publications List
2 Supervised Learning of Individual Characteristics: Affective States
– Daily Stress Recognition from Mobile Phone Data
– Happiness Recognition from Telecomunications Metadata
3 Predictive Modeling of Large Scale Group Behavior
– Predicting Crime Hotspots Based on Aggregated Anonymized People
Dynamics
– Predicting Electric Energy Consumption Using Aggregated Telecom
Data2 / 56
![Page 3: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/3.jpg)
Outline II
4 Summary and Future Work
– Computing Human Behavior: The New Kind of Science
– Supervised Learning of Individual Characteristics
– Supervised Learning of Large Scale Group Behavior
– Challenges
3 / 56
![Page 4: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/4.jpg)
Introduction
![Page 5: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/5.jpg)
Motivation - The Problem of Human Behavior Computation
Advancement of artificial intelligence and machine learning for
social good
• Computational Social Science
• Social physics
• Interdisciplinary Research
New Kind of Science
• Data Driven
• Ubiquitous Big Data
• High Performance Computing
• Complex Structural and Causal Relationships
4 / 56
![Page 6: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/6.jpg)
Research Goals
Tasks
• Human behavior prediction vs recognition
• From individual behavior to large scale group behavior
Use cases
• Affective computing – daily happiness and stress recognition
• Crime hotspots prediction
• Energy consumption prediction
Scope
• Limited to Telecom Metadata and relevant multimodal data
• Data Driven
• Interpretable by human (=no deep learning)
5 / 56
![Page 7: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/7.jpg)
Contribution and Structure
Contribution
• Algorithmic recognition of individuals’ affective states based on the
data collected on their smart phones.
• Formulated and algorithmically solved the problem of predicting
crime hot spots from telecom metadata in a city.
• Proposed new methods for learning the intrinsic relationships
between the characteristics given observations (logs) framed as
machine learning problems with reduced and highly compactified
feature space for energy consumption prediction problem.
• Provided insight into developing novel signal processing and machine
learning techniques for human behavior understanding.
6 / 56
![Page 8: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/8.jpg)
Main Publications List I
1. Andrey Bogomolov, Bruno Lepri, and Fabio Pianesi.
Happiness recognition from mobile phone data.
In Social Computing (SocialCom), 2013 International Conference on,
pages 790–795. IEEE, 2013.
2. Andrey Bogomolov, Bruno Lepri, Michela Ferron, Fabio Pianesi, and
Alex Sandy Pentland.
Daily stress recognition from mobile phone data, weather
conditions and individual traits.
In Proceedings of the 22nd ACM international conference on
Multimedia, pages 477–486. ACM, 2014.
7 / 56
![Page 9: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/9.jpg)
Main Publications List II
3. Andrey Bogomolov, Bruno Lepri, Michela Ferron, Fabio Pianesi, and
Alex Sandy Pentland.
Pervasive stress recognition for sustainable living.
In Pervasive Computing and Communications (PERCOM), 2014
IEEE International Conference on, pages 345–350. IEEE, 2014.
4. Andrey Bogomolov, Bruno Lepri, Jacopo Staiano, Nuria Oliver,
Fabio Pianesi, and Alex Pentland.
Once upon a crime: towards crime prediction from
demographics and mobile data.
In Proceedings of the 16th international conference on multimodal
interaction, pages 427–434. ACM, 2014.
8 / 56
![Page 10: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/10.jpg)
Main Publications List III
5. Andrey Bogomolov, Bruno Lepri, and Fabio Pianesi.
Generalized compression dictionary distance as universal
similarity measure.
In Big Data from Space (BiDS’14), number doi: 10.2788/1823,
pages 63–66. Publications Office of the European Union, 2014.
6. Andrey Bogomolov, Bruno Lepri, Jacopo Staiano, Emmanuel
Letouze, Nuria Oliver, Fabio Pianesi, and Alex Pentland.
Moves on the street: Classifying crime hotspots using
aggregated anonymized data on people dynamics.
Big Data, 3(3):148–158, 2015.
7. Andrey Bogomolov, Bruno Lepri, Roberto Larcher, Fabrizio
Antonelli, Fabio Pianesi, and Alex Pentland.
Energy consumption prediction using people dynamics derived
from cellular network data.
EPJ Data Science, 5(1):13, 2016.
9 / 56
![Page 11: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/11.jpg)
Supervised Learning of Individual
Characteristics: Affective States
![Page 12: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/12.jpg)
Daily Stress Recognition from Mobile Phone Data:
Our Multi-factorial Problem Statement
Inputs
• Smartphone call log
• Smartphone sms log
• Smartphone Bluetooth proximity hits
• Weather conditions
• Personality traits
Outputs
• Daily stress recognition
• 2-classes: {0 = “not stressed”, 1 = “stressed”}.
10 / 56
![Page 13: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/13.jpg)
Stress: Data - the Living Laboratory
• 111 subjects
• dates: 12 November, 2010 – 21 May, 2011
• daily surveys on subjective stress
• Big Five personality traits
• weather data
Mobile Phone Data
phone calls 33497
sms 22587
Bluetooth hits 1460939
11 / 56
![Page 14: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/14.jpg)
Stress: Data
Descriptive StatisticsHistogram of r$stress
Stress Value
Count
0 1 2 3 4 5 6 7
0500
1000
1500
2000
2500
12 / 56
![Page 15: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/15.jpg)
Stress: Data
Within-person and between-subject variance
0 1 2 3 4 5 6
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
density.default(x = r1[, 1])
Variance
Den
sity
Within−Person VarianceBetween−Person Variance
13 / 56
![Page 16: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/16.jpg)
Stress: Feature Space Assumptions
Basic Mobile Phone Features
I General Phone Usage
I Diversity
I Active Behaviors
I Regularity
14 / 56
![Page 17: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/17.jpg)
Stress: Feature Space Assumptions
Basic Mobile Phone Features: General Phone Usage
1. Total Number of Calls (Outgoing+Incoming)
2. Total Number of Incoming Calls
3. Total Number of Outgoing Calls
4. Total Number of Missed Calls
5. Number of SMS received
6. Number of SMS sent
15 / 56
![Page 18: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/18.jpg)
Stress: Feature Space Assumptions
Basic Mobile Phone Features: Diversity
7. Number of Unique Contacts Called
8. Number of Unique Contacts who Called
9. Number of Unique Contacts Communicated with (Incoming+Outgoing)
10. Number of Unique Contacts Associated with Missed Calls
11. Entropy of Call Contacts
12. Call Contacts to Interactions Ratio
13. Number of Unique Contacts SMS received from
14. Number of Unique Contacts SMS sent to
15. Entropy of SMS Contacts
16. Sms Contacts to Interactions Ratio
16 / 56
![Page 19: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/19.jpg)
Stress: Feature Space Assumptions
Basic Mobile Phone Features: Active Behaviors
17. Percent Call During the Night
18. Percent Call Initiated
19. Sms response rate
20. Sms response latency
21. Percent SMS Initiated
17 / 56
![Page 20: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/20.jpg)
Stress: Feature Space Assumptions
Basic Mobile Phone Features: Regularity
22. Average Inter-event Time for Calls (time elapsed between two events)
23. Average Inter-event Time for SMS (time elapsed between two events)
24. Variance Inter-event Time for Calls (time elapsed between two events)
25. Variance Inter-event Time for SMS (time elapsed between two events)
18 / 56
![Page 21: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/21.jpg)
Stress: Feature Space Assumptions
Basic Proximity Features
General Bluetooth Proximity
1. Number of Bluetooth IDs
2. Times most common Bluetooth ID is seen
3. Bluetooth IDs accounting for n% of IDs seen
4. Bluetooth IDs seen for more than k time slots
5. Time interval for which a Bluetooth ID is seen
6. Entropy of Bluetooth contacts
Diversity
7. Contacts to interactions ratio
Regularity
8. Average Bluetooth interactions inter-event time
(time elapsed between two events)
9. Variance of the Bluetooth interactions inter-event time
(time elapsed between two events)
19 / 56
![Page 22: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/22.jpg)
Stress: Feature Space Innovation
“Background Noise” Features
I a) peoples activity, as detected through their smartphones
I b) the weather conditions {humidity, wind speed, pressure, total
precipitation and visibility}I c) personality traits {“Big Five”}
Functional Innovation
I a) time domain – sliding window functions
I b) Miller-Madow correction for entropy calculation
HMM(θ) ≡ −p∑
i=1
θML,i log θML,i +m − 1
2N
20 / 56
![Page 23: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/23.jpg)
Stress: Feature Space
Rank Feature 0 1 Mean Decrease in Accuracy Mean Decrease in Gini Index
1 personality.Conscientiousness 13.65 18.04 23.35 159.96
2 personality.Agreeableness 14.22 19.73 22.92 167.30
3 personality.Neuroticism 15.96 21.04 22.56 183.87
4 personality.Openness 14.20 14.18 21.38 139.23
5 personality.Extraversion 15.75 15.02 21.07 158.51
6 weather.MeanTemperature 14.50 6.34 17.44 322.27
7 sms.RepliedEvents.Latency.Median 8.83 13.85 15.63 48.74
8 weather.Humidity 15.33 2.10 15.45 298.13
9 sms.AllEventsPerDay 8.61 0.56 10.50 42.91
10 bluetooth.Q95TimeForWhichIdSeen 4.99 6.05 9.94 32.47
11 bluetooth.MaxTimeForWhichIdSeen 6.24 7.23 9.47 32.12
12 sms.IncomingAndOutgoingPerDay 7.45 1.26 9.38 41.59
13 weather.Visibility 9.94 1.26 9.22 251.27
14 weather.WindSpeed 8.77 1.30 8.67 282.10
15 bluetooth.Q90TimeForWhichIdSeen 4.24 6.75 8.64 28.41
16 bluetooth.TotalEntropyShannon 5.04 3.51 8.56 31.37
17 call.EntropyMillerMadowOutgoingTotal 4.25 4.10 8.54 27.49
18 call.EntropyShannonOutgoingAndIncomingTotal 4.23 4.86 8.53 26.28
19 bluetooth.TotalEntropyMillerMadow 5.06 4.22 8.50 32.09
20 bluetooth.IdsMoreThan09TimeSlotsSeen 6.11 5.85 8.43 27.88
21 bluetooth.IdsMoreThan04TimeSlotsSeen 6.34 4.59 8.04 24.64
22 call.EntropyShannonMissedOutgoingTotal 3.13 4.92 7.85 24.34
23 bluetooth.IdsMoreThan19TimeSlotsSeen 2.97 5.16 7.78 20.87
24 call.EntropyShannonOutgoingTotal 3.10 6.45 7.78 24.79
25 bluetooth.Q75TimeForWhichIdSeen 5.16 4.70 7.76 22.07
26 call.EntropyMillerMadowMissedOutgoingTotal 4.09 5.45 7.55 24.64
27 call.EntropyMillerMadowOutgoingAndIncomingTotal 3.87 6.29 7.51 28.63
28 sms.OutgoingAndIncomingTotalEntropyMillerMadow 4.68 3.84 7.19 17.63
29 sms.OutgoingTotalEntropyMillerMadow 5.22 1.49 7.19 18.88
30 bluetooth.Q50TimeForWhichIdSeen 1.53 7.29 7.08 18.91
31 bluetooth.Q68TimeForWhichIdSeen 2.36 5.96 6.68 19.05
32 sms.OutgoingTotalEntropyShannon 2.53 2.77 5.13 17.59
21 / 56
![Page 24: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/24.jpg)
Stress: Classification Approach
• binary classifier task: stressed vs not stressed
• random split into training (80%) and testing (20%)
• classifier: ensemble of tree classifiers based on RF
• RF overcomes Generalized Boosted Model, SVMs with linear and
RBF kernels, and Neural Networks
22 / 56
![Page 25: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/25.jpg)
Stress: Performance Metrics
Final Classifier Performance Metrics Comparison
Metric Value
Accuracy 0.7228
95% CI (0.7051, 0.7399)
No Information Rate 0.6384
P-Value [Acc > NIR] < 2.2e-16
Kappa 0.3752
Sensitivity 0.5272
Specificity 0.8335
’Positive’ Class “stressed”
23 / 56
![Page 26: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/26.jpg)
Stress: Results
Our Approach Feature Subsets Comparison
Model Accuracy Kappa Sensitivity Specificity F1
Our Multifactorial Model 72.28 37.52 52.72 83.35 57.89
Baseline Majority Classifier 63.84 0.00 100.00 0.00 0.00
Weather Only 36.16 0.00 100.00 0.00 0.00
Personality Only 36.16 0.00 100.00 0.00 0.00
Bluetooth+Call+Sms 48.59 6.80 73.80 34.32 50.94
Personality+Weather 43.55 2.96 81.90 21.83 51.20
Personality+Bluetooth+Call+Sms 46.40 7.01 83.17 25.57 52.88
Weather+Bluetooth+Call+Sms 49.60 -5.45 38.45 55.91 35.55
24 / 56
![Page 27: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/27.jpg)
Stress: Relevant Predictors
Personality Traits
• all personality traits contributes significantly in predicting daily stress
• previous works focused only on Neuroticism, Extraversion and
Conscientiousness
Weather Conditions
• confirmed association between temperature and stress
• significant effects of humidity, visibility and wind speed
Mobile Phone Data
• relevant contribution of proximity features
• relevant contribution of entropy based proximity, call and sms
features
25 / 56
![Page 28: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/28.jpg)
Stress: Implications
• cost-effective, unobtrusive, widely available and reliable tool for
stress recognition
• real life multimodal data
• input for the assessment and treatment of psychological stress
systems
26 / 56
![Page 29: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/29.jpg)
Stress: Limitations
Data
• Data loss not registered as NA’s
• Battery
• Temporal resolution
Model
• Requires personality data
• Requires 1 week data collection period
• Not tested on diverse groups
27 / 56
![Page 30: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/30.jpg)
Happiness Recognition from Telecom Metadata
Similar Approach
• Data collected on Mobile Phone (Android)
• Feature Space Assumption
• Fighting noisy data – Ensemble Models
Expected Results?
• Similar but different predictive variables
• Better Metrics if more data?
28 / 56
![Page 31: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/31.jpg)
Happiness: Problem Statement
Inputs
• Smartphone call log.
• Smartphone sms log.
• Smartphone Bluetooth proximity hits.
Outputs
• Daily happiness recognition.
• 3-classes: {not happy, neutral, happy}.
29 / 56
![Page 32: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/32.jpg)
Happiness Data
Descriptive Statistics
mean 4.84
standard deviation 1.26
median 5.00
mean average deviation 1.48
min 1.00
max 7.00
range 6.00
skew -0.39
kurtosis -0.07
30 / 56
![Page 33: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/33.jpg)
Happiness: Predictive Variables
-1 0 1 MeanDecreaseAccuracy MeanDecreaseGini
meanTemperature 0.0099 0.0033 0.0094 0.0082 154.8379
humidity 0.0047 -0.0033 0.0115 0.0074 149.7668
pressure -0.0002 0.0008 0.0015 0.0011 149.0302
windSpeed 0.0028 0.0017 0.0051 0.0040 142.7727
visibility 0.0024 -0.0009 0.0051 0.0034 120.8683
neuroticism 0.0690 0.0288 0.0399 0.0419 90.5721
conscientiousness 0.0659 0.0480 0.0668 0.0627 90.2708
extraversion 0.0511 0.0357 0.0472 0.0454 76.9467
openness 0.0656 0.0340 0.0406 0.0429 73.9181
totalPrecipitation 0.0007 -0.0000 0.0012 0.0009 73.5273
agreeableness 0.0536 0.0282 0.0235 0.0289 70.7261
bluetoothQ95TimeForWhichIdSeen 0.0233 0.0120 0.0149 0.0155 22.4018
bluetoothQ90TimeForWhichIdSeen 0.0161 0.0082 0.0141 0.0131 19.8364
smsRepliedEventsLatencyMedian 0.0093 0.0085 0.0096 0.0093 18.7719
bluetoothIdsMoreThan04TimeSlotsSeen 0.0114 0.0066 0.0124 0.0110 15.6738
bluetoothMaxTimeForWhichIdSeen 0.0141 0.0072 0.0095 0.0097 15.2546
bluetoothTotalEntropyMillerMadow 0.0058 0.0017 0.0035 0.0035 13.9000
bluetoothTotalEntropyShannon 0.0065 0.0009 0.0060 0.0050 13.1826
callMeanInterEventTimePerDay -0.0003 -0.0000 0.0014 0.0009 13.1180
incomingAndOutgoingCallsPerDay -0.0000 -0.0002 0.0024 0.0015 12.3962
bluetoothQ50TimeForWhichIdSeen 0.0104 0.0104 0.0091 0.0095 12.2270
callStandardDeviationInterEventTimePerDay 0.0004 -0.0005 0.0007 0.0004 10.1723
bluetoothIdsMoreThan19TimeSlotsSeen 0.0087 0.0033 0.0089 0.0077 9.7388
incomingCallsPerDay 0.0003 -0.0005 0.0009 0.0005 9.6572
outgoingContactsToInteractionsRatioPerDay 0.0005 -0.0004 0.0013 0.0009 9.2016
callsInitiatedRatioPerDay -0.0001 -0.0001 0.0014 0.0008 9.0245
entropyMillerMadowCallsOutgoingWindow3Days 0.0001 -0.0008 0.0014 0.0008 8.7199
bluetoothIdsMoreThan09TimeSlotsSeen 0.0074 0.0046 0.0040 0.0046 8.6006
bluetoothQ75TimeForWhichIdSeen 0.0027 0.0018 0.0032 0.0028 8.4454
outgoingCallsPerDay -0.0001 -0.0001 0.0012 0.0007 8.336831 / 56
![Page 34: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/34.jpg)
Happiness: Ensemble of Decision Trees Classifier
The decision boundary => margin function:
mg(X,Y ) = avgk I (hk(X) = Y )−maxj!=Y avgk I (hk(X) = j),
(1)
Generalization error function is:
PE∗ = PX,Y (mg(X,Y ) < 0), (2)
where PX,Y is the probability over 〈X,Y 〉 space. Characteristic
function I (·) of A is:
IA(x) =
{1 ⇐⇒ (x ⊂ A)
0 otherwise
}{1 ⇐⇒ ∃x0 otherwise
}(3)
32 / 56
![Page 35: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/35.jpg)
Happiness: Results
Final Classifier Performance Metrics Comparison
Training set Test set
Accuracy 0.8081 0.8036
Kappa 0.5879 0.5743
AccuracyLower 0.8004 0.7878
AccuracyUpper 0.8156 0.8187
AccuracyNull 0.6415 0.6419
AccuracyPValue 2.139e-303 8.826e-73
McnemarPValue 5.647e-208 1.738e-57
33 / 56
![Page 36: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/36.jpg)
Happiness: Results
What We Learnt: Final Model ROC curve
Specificity
Sen
sitiv
ity
0.0
0.2
0.4
0.6
0.8
1.0
1.0 0.8 0.6 0.4 0.2 0.0
AUC: 0.844
34 / 56
![Page 37: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/37.jpg)
Predictive Modeling of Large
Scale Group Behavior
![Page 38: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/38.jpg)
Predicting Crime Hotspots Based on Aggregated Anonymized
People Dynamics
SmartSteps Data Description
Type Data
Origin-based total # people
# residents
# workers
# visitors
Gender-based # males
# females
Age-based # people aged up to 20
# people aged 21 to 30
# people aged 31 to 40
# people aged 41 to 50
# people aged 51 to 60
# people aged over 60
Criminal Cases Dataset
London Borough Profiles Dataset
35 / 56
![Page 39: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/39.jpg)
Predicting Crime Hotspots: Methodology
Problem statement
• Binary classification task For each Smartsteps cell, we predict
whether that particular cell will be a crime hotspot or not in the
next month.
Model Building
• Data Preprocessing
• Referencing all geo-tagged data to the Smartsteps cells
• Feature Extraction
• Tree-based Ensemble Model
36 / 56
![Page 40: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/40.jpg)
Predicting Crime Hotspots: Visualization
Ground Truth vs Predicted by the Algorithm
51.3
51.4
51.5
51.6
51.7
−0.50 −0.25 0.00 0.25lon
lat
Crime Level01
Radius0.51.01.5
51.3
51.4
51.5
51.6
51.7
−0.50 −0.25 0.00 0.25lon
lat
Crime Level01
Radius0.51.01.5
37 / 56
![Page 41: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/41.jpg)
Predicting Crime Hotspots: Top-20 Features Ranked by Mean
Decrease in Accuracy
Rank Features 0 1 MeanDecreaseAccuracy MeanDecreaseGini
1 smartSteps.daily.ageover60.entropy.empirical.entropy.empirical 4.48 5.43 9.02 18.75
2 smartSteps.daily.athome.mean.sd 3.20 7.60 8.91 27.13
3 smartSteps.daily.age020.sd.entropy.empirical 5.69 3.97 8.85 16.88
4 smartSteps.daily.age020.mean.entropy.empirical 3.09 5.88 8.85 17.26
5 smartSteps.daily.age020.mean.sd 4.50 5.27 8.65 16.03
6 smartSteps.daily.athome.min.entropy.empirical 6.39 2.32 8.61 15.99
7 smartSteps.daily.athome.sd.sd 3.22 8.58 8.60 45.82
8 smartSteps.daily.athome.sd.mean 3.35 5.83 8.57 24.93
9 smartSteps.daily.ageover60.entropy.empirical.sd 4.62 4.95 8.56 20.45
10 smartSteps.daily.athome.sd.median 5.41 5.04 8.50 26.48
11 smartSteps.daily.age3140.entropy.empirical.max 2.33 5.79 8.44 16.24
12 smartSteps.daily.age3140.min.sd 6.81 4.06 8.31 36.52
13 smartSteps.daily.athome.min.sd 4.36 6.85 8.29 34.26
14 smartSteps.daily.athome.sd.max 4.13 6.87 8.27 34.89
15 smartSteps.monthly.athome.max 3.92 5.42 8.26 29.86
16 smartSteps.monthly.athome.sd 4.43 4.17 8.21 39.70
17 smartSteps.daily.age5160.entropy.empirical.entropy.empirical 4.74 4.11 8.13 16.64
18 smartSteps.daily.age020.sd.sd 3.67 5.88 8.12 16.86
19 smartSteps.daily.athome.entropy.empirical.entropy.empirical 5.13 4.82 8.08 18.55
20 smartSteps.daily.athome.max.sd 2.83 6.29 8.07 26.85
38 / 56
![Page 42: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/42.jpg)
Predicting Crime Hotspots: Results
Crime Hotspots Prediction Problem Metrics Comparison
Model Acc.,% Acc. CI, 95% F1,% AUC
Baseline Majority Classifier 53.15 (0.53, 0.53) 0 0.50
Borough Profiles Model (BPM) 62.18 (0.61, 0.64) 57.52 0.58
Smartsteps 68.37 (0.67, 0.70) 65.43 0.63
Smartsteps + BPM 69.54 (0.68, 0.71) 67.23 0.64
39 / 56
![Page 43: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/43.jpg)
Predicting Electric Energy Consumption Using Aggregated Tele-
com Data:
Energy Consumption Dataset
2 months and a territory of 6000 square kilometers in the Northern
Italy. The telecommunication and the energy consumption datasets
have the same spatio-temporal aggregation. The temporal aggregation
is ten minute intervals while the spatial one is obtained by partitioning
the territory using a regular square grid 1 square kilometer.
Telecommunication dataset
Structure of the electrical grid
40 / 56
![Page 44: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/44.jpg)
Electric Energy Consumption: Data
Energy Consumption Time Series Decomposition for a Residential
Area
41 / 56
![Page 45: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/45.jpg)
Electric Energy Consumption: Data
Energy Consumption Time Series Decomposition for a Touristic
Area
42 / 56
![Page 46: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/46.jpg)
Electric Energy Consumption: Data
Energy Consumption Time Series Decomposition for a City Center
43 / 56
![Page 47: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/47.jpg)
Electric Energy Consumption: Methodology
Problem statement
• Regression task (1) – daily average energy consumption for the next
7 days
• Regression task (2) – daily peak energy consumption for the next 7
days
Model Building
• Spatial Mapping
• Data Aggregation
• Feature Extraction: spectral domain
• Ensemble Models
44 / 56
![Page 48: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/48.jpg)
Electric Energy Consumption: Spectral
Spectral Characteristics of Typical Energy Consumption Response
Variables
45 / 56
![Page 49: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/49.jpg)
Mean Daily Energy Consumption Predictor Variables – Top 24
Rank Feature Decrease in Decrease in
MSE Node Impurity
1 Number of consumers per grid powerline 19.64 69109.19
2 Variance of real numbers part of Fourier transform of area codes 4.43 6534.18
3 Variance of real numbers part of Fourier transform of outgoing sms activity 3.89 6811.37
4 Variance of calling direction area codes 3.56 7964.44
5 Variance of entropy of outgoing call activity in time domain 3.47 11568.62
6 Entropy of first harmonic of outgoing calls 3.47 3671.82
7 Entropy of internet activity summed in time domain 3.21 2750.14
8 Kurtosis of entropy distribution of outgoing calls 3.14 1472.87
9 Variance of skewness of temporal distribution of outgoing calls 3.10 2111.95
10 Entropy of fundamental frequency (first harmonic) of internet activity 3.03 1668.04
11 Standard deviation of frequencies distribution skewness of incoming sms 3.03 2848.24
12 Entrophy of sum in time domain of outgoing calls 2.98 3651.77
13 Variance of the kurtosis of outgpoung calls in time domain 2.98 2136.39
14 Kurtosis of time entropy of mobile internet activity 2.96 4409.30
15 Median of outgoing calls variance in frequiency domain 2.92 1317.35
16 Sum of 4 harmonic of incoming calls 2.88 5035.26
17 Sum of outgoing calls skewness in frequency domain 2.88 825.13
18 Median of outgoing sms temporal distribution kurtosis 2.88 1260.23
19 Kurtosis of outgoing calls 32 harmonic 2.81 947.15
20 Median of 11 harmonic of internet activity 2.72 185.14
21 Variance of 29 harmonic of outgoing sms 2.71 1243.93
22 Variance of outgoing sms fundamental frequnecy 2.69 5418.83
23 Sum of intertemporal mean of outgoing sms 2.67 182.93
24 Sum of calling direction area codes 2.63 4075.28
46 / 56
![Page 50: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/50.jpg)
Peak Daily Energy Consumption Predictor Variables – Top 24
Rank Feature Decrease in Decrease in
MSE Node Impurity
1 Number of consumers per grid powerline 21.22 132035.53
2 Entropy of temporal sum of outgoung calls 5.86 24013.99
3 Kurtosis of temporal entropy of internet activity 4.33 15479.71
4 Entropy outgoing calls fundamental frequnecy 4.10 14650.13
5 Sum of 4 harmonic of incoming calls 3.96 11449.25
6 Skewness of 5 harmonic of incoming calls 3.66 9595.91
7 Skewness of temporal entropy of internet activity 3.51 9498.52
8 Sum of spectral distribution skewness of outgoing calls 3.47 2108.19
9 Sum of temporal distribution kurtosis of internet activity 3.35 2267.17
10 Spatial variance of spectral variance of incoming sms 3.30 3971.29
11 Sum of 5 harmonic of incoming calls 3.27 9468.86
12 Sum of calling direction area codes 3.25 6815.90
13 Spectral varince of outgoing sms activity total in time 3.20 13899.48
14 Spatial skewness spectral skewness distribution of internet activity 3.19 3512.42
15 Kurtosis of temporal mean distribition of outgoing sms 3.17 27704.10
16 Spectral varince of temporal median distribition of outgoing sms 3.02 15366.03
17 Spatial median of incoming temporal sum 2.99 2642.34
18 Spectral varince of outgoing sms fundamental frequency 2.96 11624.28
19 Skewness of incoming calls temporal entropy 2.92 4174.09
20 Sum of outgoing calls 5 harmonic 2.89 10251.36
21 Variance of internet activity temporal entropy 2.86 5382.14
22 Median of calling direction area codes 2.85 3591.92
23 Sum of outgoing calls 4 harmonic 2.84 4440.13
24 Median of incoming calls 16 harmonic 2.74 943.9847 / 56
![Page 51: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/51.jpg)
Spatially Mapped Max Energy Consumption vs Telecom Pat-
terns for a Weekday on a SET grid
48 / 56
![Page 52: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/52.jpg)
Energy Consumption Prediction: Results
Energy Consumption Problem Metrics Comparison
Mean Daily Consumption Model
Metric Baseline Model
MAE 20.8468 12.3683
RAE 98.9169 58.6869
MSE 790.6041 325.2679
RMSE 28.1177 18.0352
RSE 100.9551 41.5346
R2 -0.0096 0.5847
Peak Daily Consumption Model
Metric Baseline Model
MAE 186.1440 17.3112
RAE 621.7292 57.8201
MSE 36062.7851 601.7531
RMSE 189.9020 24.5307
RSE 2551.8602 42.5810
R2 -24.5186 0.5742
49 / 56
![Page 53: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/53.jpg)
Summary and Future Work
![Page 54: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/54.jpg)
Computing Human Behavior: The New Kind of Science
The New Kind of Science
• Analytical mathematical solution?
• Numerical optimization to find the best parameters?
• Learn model from data?
• Algorithmic representation of the source processes that drive the
data generation
New paradigm?
• “computationally irreducible”
• “intrinsically random”
• “heterogeneous parallel computing”
• “compressed representation”
• “deep neural representation”,
• “antifragile systems”50 / 56
![Page 55: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/55.jpg)
Supervised Learning of Individual Characteristics
Feasibility
The analysis of spatio-temporal dynamics of human interactions,
observed through different sensory modalities, allows decent inference
and customization.
• Which types of messages are communicated by the data driven
behavioral signals?
• Which human communication patterns and human dynamics cues
convey information about a certain type of behavioral signals?
• Is it practically applicable to create intrinsically data driven models
of individual human behavior based on communication patterns
extracted from telecom metadata?
51 / 56
![Page 56: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/56.jpg)
Supervised Learning of Large Scale Group Behavior
Feasibility
Group social signals are characterized by what “directly or indirectly
provides information about social facts”
• Human dynamics, extracted from aggregated and anonymized
mobile phone data, is a good proxies for modelling energy
consumption and crime hotspots prediction.
• Could be learned from data in human interpretable way without
using autoencoding techniques or neural feature representations.
• Supervised learning of large scale group behavior is feasible from
telecom metadata.
52 / 56
![Page 57: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/57.jpg)
Challenges I
1. Multimodal nature of human behavior understanding
• Volatile data collection
• How to interpret if feature representation is learned?
• How to back track the decision rule?
2. Adversarial nature of human behavior
• Adversarial planning, multi-agent pathfinding, heuristic search?
• Reinforcement learning
• Inverse reinforcement learning (no reward function is given)
• Generative adversarial networks
3. Partially observable data
53 / 56
![Page 58: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/58.jpg)
Challenges II
4. Big Data Privacy
• Embedded privacy in machine learning algorithms
• Ethics of automated human behavior understanding
5. Reverse engineering of the new kind of science
• Finding the possible simple algorithms?
• Causality?
54 / 56
![Page 59: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/59.jpg)
Questions?
55 / 56
![Page 61: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/61.jpg)
Appendix
![Page 62: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/62.jpg)
References I
http://archive.ics.uci.edu/ml/datasets/Human+Activity+
Recognition+Using+Smartphones, 2013.
Federation Aeronautique Internationale. Annex B.
http://www.fai.org/downloads/igc/SC3_AnnexB, Access
date: 04-SEP-2013.
Funf open sensing framework.
http://www.funf.org, Access date: 12-JUN-2013.
Track your happiness web application.
http://www.trackyourhappiness.org, Access date:
12-JUN-2013.
Friends and Family Dataset.
http:
//realitycommons.media.mit.edu/friendsdataset.html,
Access date: 15-JAN-2014.
![Page 63: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/63.jpg)
References II
N. Aharony, W. Pan, C. Ip, I. Khayal, and A. Pentland.
Social fmri: Investigating and shaping social mechanisms in
the real world.
Pervasive and Mobile Computing, 7(6):643–659, 2011.
N. Aharony, W. Pan, C. Ip, I. Khayal, and A. Pentland.
The social fmri: measuring, understanding, and designing
social mechanisms in the real world.
In Proceedings of the 13th international conference on Ubiquitous
computing, pages 445–454. ACM, 2011.
H. Allcott and S. Mullainathan.
Behavior and energy policy.
Science, 327(5970):1204–1205, 2010.
![Page 64: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/64.jpg)
References III
C. A. Anderson.
Temperature and aggression: ubiquitous effects of heat on
occurrence of human violence.
Psychological bulletin, 106(1):74, 1989.
D. Anguita, A. Ghio, L. Oneto, X. Parra, and J. L. Reyes-Ortiz.
Human activity recognition on smartphones using a multiclass
hardware-friendly support vector machine.
In IWAAL, pages 216–223, 2012.
P. L. Bartlett and M. Traskin.
Adaboost is consistent.
Journal of Machine Learning Research, 8(2):347–2, 2007.
![Page 65: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/65.jpg)
References IV
G. Bauer and P. Lukowicz.
Can smartphones detect stress-related changes in the
behaviour of individuals?
In Pervasive Computing and Communications Workshops (PERCOM
Workshops), 2012 IEEE International Conference on, pages 423–426.
IEEE, 2012.
M. Berlingerio, F. Calabrese, G. Lorenzo, R. Nair, F. Pinelli, and
M. Sbodio.
Allaboard: A system for exploring urban mobility and
optimizing public transport using cellphone data.
In Machine Learning and Knowledge Discovery in Databases, volume
8190 of Lecture Notes in Computer Science, pages 663–666.
Springer Berlin Heidelberg.
![Page 66: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/66.jpg)
References V
M. Berlingerio, F. Calabrese, G. Lorenzo, R. Nair, F. Pinelli, and
M. Sbodio.
Allaboard: A system for exploring urban mobility and
optimizing public transport using cellphone data.
In Machine Learning and Knowledge Discovery in Databases, pages
663–666. Springer, 2013.
G. Biau.
Analysis of a random forests model.
The Journal of Machine Learning Research, 98888(1):1063–1095,
2012.
T. W. Bickmore, Rosalind, and W. Picard.
Establishing and maintaining long-term human-computer
relationships.
ACM Transactions on Computer Human Interaction, 12:293–327,
2005.
![Page 67: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/67.jpg)
References VI
C. M. Bishop.
Training with noise is equivalent to Tikhonov regularization.
Neural Computation, 7(1):108–116, 1995.
V. D. Blondel, M. Esch, C. Chan, F. Clerot, P. Deville, E. Huens,
F. Morlot, Z. Smoreda, and C. Ziemlicki.
Data for development: the d4d challenge on mobile phone
data.
arXiv preprint arXiv:1210.0137, 2012.
S. L. Boggs.
Urban crime patterns.
American Sociological Review, 30(6):pp. 899–908, 1965.
![Page 68: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/68.jpg)
References VII
A. Bogomolov, B. Lepri, M. Ferron, F. Pianesi, and A. S. Pentland.
Daily stress recognition from mobile phone data, weather
conditions and individual traits.
In Proceedings of the 22nd ACM international conference on
Multimedia, pages 477–486. ACM, 2014.
A. Bogomolov, B. Lepri, M. Ferron, F. Pianesi, and A. S. Pentland.
Pervasive stress recognition for sustainable living.
In Pervasive Computing and Communications (PERCOM), 2014
IEEE International Conference on, pages 345–350. IEEE, 2014.
A. Bogomolov, B. Lepri, R. Larcher, F. Antonelli, F. Pianesi, and
A. Pentland.
Energy consumption prediction using people dynamics derived
from cellular network data.
EPJ Data Science, 5(1):13, 2016.
![Page 69: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/69.jpg)
References VIII
A. Bogomolov, B. Lepri, and F. Pianesi.
Happiness recognition from mobile phone data.
In Social Computing (SocialCom), 2013 International Conference on,
pages 790–795. IEEE, 2013.
A. Bogomolov, B. Lepri, and F. Pianesi.
Happiness recognition from mobile phone data.
In Social Computing (SocialCom), 2013 International Conference on,
pages 790–795, 2013.
A. Bogomolov, B. Lepri, and F. Pianesi.
Happiness recognition from mobile phone data.
Proceedings of the 2013 International Conference on Social
Computing (SocialCom 2013), pages 790–795, 2013.
A. Bogomolov, B. Lepri, and F. Pianesi.
Happiness recognition from mobile phone data.
In SocialCom 2013, pages 790–795, 2013.
![Page 70: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/70.jpg)
References IX
A. Bogomolov, B. Lepri, and F. Pianesi.
Generalized compression dictionary distance as universal
similarity measure.
In Big Data from Space (BiDS’14), number doi: 10.2788/1823,
pages 63–66. Publications Office of the European Union, 2014.
A. Bogomolov, B. Lepri, J. Staiano, E. Letouze, N. Oliver,
F. Pianesi, and A. Pentland.
Moves on the street: Classifying crime hotspots using
aggregated anonymized data on people dynamics.
Big Data, 3(3):148–158, 2015.
A. Bogomolov, B. Lepri, J. Staiano, E. Letouze, N. Oliver,
F. Pianesi, and A. Pentland.
Moves on the street: Predicting crime hotspots using
aggregated anonymized data on people dynamics.
Big Data Journal. Forthcoming., 2015.
![Page 71: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/71.jpg)
References X
A. Bogomolov, B. Lepri, J. Staiano, N. Oliver, F. Pianesi, and
A. Pentland.
Once upon a crime: towards crime prediction from
demographics and mobile data.
In Proceedings of the 16th international conference on multimodal
interaction, pages 427–434. ACM, 2014.
A. Bogomolov and G. Medvedev.
Mathematical properties of option pricing under gamma
distribution process.
Available at SSRN 1456960, 2009.
N. Bolger and E. A. Schilling.
Personality and the problems of everyday life: The role of
neuroticism in exposure and reactivity to daily stressors.
Journal of Personality, 59(3):355–386, 1991.
![Page 72: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/72.jpg)
References XI
N. Bolger and A. Zuckerman.
A framework for studying personality in the stress process.
Journal of Personality and Social Psychology, 69:890–902, 1995.
J. Bollen, H. Mao, and X. Zeng.
Twitter mood predicts the stock market.
Journal of Computational Science, 2(1):1–8, 2011.
F. Both, M. Hoogendoorn, M. C. A. Klein, and J. Treur.
Computational modeling and analysis of the role of physical
activity in mood regulation and depression.
In Proceedings of the 17th international conference on Neural
information processing: theory and algorithms - Volume Part I,
ICONIP’10, pages 270–281, Berlin, Heidelberg, 2010.
Springer-Verlag.
![Page 73: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/73.jpg)
References XII
K. Bousmalis, S. Zafeiriou, L. Morency, and M. Pantic.
Innite hidden conditional random fields for human behavior
analysis.
IEEE Transactions on Neural Networks and Learning Systems,
24:170–177, 2013.
G. E. P. Box and D. R. Cox.
An Analysis of Transformations.
Journal of the Royal Statistical Society. Series B (Methodological),
26(2):211–252, 1964.
J. Braithwaite.
Crime, Shame and Reintegration.
Ambridge: Cambridge University Press, 1989.
P. J. Brantingham and P. L. Brantingham.
Environmental criminology.
Sage Publications Beverly Hills, CA, 1981.
![Page 74: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/74.jpg)
References XIII
P. J. Brantingham and P. L. Brantingham.
Patterns in crime.
Macmillan New York, 1984.
P. L. Brantingham and P. J. Brantingham.
A theoretical model of crime hot spot generation.
Studies on Crime & Crime Prevention, 1999.
S. Brave and C. Nass.
The human-computer interaction handbook.
chapter Emotion in human-computer interaction, pages 81–96. L.
Erlbaum Associates Inc., Hillsdale, NJ, USA, 2003.
L. Breiman.
L. Breiman.
Bagging predictors.
Machine Learning, 24(2):123–140, 1996.
![Page 75: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/75.jpg)
References XIV
L. Breiman.
Random forests-random features.
Technical report, Technical Report 567, Department of Statistics,
UC Berkeley, 1999. 31, 1999.
L. Breiman.
Random forests.
Mach. Learn., 45(1):5–32, Oct. 2001.
E. L. Broek.
Ubiquitous emotion-aware computing.
Personal Ubiquitous Comput., 17(1):53–67, Jan. 2013.
A. L. Buczak and C. M. Gifford.
Fuzzy association rule mining for community crime pattern
discovery.
In ACM SIGKDD Workshop on Intelligence and Security Informatics,
page 2. ACM, 2010.
![Page 76: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/76.jpg)
References XV
M. D. Buhmann and M. D. Buhmann.
Radial Basis Functions.
Cambridge University Press, New York, NY, USA, 2003.
C. Busso, Z. Deng, S. Yildirim, M. Bulut, C. M. Lee,
A. Kazemzadeh, S. Lee, U. Neumann, and S. Narayanan.
Analysis of emotion recognition using facial expressions,
speech and multimodal information.
In in Sixth International Conference on Multimodal Interfaces ICMI
2004, pages 205–211. ACM Press, 2004.
S. Butterworth.
On the theory of filter amplifiers.
Experimental Wireless and the Wireless Engineer, 7:536–541, 1930.
![Page 77: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/77.jpg)
References XVI
A. Campbell, P. E. Converse, and W. L. Rodgers.
The quality of American life : perceptions, evaluations, and
satisfactions / Angus Campbell, Philip E. Converse, Willard L.
Rodgers.
Russell Sage Foundation New York, 1976.
J. Caplan and L. Kennedy.
Risk terrain modeling manual: Theoretical framework and
technical steps of spatial risk assessment for crime analysis.
Rutgers Center on Public Security, 2010.
J. M. Caplan, L. W. Kennedy, and J. Miller.
Risk terrain modeling: Brokering criminological theory and gis
methods for crime forecasting.
Justice Quarterly, 28, 2011.
![Page 78: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/78.jpg)
References XVII
C. R. Carlson, F. P. Gantz, and J. C. Masters.
Adults’ emotional states and recognition of emotion in young
children.
Motivation and Emotion, 7(1):81–101, 1983.
C. S. Carver and M. F. Scheier.
Origins and Functions of Positive and Negative Affect: A
Control-Process View.
Psychological Review, 97(1):19–35, Jan. 1990.
S. Chainey, L. Tompson, and S. Uhlig.
The utility of hotspot mapping for predicting spatial patterns
of crime.
Security Journal, 21:4–28, 2008.
G. Chittaranjan, J. Blom, and D. Gatica-Perez.
Mining large-scale smartphone data for personality studies.
Personal Ubiquitous Comput., 17(3):433–450, Mar. 2013.
![Page 79: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/79.jpg)
References XVIII
G. Chittaranjan, J. Blom, and D. Gatica-Perez.
Mining large-scale smartphone data for personality studies.
Personal and Ubiquitous Computing, 17:433–450, 2013.
J. Cohen.
A Coefficient of Agreement for Nominal Scales.
Educational and Psychological Measurement, 20(1):37–46, Apr.
1960.
L. Cohen and M. Felson.
Social change and crime rate trends: A routine activity
approach.
American Sociological Review, 44:588–608, 1979.
S. Cohen, K. R. C., and L. U. Gordon.
Measuring stress: A guide for health and social scientists.
Oxford University Press, USA, 1997.
![Page 80: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/80.jpg)
References XIX
J. F. Cohn.
Foundations of human computing: facial expression and
emotion.
In Proceedings of the ICMI 2006 and IJCAI 2007 international
conference on Artifical intelligence for human computing,
ICMI’06/IJCAI’07, pages 1–16, Berlin, Heidelberg, 2007.
Springer-Verlag.
P. T. Costa and R. R. McCrae.
Influence of extraversion and neuroticism on subjective
well-being: Happy and unhappy people.
Journal of Personality and Social Psychology, 38:668–678, 1980.
P. T. Costa and R. R. McCrae.
Revised NEO personality inventory (NEO PI-R) and neo
five-factor inventory (NEO-FFI), volume 0.
Psychological Assessment Resources, 1992.
![Page 81: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/81.jpg)
References XX
P. T. Costa and R. R. McCrae.
Domains and facets: hierarchical personality assessment using
the revised NEO personality inventory.
Journal of Personality Assessment, 64(1):21–50, 1995.
R. Cowie, E. Douglas-Cowie, K. Karpouzis, G. Caridakis,
M. Wallace, and S. Kollias.
Recognition of emotional states in natural human-computer
interaction.
In D. Tzovaras, editor, Multimodal User Interfaces, Signals and
Commmunication Technologies, pages 119–153. Springer Berlin
Heidelberg, 2008.
R. Cowie, E. Douglas-Cowie, N. Tsapatsoulis, G. Votsis, S. Kollias,
W. Fellenz, and J. G. Taylor.
Emotion recognition in human-computer interaction.
Signal Processing Magazine, IEEE, 18(1):32–80, 2001.
![Page 82: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/82.jpg)
References XXI
J. Cullen and S. Levitt.
Crime, urban flight, and the consequences for the cities.
Review of Economics and Statistics, (81):159–169, 2009.
Y.-A. de Montjoye, J. Quoidbach, F. Robic, and A. Pentland.
Predicting personality using novel mobile phone-based metrics.
In Greenberg et al. [109], pages 48–55.
R. de Oliveira, A. Karatzoglou, P. Concejero Cerezo, A. Armenta
Lopez de Vicuna, and N. Oliver.
Towards a psychographic user model from mobile phone usage.
In CHI’11 Extended Abstracts on Human Factors in Computing
Systems, pages 2191–2196. ACM, 2011.
![Page 83: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/83.jpg)
References XXII
K. DeNeve and H. Cooper.
The happy personality: a meta-analysis of 137 personality
traits and subjective well-being.
Psychol Bull, 124(2):197–229, 1998.
J. Denissen, L. Butalid, L. Penke, and M. Van Aken.
The effects of weather on daily mood: A multilevel approach.
Emotion Researcher, 8(5):662–667, 2008.
J. J. Denissen, L. Butalid, L. Penke, and M. A. Van Aken.
The effects of weather on daily mood: a multilevel approach.
Emotion, 8(5):662, 2008.
P. J. Denning.
Great principles of computing.
Communications of the ACM, 46(11):15–20, 2003.
![Page 84: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/84.jpg)
References XXIII
E. Diener.
Assessing subjective well-being: Progress and opportunities.
Social Indicators Research, 31(2):103–157, 1994.
E. Diener and M. E. Seligman.
Very happy people.
Psychological Science, 13(1):81–84, 2002.
R. J. Dolan.
Emotion, cognition, and behavior.
Science (New York, N.Y.), 298(5596), Nov. 2002.
W. Dong, B. Lepri, and A. Pentland.
Modeling the co-evolution of behaviors and social relationships
using mobile phone data.
In MUM, pages 134–143, 2011.
![Page 85: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/85.jpg)
References XXIV
W. Dong, B. Lepri, and A. Pentland.
Modeling the co-evolution of behaviors and social relationships
using mobile phone data.
In MUM 2011, 2011.
C. Duggan, P. Sham, A. Lee, C. Minne, and R. Murray.
Neuroticism: a vulnerability marker for depression evidence
from a family study.
Journal of Affective Disorders, 35(3):139 – 143, 1995.
N. Eagle, M. Macy, and R. Claxton.
Network diversity and economic development.
Science, 328(5981):1029–1031, 2010.
N. Eagle and A. Pentland.
Reality mining: sensing complex social systems.
Personal and Ubiquitous Computing, 10(4):255–268, 2006.
![Page 86: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/86.jpg)
References XXV
N. Eagle, A. S. Pentland, and D. Lazer.
Inferring friendship network structure by using mobile phone
data.
Proceedings of the National Academy of Sciences,
106(36):15274–15278, 2009.
N. Eagle, A. S. Pentland, and D. Lazer.
Inferring friendship network structure by using mobile phone
data.
Proceedings of the National Academy of Sciences,
106(36):15274–15278, 2009.
J. Eck, S. Chainey, J. Cameron, and R. Wilson.
Mapping crime: understanding hotspots.
National Institute of Justice: Washington DC, 2005.
![Page 87: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/87.jpg)
References XXVI
B. Efron and R. J. Tibshirani.
An Introduction to the Bootstrap.
Chapman & Hall, New York, 1993.
I. Ehrlich.
On the relation between education and crime.
1975.
L. Ellis, K. Beaver, and J. Wright.
Handbook of crime correlates.
Academic Press, 2009.
T. Esch, G. L. Fricchione, G. B. Stefano, et al.
The therapeutic use of the relaxation response in
stress-related diseases.
Med Sci Monit, 9(2):23–34, 2003.
![Page 88: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/88.jpg)
References XXVII
V. Faust, M. Weidmann, and W. Wehner.
The influence of meteorological factors on children and youths:
A 10% random selection of 16,000 pupils and apprentices of
basle city (switzerland).
Acta Paedopsychiatrica: International Journal of Child & Adolescent
Psychiatry, 1974.
L. A. Feldman.
Valence focus and arousal focus: Individual differences in the
structure of affective experience.
Journal of personality and social psychology, 69:153–153, 1995.
G. Fink.
Stress consequences: Mental, neuropsychological and
socioeconomic.
Elsevier, San Diego, CA, 2010.
![Page 89: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/89.jpg)
References XXVIII
G. Fink.
Stress consequences: Mental, neuropsychological and
socioeconomic.
Elsevier, San Diego, CA, 2010.
R. B. Freeman.
The economics of crime.
Handbook of labor economics, 3:3529–3571, 1999.
Y. Freund and R. E. Schapire.
A decision-theoretic generalization of on-line learning and an
application to boosting.
Journal of Computer and System Sciences, 55(1):119 – 139, 1997.
![Page 90: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/90.jpg)
References XXIX
E. Frias-Martinez, G. Williamson, and V. Frias-Martinez.
An agent-based model of epidemic spread using human
mobility and social network information.
In Social Computing (SocialCom), 2011 International Conference on,
pages 57–64. IEEE, 2011.
J. H. Friedman.
Greedy function approximation: a gradient boosting machine.
Annals of Statistics, pages 1189–1232, 2001.
J. H. Friedman.
Stochastic gradient boosting.
Computational Statistics & Data Analysis, 38(4):367–378, 2002.
![Page 91: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/91.jpg)
References XXX
M. Frigo and S. G. Johnson.
The design and implementation of FFTW3.Proceedings of the IEEE, 93(2):216–231, 2005.
Special issue on ’Program Generation, Optimization, and Platform
Adaptation’.
R. Gallotti, A. Bazzani, M. Degli Esposti, and S. Rambaldi.
Entropic measures of individual mobility patterns.
Journal of Statistical Mechanics: Theory and Experiment, 2013.
S. Garcia and F. Herrera.
An Extension on ”Statistical Comparisons of Classifiers over
Multiple Data Sets” for all Pairwise Comparisons.
Journal of Machine Learning Research, 9:2677–2694, Dec. 2008.
G. George, M. R. Haas, and A. Pentland.
Big data and management.
Academy of Management Journal, 57(2):321–326, 2014.
![Page 92: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/92.jpg)
References XXXI
D. Giakoumis, A. Drosou, P. Cipresso, D. Tzovaras, G. Hassapis,
A. Gaggioli, and G. Riva.
Real-time monitoring of behavioural parameters related to
psychological stress.
Studies in health technology and informatics, 181:287, 2012.
C. Gini.
Concentration and dependency ratios.
Rivista di Politica Economica, 87:769–792, 1997.
M. Gonzalez, C. Hidalgo, and L. Barabasi.
Understanding individual mobility patterns.
Nature, 453(7196):779–782, 2008.
W. J. Goode.
A theory of role strain.
American Sociological Review, 25(4):pp. 483–496, 1960.
![Page 93: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/93.jpg)
References XXXII
J. A. Gray.
Neural systems, emotion, and personality.
In Neurobiology of learning, emotion, and affect / editor, John
Madden IV, pages 273–306. Raven Press New York, 1991.
A. M. Greenberg, W. G. Kennedy, and N. Bos, editors.
Social Computing, Behavioral-Cultural Modeling and
Prediction - 6th International Conference, SBP 2013,
Washington, DC, USA, April 2-5, 2013. Proceedings, volume
7812 of Lecture Notes in Computer Science. Springer, 2013.
A. Gudmundsson and N. Mohajeri.
Entropy and order in urban street networks.
Scientific Reports, (3324), 2013.
H. Gunes and M. Pantic.
Automatic, dimensional and continuous emotion recognition.
Int’l Journal of Synthetic Emotion, 1(1):68–99, 2010.
![Page 94: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/94.jpg)
References XXXIII
H. Gunes, B. Schuller, M. Pantic, and R. Cowie.
Emotion representation, analysis and synthesis in continuous
space: A survey.
In Proceedings of IEEE International Conference on Automatic Face
and Gesture Recognition (FG’11), EmoSPACE 2011 - 1st
International Workshop on Emotion Synthesis, rePresentation, and
Analysis in Continuous spacE, pages 827–834, Santa Barbara, CA,
USA, March 2011.
L. H. A. S. Gunthert, Kathleen Cimbolic; Cohen.
The role of neuroticism in daily stress and coping.
Journal of Personality and Social Psychology, 77:1087–1100, 1999.
I. Guyon and A. Elisseeff.
An introduction to variable and feature selection.
J. Mach. Learn. Res., 3:1157–1182, Mar. 2003.
![Page 95: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/95.jpg)
References XXXIV
D. Hand and R. Till.
A simple generalisation of the area under the ROC curve for
multiple class classification problems.
Machine Learning, 45:171–186, 2001.
J. Hardt and H. Gerbershagen.
No changes in mood with the seasons: observations in 3000
chronic pain patients.
Acta Psychiatrica Scandinavica, 100(4):288–294, 1999.
J. Healey and R. Picard.
Detecting stress during real-world driving tasks using
physiological sensors.
Intelligent Transportation Systems, IEEE Transactions on,
6(2):156–166, 2005.
![Page 96: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/96.jpg)
References XXXV
J. Hernandez, M. E. Hoque, and R. W. Picard.
Mood meter: large-scale and long-term smile monitoring
system.
In ACM SIGGRAPH 2012 Emerging Technologies, SIGGRAPH ’12,
pages 15:1–15:1, New York, NY, USA, 2012. ACM.
E. Howarth and M. S. Hoffman.
A multidimensional approach to the relationship between
mood and weather.
British Journal of Psychology, 75(1):15–23, 1984.
T. Hua, F. Chen, L. Zhao, C.-T. Lu, and N. Ramakrishnan.
Sted: Semi-supervised targeted-interest event detection in
twitter.
In Proceedings of the 19th ACM SIGKDD International Conference
on Knowledge Discovery and Data Mining, KDD ’13, pages
1466–1469, New York, NY, USA, 2013. ACM.
![Page 97: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/97.jpg)
References XXXVI
A. G. Janecek, W. N. Gansterer, M. A. Demel, and G. F. Ecker.
On the Relationship Between Feature Selection and
Classification Accuracy.
In JMLR: Workshop and Conference Proceedings 4, pages 90–105,
2008.
O. P. John and S. Srivastava.
The Big Five trait taxonomy: History, measurement, and
theoretical perspectives.
In L. A. Pervin and O. P. John, editors, Handbook of Personality:
Theory and Research, pages 102–138. Guilford Press, New York,
second edition, 1999.
![Page 98: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/98.jpg)
References XXXVII
E. Jovanov, A. O’Donnell Lords, D. Raskovic, P. Cox, R. Adhami,
and F. Andrasik.
Stress monitoring using a distributed wireless intelligent sensor
system.
Engineering in Medicine and Biology Magazine, IEEE, 22(3):49–55,
2003.
A. Kapoor, W. Burleson, and R. W. Picard.
Automatic prediction of frustration.
Int. J. Hum.-Comput. Stud., 65(8):724–736, Aug. 2007.
A. Kapoor and R. W. Picard.
Multimodal affect recognition in learning environments.
In Proceedings of the 13th annual ACM international conference on
Multimedia, MULTIMEDIA ’05, pages 677–682, New York, NY,
USA, 2005. ACM.
![Page 99: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/99.jpg)
References XXXVIII
M. C. Keller, B. L. Fredrickson, O. Ybarra, S. Cote, K. Johnson,
J. Mikels, A. Conway, and T. Wager.
A Warm Heart and a Clear Head: The Contingent Effects of
Weather on Mood and Cognition.
Psychological Science, 16(9):724–731, Sept. 2005.
J. Z. Kolter and J. Ferreira Jr.
A large-scale study on predicting and contextualizing building
energy usage.
2011.
D. Krackhardt.
The strength of strong ties: The importance of philos in
organizations.
pages 216–239, 1992.
![Page 100: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/100.jpg)
References XXXIX
C. Krumme, A. Llorente, M. Cebrian, A. Pentland, and E. Moro.
The predictability of consumer visitation patterns.
Scientific Reports, (1645), 2013.
W. J. Krzanowski.
Misclassification rates.
In Encyclopedia of Statistics in Behavioral Science. 2005.
N. Lane, M. Mohammod, M. Lin, X. Yang, H. Lu, S. Ali, A. Doryab,
E. Berke, T. Choudhury, and A. Campbell.
Bewell: A smartphone application to monitor, model and
promote wellbeing.
IEEE, 4 2012.
N. D. Lane, E. Miluzzo, H. Lu, D. Peebles, T. Choudhury, and A. T.
Campbell.
A survey of mobile phone sensing.
Comm. Mag., 48(9):140–150, Sept. 2010.
![Page 101: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/101.jpg)
References XL
N. D. Lane, E. Miluzzo, H. Lu, D. Peebles, T. Choudhury, and A. T.
Campbell.
A survey of mobile phone sensing.
IEEE Communications Magazine, 48(9):140–150, 2010.
N. Lathia, K. K. Rachuri, C. Mascolo, and P. J. Rentfrow.
Contextual dissonance: Design bias in sensor-based experience
sampling methods.
Proceedings of Ubicomp 2013, pages 183–192, 2013.
J. Laurila, D. Gatica-Perez, I. Aad, J. Blom, O. Bornet, T. Do,
O. Dousse, J. Eberle, and M. Miettinen.
From big smartphone data to worldwide research: The mobile
data challenge.
Pervasive and Mobile Computing, 9:752–771, 2013.
![Page 102: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/102.jpg)
References XLI
D. Lazer, A. Pentland, L. Adamic, S. Aral, A.-L. Barabasi,
D. Brewer, N. Christakis, N. Contractor, J. Fowler, M. Gutmann,
T. Jebara, G. King, M. Macy, D. Roy, and M. Van Alstyne.
Computational social science.
Science, 323(5915):721–723, 2009.
G. R. Lee and M. Ishii-Kuntz.
Social Interaction, Loneliness, and Emotional Well-Being
among the Elderly.
Research on Aging, 9(4):459–482, Dec. 1987.
J. Lester, T. Choudhury, and G. Borriello.
A practical approach to recognizing physical activities.
In K. Fishkin, B. Schiele, P. Nixon, and A. Quigley, editors, Pervasive
Computing, volume 3968 of Lecture Notes in Computer Science,
pages 1–16. Springer Berlin Heidelberg, 2006.
![Page 103: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/103.jpg)
References XLII
M. LEWIS and J. HAVILAND-JONES.
Handbook of Emotions.
Guilford Publications, Incorporated, 2004.
R. Li, K. H. Lei, R. Khadiwala, and K.-C. Chang.
Tedas: A twitter-based event detection and analysis system.
In Data Engineering (ICDE), 2012 IEEE 28th International
Conference on, pages 1273–1276. IEEE, 2012.
R. LiKamWa, Y. Liu, N. D. Lane, and L. Zhong.
Moodscope: building a mood sensor from smartphone usage
patterns.
In Proceeding of the 11th annual international conference on Mobile
systems, applications, and services, MobiSys ’13, pages 389–402,
New York, NY, USA, 2013. ACM.
![Page 104: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/104.jpg)
References XLIII
A. Lima, M. De Domenico, V. Pejovic, and M. Musolesi.
Exploiting cellular data for disease containment and
information campaigns strategies in country-wide epidemics.
arXiv preprint arXiv:1306.4534, 2013.
W.-J. Lin and J. J. Chen.
Class-imbalanced classifiers for high-dimensional data.
Briefings in Bioinformatics, 14(1):13–26, Jan. 2013.
H. Lu, D. Frauendorfer, M. Rabbi, M. S. Mast, G. T. Chittaranjan,
A. T. Campbell, D. Gatica-Perez, and T. Choudhury.
Stresssense: detecting stress in unconstrained acoustic
environments using smartphones.
In Proceedings of the 2012 ACM Conference on Ubiquitous
Computing, UbiComp ’12, pages 351–360, New York, NY, USA,
2012. ACM.
![Page 105: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/105.jpg)
References XLIV
S. Lyubomirsky, L. King, and E. Diener.
The benefits of frequent positive affect: Does happiness lead
to success?
Psychological Bulletin, 131(6):803–855, Nov. 2005.
S. Lyubomirsky, C. Tkach, and R. M. Dimatteo.
What are the differences between happiness and self-esteem?
Social Indicators Research, 78:363–404, 2006.
Y. Ma, B. Xu, Y. Bai, G. Sun, and R. Zhu.
Daily mood assessment based on mobile phone sensing.
In Wearable and Implantable Body Sensor Networks (BSN), 2012
Ninth International Conference on, pages 142–147, 2012.
A. Madan, M. Cebrian, D. Lazer, and A. Pentland.
Social sensing for epidemiological behavior change.
In UbiComp, pages 291–300, 2010.
![Page 106: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/106.jpg)
References XLV
A. Madan, M. Cebrian, S. T. Moturu, K. Farrahi, and A. Pentland.
Sensing the ”health state” of a community.
IEEE Pervasive Computing, 11(4):36–45, 2012.
D. Majoe, P. Bonhof, T. Kaegi-Trachsel, J. Gutknecht, and
L. Widmer.
Stress and sleep quality estimation from a smart wearable
sensor.
In Pervasive Computing and Applications (ICPCA), 2010 5th
International Conference on, pages 14–19, 2010.
R. R. McCrae and P. T. Costa.
Adding liebe und arbeit: The full five-factor model and
well-being.
Personality and Social Psychology Bulletin, 17(2):227–232, 1991.
![Page 107: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/107.jpg)
References XLVI
H. Mehlum, K. Moene, and R. Torvik.
Crime induced poverty traps.
Journal of Development Economics, (77):325–340, 2005.
S. F. Messner, L. Anselin, R. D. Baller, D. F. Hawkins, G. Deane,
and S. E. Tolnay.
The spatial patterning of county homicide rates: An
application of exploratory spatial data analysis.
Journal of Quantitative criminology, 15(4):423–450, 1999.
G. Miller.
Note on the bias of information estimates.
Information theory in psychology: Problems and methods, 2(95):100,
1955.
G. Miller.
The smartphone psychology manifesto.
Perspectives on Psychological Science, 7(3):221–237, 2012.
![Page 108: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/108.jpg)
References XLVII
G. Mohler, M. Short, P. Brantingham, F. Schoenberg, and G. Tita.
Self-exciting point process modeling of crime.
Journal of the American Statistical Association, (106):100–108,
2011.
Y.-A. Montjoye, J. Quoidbach, F. Robic, and A. Pentland.
Predicting personality using novel mobile phone-based metrics.
In A. Greenberg, W. Kennedy, and N. Bos, editors, Social
Computing, Behavioral-Cultural Modeling and Prediction, volume
7812 of Lecture Notes in Computer Science, pages 48–55. Springer
Berlin Heidelberg, 2013.
![Page 109: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/109.jpg)
References XLVIII
Y.-A. Montjoye, J. Quoidbach, F. Robic, and A. Pentland.
Predicting personality using novel mobile phone-based metrics.
In A. Greenberg, W. Kennedy, and N. Bos, editors, Social
Computing, Behavioral-Cultural Modeling and Prediction, volume
7812 of Lecture Notes in Computer Science, pages 48–55. Springer
Berlin Heidelberg, 2013.
Y.-A. Montjoye, J. Quoidbach, F. Robic, and A. Pentland.
Predicting personality using novel mobile phone-based metrics.
In A. Greenberg, W. Kennedy, and N. Bos, editors, Social
Computing, Behavioral-Cultural Modeling and Prediction, volume
7812 of Lecture Notes in Computer Science, pages 48–55. Springer
Berlin Heidelberg, 2013.
![Page 110: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/110.jpg)
References XLIX
S. Moran, I. de Vallejo, K. Nakata, R. Conroy-Dalton, R. Luck,
P. McLennan, and S. Hailes.
Studying the impact of ubiquitous monitoring technology on
office worker behaviours: The value of sharing research data.
In Pervasive Computing and Communications Workshops (PERCOM
Workshops), 2012 IEEE International Conference on, pages 902–907,
2012.
S. T. Moturu, I. Khayal, N. Aharony, W. Pan, and A. Pentland.
Using social sensing to understand the links between sleep,
mood, and sociability.
In SocialCom/PASSAT, pages 208–214, 2011.
A. Muaremi, B. Arnrich, and G. Troster.
A survey on measuring happiness with smart phones.
In 6th International Workshop on Ubiquitous Health and Wellness
(UbiHealth 2012), 2012.
![Page 111: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/111.jpg)
References L
D. G. Myers and E. Diener.
Who is happy?
Psychological Science, 6(1):10–19, 1995.
I. Myers and P. Myers.
Gifts differing: understanding personality type.
Davies-Black Pub., 1980.
M. A. Nicolaou, H. Gunes, and M. Pantic.
Output-associative rvm regression for dimensional and
continuous emotion prediction.
In Proceedings of IEEE International Conference on Automatic Face
and Gesture Recognition (FG’11), pages 16–23, Santa Barbara, CA,
USA, March 2011.
I. C. T. G. Nicu Sebe, Erwin Bakker and T. Huang.
Bimodal emotion recognition.
![Page 112: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/112.jpg)
References LI
M. Nilsson, P. Funk, E. M. Olsson, B. von Schele, and N. Xiong.
Clinical decision-support for diagnosing stress-related disorders
by applying psychophysiological medical knowledge to an
instance-based learning system.
Artificial Intelligence in Medicine, 36(2):159 – 176, 2006.
A. Oikonomopoulos, I. Patras, M. Pantic, and N. Paragios.
Trajectory-based Representation of Human Actions, volume
4451, pages 133–154.
2007.
J. Onnela, J. Saramaki, J. Hyvonen, G. Szabo, D. Lazer, K. Kaski,
J. Kertesz, and A. Barabasi.
Structure and tie strengths in mobile communication networks.
PNAS, 104(18), 2007.
![Page 113: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/113.jpg)
References LII
L. Paninski.
Estimation of entropy and mutual information.
Neural Comput., 15(6):1191–1253, June 2003.
M. Pantic.
Affective Computing, volume 1, pages 8–14.
2005.
M. Pantic.
Affective Computing (revisited), volume 1, pages 15–21.
2009.
M. Pantic.
Machine analysis of facial behaviour: Naturalistic and dynamic
behaviour.
Philosophical Transactions of Royal Society B, 364:3505–3513, 2009.
![Page 114: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/114.jpg)
References LIII
M. Pantic, R. Cowie, F. D’ericco, D. Heylen, M. Mehu,
C. Pelachaud, I. Poggi, M. Schroder, and A. Vinciarelli.
Social Signal Processing: The Research Agenda, pages
511–538.
2011.
M. Pantic, A. Pentland, A. Nijholt, and T. Huang.
Human Computing and Machine Understanding of Human
Behavior: A Survey, volume 4451, pages 47–71.
2007.
E. B. Patterson.
Poverty, income inequality, and community crime rates.
Criminology, 29(4):755–776, 1991.
A. Pentland and A. Liu.
Modeling and prediction of human behavior.
Neural Comput., 11(1):229–242, Jan. 1999.
![Page 115: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/115.jpg)
References LIV
R. W. Picard.
Affective computing.
MIT Press, Cambridge, MA, USA, 1997.
K. Plarre, A. Raij, S. Hossain, A. Ali, M. Nakajima, M. Al’absi,
E. Ertin, T. Kamarck, S. Kumar, M. Scott, D. Siewiorek,
A. Smailagic, and L. Wittmers.
Continuous inference of psychological stress from sensory
measurements collected in the natural environment.
In Information Processing in Sensor Networks (IPSN), 2011 10th
International Conference on, pages 97–108, 2011.
![Page 116: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/116.jpg)
References LV
D. Quercia.
Dont worry, be happy: The geography of happiness on
facebook.
http:
//profzero.org/publications/websci13_dontworry.pdf,
2013.
D. Quercia, J. Ellis, L. Capra, and J. Crowcroft.
Tracking ”gross community happiness” from tweets.
In CSCW, pages 965–968, 2012.
M. Rabbi, S. Ali, T. Choudhury, and E. Berke.
Passive and in-situ assessment of mental and physical
well-being using mobile sensors.
In Proceedings of the 13th international conference on Ubiquitous
computing, UbiComp ’11, pages 385–394, New York, NY, USA,
2011. ACM.
![Page 117: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/117.jpg)
References LVI
K. K. Rachuri, M. Musolesi, C. Mascolo, P. J. Rentfrow,
C. Longworth, and A. Aucinas.
Emotionsense: a mobile phones based adaptive platform for
experimental social psychology research.
In UbiComp, pages 281–290, 2010.
S. Raphael and R. Winter-Ebmer.
Identifying the effect of unemployment on crime.
Journal of Law and Economics, 44(1), 2001.
J. H. Ratcliffe.
A temporal constraint theory to explain opportunity-based
spatial offending patterns.
Journal of Research in Crime and Delinquency, 43(3):261–291, 2006.
![Page 118: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/118.jpg)
References LVII
N. D. L. L. Z. Robert LiKamWa, Yunxin Liu.
Moodscope: Building a mood sensor from smartphone usage
patterns.
http://niclane.org/pubs/mobisys_moodscope.pdf, 2013.
N. E. Rosenthal, D. A. Sack, J. C. Gillin, A. J. Lewy, F. K. Goodwin,
Y. Davenport, P. S. Mueller, D. A. Newsome, and T. A. Wehr.
Seasonal affective disorder: a description of the syndrome and
preliminary findings with light therapy.
Archives of General Psychiatry, 41(1):72, 1984.
C. S, J.-D. D, and M. GE.
Psychological stress and disease.
JAMA, 298(14):1685–1687, 2007.
![Page 119: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/119.jpg)
References LVIII
T. Sakaki, M. Okazaki, and Y. Matsuo.
Earthquake shakes twitter users: Real-time event detection by
social sensors.
In Proceedings of the 19th International Conference on World Wide
Web, WWW ’10, pages 851–860, New York, NY, USA, 2010. ACM.
A. Salah, J. Ruiz-del Solar, . Merili, and P.-Y. Oudeyer.
Human behavior understanding for robotics.
In A. Salah, J. Ruiz-del Solar, . Merili, and P.-Y. Oudeyer, editors,
Human Behavior Understanding, volume 7559 of Lecture Notes in
Computer Science, pages 1–16. Springer Berlin Heidelberg, 2012.
A. A. Salah and B. Lepri.
Challenges of human behavior understanding for inducing
behavioral change.
In AmI Workshops, pages 249–251, 2011.
![Page 120: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/120.jpg)
References LIX
J. L. Sanders and M. S. Brizzolara.
Relationships between weather and mood.
The Journal of General Psychology, 107(1):155–156, 1982.
C. N. Sanderson, C.A.
A life task perspective on personality coherence: Stability
versus change in tasks, goals, strategies, and outcomes.
In D. Cervone & Y. Shoda (Eds.), The Coherence of Personality:
Social-cognitive Bases of Consistency, Variability, and Organization,
pages 372–392, New York, NY, USA, 1999. Guilford Press.
A. Sano and R. W. Picard.
Stress recognition using wearable sensors and mobile phones.
In Affective Computing and Intelligent Interaction (ACII), 2013
Humaine Association Conference on, pages 671–676. IEEE, 2013.
![Page 121: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/121.jpg)
References LX
A. Sano and R. W. Picard.
Stress recognition using wearable sensors and mobile phones.
Proceedings of the 2013 Humaine Association Conference on
Affective Computing and Intelligent Interaction, pages 671–676,
2013.
K. R. Scherer, D. Grandjean, T. Johnstone, G. Klasmeyer, and
T. Banziger.
Acoustic correlates of task load and stress.
In INTERSPEECH, 2002.
B. Schuller, R. Villar, G. Rigoll, and M. Lang.
Meta-classifiers in acoustic and linguistic feature fusion-based
affect recognition.
In Acoustics, Speech, and Signal Processing, 2005. Proceedings.
(ICASSP ’05). IEEE International Conference on, volume 1, pages
325–328, 2005.
![Page 122: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/122.jpg)
References LXI
N. Sebe, M. S. Lew, I. Cohen, A. Garg, and T. S. Huang.
Emotion recognition using a cauchy naive bayes classifier.
In in Proc. ICPR, pages 17–20, 2002.
J. Shang, T. W. Zheng, Y., and E. Chang.
Inferring gas consumption and pollution emission of vehicles
throughout a city.
In KDD 2014, 2014.
C. Shannon.
A mathematical theory of communication.
Bell System Technical Journal, 27:379–423, 623–656, July, October
1948.
![Page 123: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/123.jpg)
References LXII
M. B. Short, M. R. D’Orsogna, V. B. Pasour, G. E. Tita, P. J.
Brantingham, A. L. Bertozzi, and L. B. Chayes.
A statistical model of criminal behavior.
Mathematical Models and Methods in Applied Sciences,
18(supp01):1249–1267, 2008.
R. Sinatra and M. Szell.
Entropy and the predictability of online life.
Entropy, (16):543–556, 2014.
S. R. Singh, H. A. Murthy, and T. A. Gonsalves.
Feature selection for text classification based on gini
coefficient of inequality.
Journal of Machine Learning Research-Proceedings Track, 10:76–85,
2010.
![Page 124: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/124.jpg)
References LXIII
V. K. Singh, L. Freeman, B. Lepri, and A. S. Pentland.
Predicting spending behavior using socio-mobile features.
In Social Computing (SocialCom), 2013 International Conference on,
pages 174–179. IEEE, 2013.
C. Smith-Clarke, A. Mashhadi, and L. Capra.
Poverty on the cheap: Estimating poverty maps using
aggregated mobile communication networks.
In 32nd ACM Conference on Human Factors in Computing Systems
(CHI2014), 2014.
M. Soleymani and M. Pantic.
Emotionally aware tv.
In Proceedings of TVUX-2013: Workshop on Exploring and
Enhancing the User Experience for TV at ACM CHI 2013, Paris,
France, April 2013.
![Page 125: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/125.jpg)
References LXIV
M. Soleymani, M. Pantic, and T. Pun.
Multimodal emotion recognition in response to videos.IEEE Trans. on Affective Computing, 3(2):211–223, November 2012.
in press.
C. Song, Z. Qu, N. Blumm, and A. Barabasi.
Limits of predictability in human mobility.
Science, (327):1018–1021, 2010.
V. Soto, V. Frias-Martinez, J. Virseda, and E. Frias-Martinez.
Prediction of socioeconomic levels using cell phone records.
In UMAP 2011, pages 377–388, 2011.
![Page 126: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/126.jpg)
References LXV
J. Staiano, B. Lepri, N. Aharony, F. Pianesi, N. Sebe, and
A. Pentland.
Friends don’t lie - inferring personality traits from social
network structure.
In ACM Ubicomp 2012, pages 321–330, 2012.
J. Staiano, B. Lepri, N. Aharony, F. Pianesi, N. Sebe, and
A. Pentland.
Friends don’t lie: inferring personality traits from social
network structure.
In Proceedings of the 2012 ACM Conference on Ubiquitous
Computing, pages 321–330. ACM, 2012.
![Page 127: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/127.jpg)
References LXVI
J. Staiano, B. Lepri, R. Subramanian, N. Sebe, and F. Pianesi.
Automatic modeling of personality states in small group
interactions.
In Proceedings of the 19th ACM international conference on
Multimedia, MM ’11, pages 989–992, New York, NY, USA, 2011.
ACM.
S. V. Stehman.
Selecting and interpreting measures of thematic classification
accuracy.
Remote Sensing of Environment, 62(1):77–89, 1997.
M. Stone.
Cross-Validatory Choice and Assessment of Statistical
Predictions.
Journal of the Royal Statistical Society. Series B (Methodological),
36(2):111–147, 1974.
![Page 128: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/128.jpg)
References LXVII
A. Stopczynski, V. Sekara, P. Sapiezynski, A. Cuttone, M. M.
Madsen, J. E. Larsen, and S. Lehmann.
Measuring large-scale social networks with high resolution.
PLoS ONE, (9), 2014.
J. Suls, P. Green, and S. Hillis.
Emotional reactivity to everyday problems, affective inertia,
and neuroticism.
Personality and Social Psychology Bulletin, 24(2):127–136, 1998.
R. Tarling and K. Morris.
Reporting crime to the police.
British Journal of Criminology, (50):474–479, 2010.
![Page 129: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/129.jpg)
References LXVIII
G. Tartarisco, G. Baldus, D. Corda, R. Raso, A. Arnao, M. Ferro,
A. Gaggioli, and G. Pioggia.
Personal health system architecture for stress monitoring and
support to clinical decisions.
Computer Communications, 35(11):1296 – 1305, 2012.
R. C. Team.
R: A language and environment for statistical computing.
http://www.R-project.org, 2012.
J. L. Toole, N. Eagle, and J. B. Plotkin.
Spatiotemporal correlations in criminal offense records.
ACM Trans. Intell. Syst. Technol., 2(4):38:1–38:18, July 2011.
Y. Tsutsui.
Weather and individual happiness.
Weather, Climate, and Society, 5(1):70–82, 2013/06/12 2012.
![Page 130: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/130.jpg)
References LXIX
A. Tumasjan, T. O. Sprenger, P. G. Sandner, and I. M. Welpe.
Predicting elections with twitter: What 140 characters reveal
about political sentiment.
ICWSM, 10:178–185, 2010.
E. Tuv, A. Borisov, G. Runger, and K. Torkkola.
Feature selection with ensembles, artificial variables, and
redundancy elimination.
The Journal of Machine Learning Research, 10:1341–1366, 2009.
M. Unser.
Sampling–50 years after Shannon.
In Proceedings of the IEEE, pages 569–587, 2000.
M. Unser.
Sampling–50 years after Shannon.
In Proceedings of the IEEE, pages 569–587, 2000.
![Page 131: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/131.jpg)
References LXX
M. Unser.
Sampling–50 years after Shannon.
In Proceedings of the IEEE, pages 569–587, 2000.
V. N. Vapnik.
The nature of statistical learning theory.
Springer-Verlag New York, Inc., New York, NY, USA, 1995.
J. Vitters.
Personality traits and subjective well-being: emotional
stability, not extraversion, is probably the important predictor.
Personality and Individual Differences, 31(6):903 – 914, 2001.
M. Vollrath and S. Torgersen.
Personality types and coping.
Personal. Individ. Differ., 29:367–378, 2000.
![Page 132: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/132.jpg)
References LXXI
J. Wagner, E. Andre, F. Lingenfelser, and J. Kim.
Exploring fusion methods for multimodal emotion recognition
with missing data.
IEEE Trans. Affect. Comput., 2(4):206–218, Oct. 2011.
J. Wagner, F. Lingenfelser, E. Andre, J. Kim, and T. Vogt.
Exploring fusion methods for multimodal emotion recognition
with missing data.
IEEE Transactions on Affective Computing, 2(4):206–218, 2011.
M. S. Wandishin and S. J. Mullen.
Multiclass ROC Analysis.
American Meteorological Society, 2009.
T. Wang, C. Rudin, D. Wagner, and R. Sevieri.
Learning to detect patterns of crime.
In Machine Learning and Knowledge Discovery in Databases, pages
515–530. Springer, 2013.
![Page 133: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/133.jpg)
References LXXII
X. Wang, D. E. Brown, and M. S. Gerber.
Spatio-temporal modeling of criminal incidents using
geographic, demographic, and twitter-derived information.
In Intelligence and Security Informatics (ISI), 2012 IEEE
International Conference on, pages 36–41. IEEE, 2012.
X. Wang, M. S. Gerber, and D. E. Brown.
Automatic crime prediction using events extracted from
twitter posts.
In Social Computing, Behavioral-Cultural Modeling and Prediction,
pages 231–238. Springer, 2012.
Y. Wang and L. Guan.
Recognizing human emotional state from audiovisual signals.
Multimedia, IEEE Transactions on, 10(4):659–668, 2008.
![Page 134: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/134.jpg)
References LXXIII
S. K. Weinberg.
Theories of criminality and problems of prediction.
The Journal of Criminal Law, Criminology, and Police Science,
45(4):412–424, 1954.
D. Weisburd.
Place-based policing.
Ideas in American Policing, 9:1–16, 2008.
D. Weisburd and L. Green.
Defining the street-level drug market.
1994.
![Page 135: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/135.jpg)
References LXXIV
D. Weisburd, L. Maher, L. Sherman, M. Buerger, E. Cohn, and
A. Petrisino.
Contrasting crime general and crime specific theory: The case
of hot spots of crime.
New Directions in Criminological Theory. Advances in Criminological
Theory, 4:45–70, 1993.
A. Wesolowski, N. Eagle, A. Tatem, D. Smith, R. Noor, and
C. Buckee.
Quantifying the impact of human mobility on malaria.
Science, 338(6104):267–270, 2012.
R. B. Zajonc and D. N. McIntosh.
Emotions research: Some promising questions and some
questionable promises.
Psychological Science, pages 70–74, 1992.
![Page 136: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/136.jpg)
References LXXV
Z. Zeng, M. Pantic, and T. Huang.
Emotion Recognition based on Multimodal Information, pages
241–266.
Springer, 2009.
Z. Zeng, M. Pantic, G. Roisman, and T. Huang.
A survey of affect recognition methods: Audio, visual, and
spontaneous expressions.
IEEE Transactions on Pattern Analysis and Machine Intelligence,
31(1):39–58, 2009.
Z. Zeng, J. Tu, M. Liu, T. Zhang, N. Rizzolo, Z. Zhang, T. S.
Huang, D. Roth, and S. Levinson.
Bimodal hci-related affect recognition.
In Proceedings of the 6th international conference on Multimodal
interfaces, pages 137–143, 2004.
![Page 137: Predictive Modeling of Human Behavior: Supervised Learning from Telecom Metadata [Doctoral Thesis Presentation]](https://reader031.vdocuments.site/reader031/viewer/2022030318/5a65e1fc7f8b9ad05e8b49d5/html5/thumbnails/137.jpg)
References LXXVI
H. Zhang, R. Dantu, and J. W. Cangussu.
Socioscope: Human relationship and behavior analysis in social
networks.
Systems, Man and Cybernetics, Part A: Systems and Humans, IEEE
Transactions on, 41(6):1122–1143, 2011.
Y. Zheng, L. Capra, O. Wolfson, and H. Yang.
Urban computing: concepts, methodologies, and applications.
ACM Transaction on Intelligent Systems and Technology, 2014.
Y. Zheng, T. Liu, Y. Wang, Y. Liu, Y. Zhu, and E. Chang.
Diagnosing new york city’s noises with ubiquitous data.
In ACM Ubicomp 2014, 2014.