3-visualization.qmd

---
title: "Visualization"
author: 
  - George G. Vega Yon
  - Aníbal Olivera M.
date: "2025-06-24"
date-modified: "2025-06-24"
---

```{r setup, echo=FALSE, message=FALSE, warning=FALSE}
library(netdiffuseR)
knitr::opts_chunk$set(comment = "#")
```


# Visualization methods

__netdiffuseR__ offers several methods to visualize network diffusion dynamics. Here we will focus on three of them: `plot_adopters()`, `plot_threshold()`, and `plot_diffnet()`.

You can see the [Reference Manual](https://cran.r-project.org/web/packages/netdiffuseR/netdiffuseR.pdf) on CRAN, or the Appendix from this lecture, to see the full list of methods available in the package.

### Creating a diffusion network with survey_to_diffnet()

We can select a specific village and visualize its diffusion dynamics. As we did in the previous section, we will use the `kfamily` dataset, which contains data on Korean Family Planning, and we will focus on village #21:

```{r}
#| label: kfamily-village-21
#| warning: false
#| message: false
data(kfamily)

# Subset the kfamily dataset for village 21
kfamily_21 <- subset(kfamily, village == 21)

# Convert the subsetted data to a diffnet object
kfamily_diffnet_21 <- survey_to_diffnet(
  dat      = kfamily_21,
  idvar    = "id",
  netvars  = c(
    # Neighbors talk to about FP
    "net11", "net12", "net13", "net14", "net15", 
    # Closest neighbor most frequently met
    "net21", "net22", "net23", "net24", "net25", 
    # Advice on FP sought from 
    "net31", "net32", "net33", "net34", "net35"), 
  toavar   = "toa",
  groupvar = "village"
)
```

We can visualize adopters and cumulative adopters:

```{r}
#| label: kfamily-village-21-plot_adopters
#| warning: false
#| message: false

plot_adopters(kfamily_diffnet_21)
```

or draw a graph where the x-axis coordinates are given by time of adoption, and the y-axis by the threshold level:

```{r}
#| label: kfamily-village-21-plot_threshold
#| warning: false
#| message: false

plot_threshold(kfamily_diffnet_21)
```

or create a colored network plot showing the diffusion status of the graph through time (one network plot for each time period):

```{r}
#| label: kfamily-village-21-plot_diffnet
#| warning: false
#| message: false

plot_diffnet(kfamily_diffnet_21)
```

# Problems

1.  Using the diffnet object in [`intro.rda`](files/intro.rda), use the function `plot_threshold` specifying shapes and colors according to the variables ItrustMyFriends and Age. Do you see any pattern? (<a href="files/intro-solutions.r" target="_blank">solution script</a> and <a href="files/intro-solutions.png" target="_blank">solution plot</a>)

```{r datasim, echo=FALSE, eval=TRUE}
set.seed(1252)
dat <- data.frame(
  ItrustMyFriends = sample(c(0,1), 200, TRUE),
  Age = 10 + rpois(200, 4)
  )
net <- rgraph_er(200, p = .05)
# net <- diag_expand(list(net, net))
# net[cbind(1:20, 101:120)] <- 1 

# Generating the process
diffnet <- rdiffnet(
  threshold.dist = 4 - dat$ItrustMyFriends*3,
  seed.graph = net,
  t=6,
  seed.nodes = c(9:25),
  exposure.args = list(normalized=FALSE),
  rewire = FALSE)

diffnet[["ItrustMyFriends"]] <- dat$ItrustMyFriends
diffnet[["Age"]] <- dat$Age

save(diffnet, file = "files/intro.rda")
```

# Appendix

### Creating a diffusion network with `rdiffnet()`

Another example using the `rdiffnet` function to generate random diffusion networks. This example creates a small-world network with 400 nodes and 6 time steps, where the connections are rewired based on a threshold distance of 1/4. The `seed.graph` is set to "small-world" and the `seed.nodes` to "central", which means that the initial nodes are chosen from the center of the network.


```{r}
#| label: rdiffnet-example
set.seed(12315)
diffnet_1 <- rdiffnet(
  n = 400, t = 6,
  rgraph.args    = list(k=6, p=.3),
  seed.graph     = "small-world",
  seed.nodes     = "central",
  rewire         = FALSE,
  threshold.dist = 1/4
  )
diffnet_1
```

Diffusion networks can visualized using many methods included in the package. Here are some of them:

```{r}
#| label: plot-methods

plot(diffnet_1)
plot_diffnet(diffnet_1)
plot_diffnet2(diffnet_1)
plot_adopters(diffnet_1)
plot_threshold(diffnet_1)
plot_infectsuscep(diffnet_1, K=2)
plot_hazard(diffnet_1)
```