Module # 12

 

Your assignment is to create your first visual Social Network analysis. You can select from any two platforms you feel more comfortable with. The first is called NodeXLLinks to an external site.. It is based on Microsoft Excel support platform. To download NodeXL package for Microsoft Excel click hereLinks to an external site. (Links to an external site.)Links to an external site.. Make sure you have Microsoft Excel in your computer. For full documentation on how to use NodeXL, click hereLinks to an external site. (Links to an external site.)Links to an external site.. (https://www.smrfoundation.org/nodexlLinks to an external site.)
The second option is using Rstudio. In my presentation, I used ggnet2: network visualization with ggplot2. The creator of this page is Ehsan Aghaei. (Links to an external site.)Links to an external site. You can find the full documentation on his code on his GitHub site: https://briatte.github.io/ggnet (Links to an external site.)Links to an external site.

Using the given code, this is the result after inputting into Rstudio. 

> net = rgraph(10, mode = "graph", tprob = 0.5)
> net = network(net, directed = FALSE)
> 
> network.vertex.names(net) = letters[1:10]
> 
> ggnet2(net)

After plotting, it seems that there really aren't any relationships demonstrated. However, I believe more could be added to this by using more data. 



Using the dataset titled movies.csv, I wanted to focus on a connection between directors and movies. 



> edges <- movies %>%
+   select(title, director) %>%
+   filter(!is.na(director), director != "") %>%
+   slice(1:50) 
> 
> net <- network(edges, directed = FALSE)
> 
> ggnet2(net,
+        label = TRUE,
+        label.size = 2.5,
+        color = "pink",
+        size = 5)



Using the code above, I focused on two columns, director and title. The title is the name of the film. The filter(!is.na(director),director ! = "") removes any rows where the director is missing or blank. Since the dataset is quite large, I wanted to limit it to 50 rows. Similar to the graph above, I am able to plot it using the network function. Below is the result. 




From here we are able to the directors that have led numerous films. For example, we can see that Bryan Singer had made numerous superhero films such as X Men: Days of Future Past and Superman Returns, as well as Jack the Giant Slayer. This visualization would be even more interesting if more rows were implemented however it creates a rather difficult time reading the plot. More relationships can be seen with various movies. Unfortunately the dataset only lists one director for each movie and no co-director or producers. I believe if this data was available, we would be able to see more overlap with these filmmakers. Using the given code, this is the result after inputting into Rstudio. Just for reference, I made another plot with 200 rows utilized but you would need to zoom in to see the relationships clearly. I believe this plot does a better job in portraying the power of social network analysis. Overall, SNA (social network analysis) can be quite tedious when sorting through large datasets, however I believe it can effectively demonstrate patterns of interaction.