---
title: "Network centralities in GeoHabnet"
author: "Plex, Sula (plexaaron@ufl.edu), Krishna Keshav(kkeshav@ufl.edu)"
date: "`r format(Sys.time(), '%d %B, %Y')`"
output: rmarkdown::html_vignette
vignette: >
  %\VignetteIndexEntry{Network centralities in GeoHabnet}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
bibliography: references.bib
---

This article expands on "Analyzing risk index using habitat connectivity" from [@geohabnet] to demonstrate one example on how parameters can be customized to calculate centrality and it's implication in the results.

Package installation -

```{r setup}
if (!require("geohabnet")) {
  utils::install.packages("geohabnet")
}
library(geohabnet)
```

## Betweenness

Getting the host density

```{r fetch}
avocado_mon <- geohabnet::cropharvest_rast("potato", "monfreda")
```

Running Sensitivity Analysis

```{r run}
avocado_result <- geohabnet::msean(avocado_mon, global = TRUE, link_threshold  = 0.000001,
                                   inv_pl = list(beta = c(0.5),
                                                 metrics = c("betweeness"),
                                                 weights = c(100),
                                                 cutoff = -1), res = 24,
                                   neg_exp = list(gamma = c(0.1),
                                                  metrics = c("betweeness"),
                                                  weights = c(100), cutoff = -1))
```

It is important to note that Betweenness centrality is a time intensive operation. Geohabnet uses a wrapper around `igraph::betweenness()`. Refer to [@igraph] for more details and further reading. The `cutoff` parameter can be used set the threshold which will include the link weights based on this threshold. Setting this parameter will also be reflected in closeness.

The configuration based run for the above will be as follows -

get_parameters() -\> modify values in parameters.yaml -\> set_parameters() -\> run sensitivity_analysis()

## References