mp-bib-coupling

[brieaspasia]

Create a map of bibliographic coupling and history of mp research

• [ ] #9
• [x] transform biblioNetwork analysis into df for Gephi
• [x] solve error message in historiography analysis (deleted this code in order to focus on code that was working)

[brieaspasia]

Plotting the igraph object (gD created at line 61) results in a meaningless graphic, I'm not sure how to modify it to make it meaningful.

[brieaspasia]

I was able to create the Gephi graph at line 157, but I didn't realise that it was completely different software and I don't know how to modify or read what it has created. I think for a start, I just need to look at the top 50 nodes instead of all of them.

[mlagisz]

@brieaspasia:

I don't know what this is meant to do??? See line 47 at #https://www.r-bloggers.com/network-visualization-part-2-gephi/

F1 <- function(x) {data.frame(V4 = dsAll[which(V(gD)$name == as.character(x$V1)), which(V(gD)$name == as.character(x$V2))])}

6 dataSet.ext <- plyr::ddply(out, .variables=c("V1", "V2", "V3"), function(x) data.frame(F1(x)))

It looks like this matches dataframe to graph values by references and creates a new column. You would need to adjust this function to make it work, because you probably have different column names.

igraphs allows you great level of customisation but it comes at a cost of complex code and need to understand of what id does exactly. If you are short of time just stick functions provided by blibliometrix, but explore more the parameters you can set there, e.g. see this:

NetMatrix <- biblioNetwork(bib, analysis = "coupling", network = "references", sep = ";") net = networkPlot(NetMatrix, weighted = NULL, n = 50, label.n = 10, Title = "Authors' bibliographic coupling", type = "fruchterman", cluster= "walktrap", size.cex = TRUE, remove.multiple = FALSE, remove.isolates = FALSE, labelsize = 1, edgesize=2)

You can display labels only for the top n nodes (label.n.) to make graphs readable.

To me it seems like most of the time you simply get two clusters of papers - its like tow communities of researchers in 2015-2017?

](url)

Also, you do not need to make the plots perfect in R - you can e.g. add annotations in other programs, even PowerPoint.

[brieaspasia]

I was able to make a chord diagram of bibliographic coupling, however each author is linked in exactly the same way to each other author... its such a pretty graph, too bad it doesn't tell us much!