example use of Scopus API for a citation network

scopus_bib.R

library(rscopus)
library(tidyverse)
library(igraph)
library(GGally)

faculty <- read_csv("faculty.csv") 

faculty$au_id <- NA
for (i in 1:nrow(faculty)) {
  print(i)
  Sys.sleep(.5) # to solve timeout problem
  au_id <- get_author_info(last_name = faculty$last[i], 
                           first_name = faculty$first[i],
                           affil_id = faculty$affiliation[i])$au_id[1]
  faculty$au_id[i] = ifelse(!is.null(au_id), au_id, NA)
}


# Get all articles by each faculty member. 
articles <- faculty |>
  mutate(idx = 1:n()) %>%
  split(.$idx) |>
  map_df(function(f) {    
    print(f$idx)
    Sys.sleep(.5)
    
    author_df(au_id = f$au_id,
              affil_id = f$affiliation, 
              verbose=FALSE) |>
      mutate(faculty = f$last)
  })

articles <- as_tibble(articles)

multi_author <- articles |> 
  group_by(`dc:identifier`) |>
  summarise(n_authors = length(unique(faculty))) |>
  filter(n_authors > 1)

edgelist <- left_join(multi_author, 
                      select(articles, `dc:identifier`, faculty)) |>
  rename(id = `dc:identifier`) %>%
  group_by(id) |>
  mutate(i = 1:n()) |>
  pivot_wider(names_from = i, values_from = "faculty") 

edgelist_3 <- filter(edgelist, !is.na(`3`)) %>%
  split(.$id) %>%
  map_df(function(x) {
    tibble(id = x$id, 
           n_authors = 3, 
           `1` = c(x$`1`[1], x$`1`[1]), 
           `2` = c(x$`2`[1], x$`3`[1]))
  })

edgelist_2 <- filter(edgelist, is.na(`3`)) |>
  arrange(id)

edgelist_full <- edgelist_2 |>
  bind_rows(edgelist_3) |> 
  ungroup() |>
  select(`1`,`2`) |>
  group_by(`1`,`2`) |>
  count()

mat <- as.matrix(edgelist_full[,1:2], ncol = 2)
g <- graph_from_edgelist(mat, directed=FALSE)

area <- as.factor(unlist(map(names(V(g)), 
                             function(x) {filter(faculty, last == x)$area})))

GGally::ggnet2(g,
               label = TRUE,
               mode = "kamadakawai",
               node.color = area,
               edge.size = edgelist_full$n/3)