FirstBeamerPresentationPdf1.Rmd

---
title: "Immunization Strategies in Complex Networks: Addressing the Problem of Missing Data"
author: "Sam Rosenblatt"
date: "9/12/2018"
output: 
  beamer_presentation: 
    colortheme: rose
    fonttheme: professionalfonts
    theme: Szeged
---



## Complete Data is Rare: \newline Regular Issues with Census Data

- Unwilling to be interviewed

\bigskip

- Contact details secret even among partners

\bigskip

- Population members inaccessible
    - Incarcerated
    - Homeless (no address)



## Complete Data is Rare: \newline Hidden Population Example - Colorado Springs

- At least 44% of nodes missing (out of those found eligible for the study)

\bigskip

- At least 88% of nodes with incomplete local network data

\bigskip

- These are generous lower bounds




## Complete Data is Rare: \newline Impact on Centrality Measures

\begin{picture}(320,250)
 \put(00,105){\includegraphics[height=4.5cm]{betweennessExample.jpg}}
 \put(00,90){\begin{minipage}[t]{0.9\linewidth}
 {Some centrality measures can change drastically with the addition/removal of few nodes}
 \end{minipage}}
 \end{picture}
 
## Stochastic Strategies
>- 1: Randomly sample from node list 
>- 2: Investigate local network around randomly sampled node
>- 3: Nodes are found to be central if they are frequently found to be locally important


## True Network vs Known Network

 \begin{picture}(320,250)
 \put(-25,100){\includegraphics[height=3.4cm]{trueNetwork.jpg}}
 \put(155,100){\includegraphics[height=3.4cm]{KnownNetwork.jpg}} 
 \put(00,50){\begin{minipage}[t]{1\linewidth}
 \end{minipage}}
 \end{picture}
 

## Relative Efficacy of Immunization Strategies 
	

 \begin{picture}(320,250)
 \put(0,40){\includegraphics[height=7cm]{RelativeEfficacyAll.jpg}}
 \put(50,180){\begin{minipage}[t]{0.4\linewidth}
 {\color{blue}Random \\ \color{green}Betweenness \\ \color{purple}Degree  \\ \color{red}AcquaintanceKK \\ \color{yellow}AcquaintanceKG \\ \color{orange}AcquaintanceGG}
 \end{minipage}}
 \end{picture}