Simulating the “freshers' flu”: an individual-level simulation approach utilising social networking

296 Pages·2016·2.83 MB·English

University of Southampton Research Repository
ePrints Soton
Copyright © and Moral Rights for this thesis are retained by the author and/or other
copyright owners. A copy can be downloaded for personal non-commercial
research or study, without prior permission or charge. This thesis cannot be
reproduced or quoted extensively from without first obtaining permission in writing
from the copyright holder/s. The content must not be changed in any way or sold
commercially in any format or medium without the formal permission of the
copyright holders.

When referring to this work, full bibliographic details including the author, title,
awarding institution and date of the thesis must be given e.g.
AUTHOR (year of submission) "Full thesis title", University of Southampton, name
of the University School or Department, PhD Thesis, pagination
http://eprints.soton.ac.uk
UNIVERSITY OF SOUTHAMPTON

FACULTY OF BUSINESS AND LAW

School of Management

Simulating the “Freshers’ Flu”
An individual-level simulation approach utilising social networking and
epidemiological models with a spatial component

by

Paul Davie

Thesis for the degree of Master of Philosophy

March 2015

ABSTRACT
Despite a range of epidemiological models existing, the majority of
these are cohort-level instead of individual-level models. Individual level
models allow for contact tracing, where one can see who each individual
interacts with. With the increasing popularity of social media amongst
students, most noticeably the rise of Facebook, we have chosen to
integrate an evolving social networking model with a conventional
Susceptible-Infectious-Recovered (SIR) epidemiological model in order to
simulate how infection is spread by contact with a growing netowkr of
friends within a population.
We considered the case of “Freshers’ Flu”, a form of seasonal influenza,
in a closed population simulation of new students at university. This is
a comparatively well-defined infection with known consistent values for
the rate of infection and recovery, and is primarily spread by airborne
transmission. Using the principles of discrete event simulation, and
collecting data on lectures, social events and population demographics
we created unique series of events per individual, combined with a
personality type defined by their individual average daily friendship
growth.
We ran several scenarios which examined the default case of an
infection spreading, the recommended university strategy of closing
campus during an epidemic and the effects of vaccinating specific
subsets of the population such as individuals on a particular degree
course or those living in specific halls of residences.
The model produced results which were consistent with a typical SIR
model of an influenza outbreak, although smaller and over a longer time
period. The social network and the formation of friends over time
within the model were shown to have an impact on incidence, the
number of new cases of infection per day.
Prior to lectures commencing, the greatest influence on infection were
the contacts made in halls of residences, with a background
contribution from communal and social events. Post lectures, there was
i
a consistent spike in incidence after the formation of friendships based
upon studying the same degree.
ii
iii
Contents

ABSTRACT ............................................................................................................................... i
List of figures ....................................................................................................................... viii
List of tables........................................................................................................................... x
DECLARATION OF AUTHORSHIP ......................................................................................... xii
Acknowledgements ............................................................................................................ xiv
1 Introduction ........................................................................................................................ 1
1.1 Aims of the model ....................................................................................................... 3
1.2 Purpose of the model .................................................................................................. 4
1.3 Potential benefits of individual-level models ............................................................. 5
2. Background ........................................................................................................................ 9
2.1 Epidemiological Modelling .......................................................................................... 9
2.2 Traditional compartmental models .......................................................................... 10
2.3 Types of Simulation ................................................................................................... 11
2.3.1 System dynamics ................................................................................................ 13
2.3.2 Dynamic Systems ................................................................................................ 13
2.3.3. Discrete Event Simulation ................................................................................. 14
2.3.4 Differences between Systems Dynamics and Discrete Event Simulation........ 14
2.3.5 Agent Based Models ........................................................................................... 16
2.3.6 Spatial Modelling ................................................................................................ 21
2.4 Social Network Analysis ............................................................................................ 24
2.5 Online Social Networks ............................................................................................. 29
2.6 Influenza and “freshers flu”. ..................................................................................... 32
3. Literature Review ............................................................................................................. 36
3.1 Network Modelling .................................................................................................... 36
3.2 Agent Based Modelling .............................................................................................. 44
3.3 Facebook reflecting the world .................................................................................. 62
3.4 Conclusions ................................................................................................................ 69
4. Modelling approach ........................................................................................................ 73
iv
4.1 Disease model ............................................................................................................ 73
4.2 Social networking model ........................................................................................... 79
4.3 Spatial model ............................................................................................................. 83
5. Challenges ....................................................................................................................... 90
5.1 Time handling ............................................................................................................ 90
5.2 Data requirements ..................................................................................................... 91
5.3 Potential solutions to the problems ......................................................................... 93
5.3.1 Data Collection .................................................................................................... 93
5.3.2 Time Handling ..................................................................................................... 95
6. Methodology .................................................................................................................... 99
6.1 The Programmatical Model ....................................................................................... 99
6.2 The Data ................................................................................................................... 106
6.2.1 Primary Data Sources – Social Network Model ............................................... 106
6.2.2 Secondary Data Sources ................................................................................... 109
6.3 The Disease Model .................................................................................................. 110
6.4 Location Data ........................................................................................................... 111
6.5 Data Collection ........................................................................................................ 115
6.5.1 Social Data Collection ....................................................................................... 118
6.6 Data Filtering ........................................................................................................... 127
6.7 Conducting the data collection .............................................................................. 130
6.8 Storing the Data ....................................................................................................... 138
6.9 Parameters for the model from the data ............................................................... 141
7. Results ............................................................................................................................ 147
7.1 Scenarios .................................................................................................................. 147
7.2 Replications & sensitivity ........................................................................................ 151
7.3 Input Data ................................................................................................................ 152
7.3.1 Event Data ......................................................................................................... 154
7.4 Model Validation ...................................................................................................... 165
7.4.1 Comparison with Compartmental Model ........................................................ 169
7.4.2 Varying the rate of infection ............................................................................ 173
7.4.3 Varying the rate of Friend Growth ................................................................... 178
v
7.4.4 Varying the population vaccination rate ......................................................... 186
7.5 Default Scenario ....................................................................................................... 188
7.6 Scenario 1: Closing campus .................................................................................... 194
7.7 Scenario 2: Targeting specific groups within the population .............................. 197
7.8 Scenario 3: Remove the “popular” people.............................................................. 206
8 Discussion ....................................................................................................................... 211
8.1 Friend Growth .......................................................................................................... 211
8.2 Impact of events ...................................................................................................... 213
8.3 Impact of friends ..................................................................................................... 214
9 Future Work & Development.......................................................................................... 217
9.1 Events & Timetables ................................................................................................ 217
9.2 Expanding the spatial component.......................................................................... 219
9.3 Expanding the infection .......................................................................................... 220
9.4 Personality types ...................................................................................................... 222
9.5 Improving the social network ................................................................................. 224
10 Conclusion .................................................................................................................... 226
11 Further Reflections ....................................................................................................... 237
11.1 Changes in Technology & Society ........................................................................ 237
11.2 Research Questions ............................................................................................... 238
11.3 Changing trends of social media ......................................................................... 240
11.4 Data collection from online social networks ....................................................... 241
11.4.1 Twitter, not Facebook .................................................................................... 243
11.5 Control strategies in academic environments ..................................................... 244
11.6 Research Questions Revisited ............................................................................... 245
11.7 Critical Appraisal ................................................................................................... 248
12. References ................................................................................................................... 250

vi
vii