Scientometric Analysis
Sebastian Robledo
1/10/2023
Creating the environment
library(tidyverse)
library(tidygraph)
library(igraph)
library(bibliometrix)
library(tosr)
library(here)
library(lubridate)
# library(sjrdata)
library(openxlsx)
library(zoo)
library(RSQLite)
library(journalabbr)
library(ggraph)
library(openxlsx)
library(XML)
library(plyr)
library(readxl)
source("verbs.R")
windowsFonts("Times" = windowsFont("Times"))
windowsFonts("Times New Roman" = windowsFont("Times New Roman"))
giant.component <- function(graph) {
cl <- igraph::clusters(graph)
igraph::induced.subgraph(graph,
which(cl$membership == which.max(cl$csize)))
}Data getting
wos_scopus <-
read_csv("https://docs.google.com/spreadsheets/d/1U3Pv7cAnntkcuRfuyv0ixFAtz_IBT3_sDzJYsJpm3K4/export?format=csv&gid=1759343267")
wos <-
read_csv("https://docs.google.com/spreadsheets/d/1U3Pv7cAnntkcuRfuyv0ixFAtz_IBT3_sDzJYsJpm3K4/export?format=csv&gid=185803309") # create dataframe from wos file
scopus <-
read_csv("https://docs.google.com/spreadsheets/d/1U3Pv7cAnntkcuRfuyv0ixFAtz_IBT3_sDzJYsJpm3K4/export?format=csv&gid=9089935")
reference_df <-
read_csv("https://docs.google.com/spreadsheets/d/1U3Pv7cAnntkcuRfuyv0ixFAtz_IBT3_sDzJYsJpm3K4/export?format=csv&gid=1730837925")
journal_df <-
read_csv("https://docs.google.com/spreadsheets/d/1U3Pv7cAnntkcuRfuyv0ixFAtz_IBT3_sDzJYsJpm3K4/export?format=csv&gid=654212654")
author_df <-
read_csv("https://docs.google.com/spreadsheets/d/1U3Pv7cAnntkcuRfuyv0ixFAtz_IBT3_sDzJYsJpm3K4/export?format=csv&gid=1667291175")
TC_all <-
read_csv("https://docs.google.com/spreadsheets/d/1U3Pv7cAnntkcuRfuyv0ixFAtz_IBT3_sDzJYsJpm3K4/export?format=csv&gid=1374285506")
figure_1_data <-
read_csv("https://docs.google.com/spreadsheets/d/1U3Pv7cAnntkcuRfuyv0ixFAtz_IBT3_sDzJYsJpm3K4/export?format=csv&gid=780120419")
table_2_country <- #table_2_country
read_csv("https://docs.google.com/spreadsheets/d/1U3Pv7cAnntkcuRfuyv0ixFAtz_IBT3_sDzJYsJpm3K4/export?format=csv&gid=1280545488")
figure_2_country_wos_scopus <- #figure_2_country_wos_scopus
read_csv("https://docs.google.com/spreadsheets/d/1U3Pv7cAnntkcuRfuyv0ixFAtz_IBT3_sDzJYsJpm3K4/export?format=csv&gid=633478795")
figure_2_country_wos_scopus_1 <- #figure_2_country_wos_scopus_1
read_csv("https://docs.google.com/spreadsheets/d/1U3Pv7cAnntkcuRfuyv0ixFAtz_IBT3_sDzJYsJpm3K4/export?format=csv&gid=686814371") |>
tidygraph::as_tbl_graph(directed = FALSE) |>
activate(nodes) |>
dplyr::mutate(community = tidygraph::group_louvain(),
degree = tidygraph::centrality_degree(),
community = as.factor(community))
#
table_3_journal <-
read_csv("https://docs.google.com/spreadsheets/d/1U3Pv7cAnntkcuRfuyv0ixFAtz_IBT3_sDzJYsJpm3K4/export?format=csv&gid=2006411765")
#
table_4_authors <- #table_4_authors
read_csv("https://docs.google.com/spreadsheets/d/1U3Pv7cAnntkcuRfuyv0ixFAtz_IBT3_sDzJYsJpm3K4/export?format=csv&gid=1036812008")
#
AU_CO_links <-
read_csv("https://docs.google.com/spreadsheets/d/1U3Pv7cAnntkcuRfuyv0ixFAtz_IBT3_sDzJYsJpm3K4/export?format=csv&gid=2022067094")
tos <-
read_csv("https://docs.google.com/spreadsheets/d/1U3Pv7cAnntkcuRfuyv0ixFAtz_IBT3_sDzJYsJpm3K4/export?format=csv&gid=1816339518")
edges_tos <-
read_csv("https://docs.google.com/spreadsheets/d/1U3Pv7cAnntkcuRfuyv0ixFAtz_IBT3_sDzJYsJpm3K4/export?format=csv&gid=59260896")
nodes_tos <-
read_csv("https://docs.google.com/spreadsheets/d/1U3Pv7cAnntkcuRfuyv0ixFAtz_IBT3_sDzJYsJpm3K4/export?format=csv&gid=179092650")
SO_edges <-
read_csv("https://docs.google.com/spreadsheets/d/1U3Pv7cAnntkcuRfuyv0ixFAtz_IBT3_sDzJYsJpm3K4/export?format=csv&gid=769886924")
SO_nodes <-
read_csv("https://docs.google.com/spreadsheets/d/1U3Pv7cAnntkcuRfuyv0ixFAtz_IBT3_sDzJYsJpm3K4/export?format=csv&gid=743563253")
AU_ego_edges <-
read_csv("https://docs.google.com/spreadsheets/d/1U3Pv7cAnntkcuRfuyv0ixFAtz_IBT3_sDzJYsJpm3K4/export?format=csv&gid=1916489693")
AU_ego_nodes <-
read_csv("https://docs.google.com/spreadsheets/d/1U3Pv7cAnntkcuRfuyv0ixFAtz_IBT3_sDzJYsJpm3K4/export?format=csv&gid=2059395475")Resutls
Scientometric Analysis
3.1 Scientific Production
Figure 1a - Scopus + WoS
Combine charts using Python Matplotlib & Reticulate
library(reticulate)
# create a new environment
# conda_create("r-reticulate")
# install Matplotlib
# conda_install("r-reticulate", "matplotlib")
# import Matplotlib (it will be automatically discovered in "r-reticulate")
plt <- import("matplotlib")
np <- import("numpy")# From Double get integers
# TC y
TC_all$TC_sum_all <- as.integer(TC_all$TC_sum_all)import matplotlib.pyplot as plt
import numpy as np
from matplotlib.ticker import FuncFormatter
# ax=axes
fig, ax = plt.subplots()
# First plot Total Publications - time series
ax.plot(tpx, tpy, color='r',marker='o', label='Total Publications')
ax.set_xlabel('Year')
ax.set_ylabel('Total Publications', color='r')
# Customization for bar charts
barw = 0.5
ax.bar(sx, sy, color='g', label = 'Scopus', alpha = 0.5, width=barw)## <BarContainer object of 22 artists>ax.bar(wx1, wy, color='orange', label = 'WoS', alpha=0.8, width=barw)
# Y2 - Total citations## <BarContainer object of 22 artists>twin_axes = ax.twinx()
twin_axes.plot(tcx, tcy, color = 'purple',marker='o', label='Total Citations')
twin_axes.set_ylabel('Total Citations', color='purple')
# Customize
plt.title('Total Scientific Production vs. Total Citations')
# y2 Total Citation label location
plt.legend(loc='center left')
# True or False to get the grid at the background
ax.grid(False)
# y1 label location
ax.legend(loc='upper left')
# Y2 limit depends of tcy scale in this case 1400 improves label location
plt.ylim(0, 1400) ######### <-----Important--------- """"Change Y2 Coordinate"""""
# plt.annotate() customize numbers for each position## (0.0, 1400.0)for i, label in enumerate(tcy):
plt.annotate(label, (tcx[i], tcy[i] + 0.5), color='purple', size=8)
for i, label in enumerate(tpy):
ax.annotate(label, (tpx[i], tpy[i] + 0.8), color='red', size=8)
for i, label in enumerate(wy):
ax.annotate(label, (wx1[i], wy[i] + 0.1), color='yellow', size=8)
for i, label in enumerate(sy):
ax.annotate(label, (sx[i], sy[i] + 0.2),color='green', size=8)
# Rotate x ticks
plt.xticks(tpx)## ([<matplotlib.axis.XTick object at 0x000001EC11D14580>, <matplotlib.axis.XTick object at 0x000001EC11D14400>, <matplotlib.axis.XTick object at 0x000001EC0E5DB3A0>, <matplotlib.axis.XTick object at 0x000001EC11D8DA00>, <matplotlib.axis.XTick object at 0x000001EC11D92070>, <matplotlib.axis.XTick object at 0x000001EC11D927C0>, <matplotlib.axis.XTick object at 0x000001EC11D9D070>, <matplotlib.axis.XTick object at 0x000001EC11D9D6A0>, <matplotlib.axis.XTick object at 0x000001EC11D9DDF0>, <matplotlib.axis.XTick object at 0x000001EC11D9DE50>, <matplotlib.axis.XTick object at 0x000001EC11D92CA0>, <matplotlib.axis.XTick object at 0x000001EC11D8DAC0>, <matplotlib.axis.XTick object at 0x000001EC11DA4A90>, <matplotlib.axis.XTick object at 0x000001EC11DA9220>, <matplotlib.axis.XTick object at 0x000001EC11DA9970>, <matplotlib.axis.XTick object at 0x000001EC11DB0100>, <matplotlib.axis.XTick object at 0x000001EC11DA9A30>, <matplotlib.axis.XTick object at 0x000001EC11D8DA60>, <matplotlib.axis.XTick object at 0x000001EC11DB01C0>, <matplotlib.axis.XTick object at 0x000001EC11DB0D60>, <matplotlib.axis.XTick object at 0x000001EC11DB74F0>, <matplotlib.axis.XTick object at 0x000001EC11DB7C40>], [Text(0, 0, ''), Text(0, 0, ''), Text(0, 0, ''), Text(0, 0, ''), Text(0, 0, ''), Text(0, 0, ''), Text(0, 0, ''), Text(0, 0, ''), Text(0, 0, ''), Text(0, 0, ''), Text(0, 0, ''), Text(0, 0, ''), Text(0, 0, ''), Text(0, 0, ''), Text(0, 0, ''), Text(0, 0, ''), Text(0, 0, ''), Text(0, 0, ''), Text(0, 0, ''), Text(0, 0, ''), Text(0, 0, ''), Text(0, 0, '')])fig.autofmt_xdate(rotation = 70)
# The Y1 ticks depends from tpy scale limits
yticks = [0,20,40,60,80,100] ########## <-----Important---- Choose scale .. just specify which numbers you want
ax.set_yticks(yticks)
# Export Figure as SVG
#plt.savefig("ScientificProd_4charts.svg")
plt.show()
figure_1a <-
figure_1_data |>
pivot_longer(!PY, names_to = "ref_type", values_to = "n") |>
filter(ref_type != "total") |>
ggplot(aes(x = factor(PY),
y = n,
fill = ref_type)) +
geom_bar(stat = "identity",
position = "dodge") +
geom_text(aes(label = n),
vjust = -0.3,
position = position_dodge(0.9),
size = 3,
family = "Times") +
scale_fill_manual(values = c("springgreen3",
"orange3")) +
theme(text = element_text(family = "Times",
face = "bold",
size =12),
panel.background = element_rect(fill = "white"),
legend.position = "bottom",
legend.title = element_text(size = 0),
axis.text.x = element_text(face = "bold",
angle = 45,
vjust = 0.5),
axis.line = element_line(color = "black",
size = 0.2)) +
labs(y = "Number of publications",
x = "Year")
figure_1a
Figure 1b - Total production
figure_1b <-
figure_1_data |>
ggplot(aes(x = PY, y = total)) +
geom_line(stat = "identity", color = "red") +
geom_point(stat = "identity", color = "red") +
scale_x_continuous(breaks = seq(2002, year(today()) - 1, by = 1)) +
geom_text(aes(label = total),
vjust = -0.3,
position = position_dodge(0.9),
size = 3,
family = "Times New Roman",
color = "red") +
scale_fill_manual(values = c("springgreen3",
"orange3")) +
theme(text = element_text(family = "Times New Roman",
face = "bold",
size =12),
panel.background = element_rect(fill = "white"),
legend.position = "bottom",
legend.title = element_text(size = 0),
axis.text.x = element_text(face = "bold",
angle = 45,
vjust = 0.5),
axis.line = element_line(color = "black",
size = 0.2)) +
labs(y = "Number of total publications",
x = "Year")
figure_1b
Figure 1c - Total Citations
figure_1c <-
TC_all |>
ggplot(aes(x = PY , y = TC_sum_all)) +
geom_line(stat = "identity", color = "purple") +
geom_point(color = "purple") +
scale_x_continuous(breaks = seq(2002, year(today()) - 1 , by = 1)) +
geom_text(aes(label = TC_sum_all),
vjust = -0.3,
position = position_dodge(0.9),
size = 3,
family = "Times New Roman",
color = "purple") +
scale_fill_manual(values = c("springgreen3",
"orange3")) +
theme(text = element_text(family = "Times New Roman",
face = "bold",
size =12),
panel.background = element_rect(fill = "white"),
legend.position = "bottom",
legend.title = element_text(size = 0),
axis.text.x = element_text(face = "bold",
angle = 45,
vjust = 0.5),
axis.line = element_line(color = "black",
size = 0.2)) +
labs(y = "Number of citations",
x = "Year")
figure_1c
3.2 Country analysis
Table 2 - Country production
table_2_country |>
DT::datatable(class = "cell-border stripe",
rownames = F,
filter = "top",
editable = FALSE,
extensions = "Buttons",
options = list(dom = "Bfrtip",
buttons = c("copy",
"csv",
"excel",
"pdf",
"print")))Figure 2a - Country Collaboration
# figure_2 <-
# figure_2_country_wos_scopus_1 |>
# ggraph(layout = "graphopt") +
# geom_edge_link(aes(width = weight),
# colour = "lightgray") +
# scale_edge_width(name = "Link strength") +
# geom_node_point(aes(color = community,
# size = degree)) +
# geom_node_text(aes(label = name), repel = TRUE) +
# scale_size(name = "Degree") +
# # scale_color_binned(name = "Communities") +
# theme_graph()
#
# figure_2
figure_2a <-
figure_2_country_wos_scopus_1 |>
activate(edges) |>
dplyr::rename(weight = n) |>
ggraph(layout = "graphopt") +
geom_edge_link(aes(width = weight),
colour = "lightgray") +
scale_edge_width(name = "Link strength") +
geom_node_point(aes(color = community,
size = degree)) +
geom_node_text(aes(label = name), repel = TRUE) +
scale_size(name = "Degree") +
# scale_color_binned(name = "Communities") +
theme_graph()
figure_2a
# country_collab_graphml_nodes <-
# figure_2_country_wos_scopus_1 |>
# activate(nodes) |>
# as_tibble() |>
# dplyr::rename(author = name) |>
# rownames_to_column("name")
#
# country_collab_graphml_edges <-
# figure_2_country_wos_scopus_1 |>
# activate(edges) |>
# as_tibble()
#
# AU_CO_weighted_TM <-
# graph_from_data_frame(d = country_collab_graphml_edges,
# directed = FALSE,
# vertices = country_collab_graphml_nodes)
#
# write_graph(AU_CO_weighted_TM, "AU_CO_weighted_TM.graphml", "graphml") # Export author co-citation graphFigure 2b Clusters
figure_2b <-
figure_2_country_wos_scopus_1 |>
activate(nodes) |>
data.frame() |>
group_by(community) |>
dplyr::count(community, sort = TRUE) |>
slice(1:10) |>
ggplot(aes(x = reorder(community, n), y = n)) +
geom_point(stat = "identity") +
geom_line(group = 1) +
# geom_text(label = as.numeric(community),
# nudge_x = 0.5,
# nudge_y = 0.5,
# check_overlap = T) +
labs(title = "Communities by size",
x = "communities",
y = "Countries") +
theme(text = element_text(color = "black",
face = "bold",
family = "Times New Roman"),
plot.title = element_text(size = 25),
panel.background = element_rect(fill = "white"),
axis.text.y = element_text(size = 15,
colour = "black"),
axis.text.x = element_text(size = 15,
colour = "black"),
axis.title.x = element_text(size = 20),
axis.title.y = element_text(size = 20)
)
figure_2b
Figure 2c Longitudinal
# Create a dataframe with links
figure_2c_edges <-
figure_2_country_wos_scopus |>
dplyr::filter(from != to) |>
tidygraph::as_tbl_graph() |>
activate(edges) |>
as_tibble() |>
dplyr::select(year = PY) |>
dplyr::count(year) |>
dplyr::filter(year >= 2002,
year <= 2022) |>
dplyr::mutate(percentage = n/max(n)) |>
dplyr::select(year, percentage)
# Create a data frame with author and year
figure_2c_nodes <- # 21 row
figure_2_country_wos_scopus |>
dplyr::filter(from != to) |>
tidygraph::as_tbl_graph() |>
activate(edges) |>
as_tibble() |>
dplyr::select(CO = from,
year = PY) |>
bind_rows(figure_2_country_wos_scopus |>
tidygraph::as_tbl_graph() |>
tidygraph::activate(edges) |>
tidygraph::as_tibble() |>
dplyr::select(CO = to,
year = PY)) |>
unique() |>
dplyr::group_by(CO) |>
dplyr::slice(which.min(year)) |>
dplyr::ungroup() |>
dplyr::select(year) |>
dplyr::group_by(year) |>
dplyr::count(year) |>
dplyr::filter(year >= 2002,
year <= 2022) |>
dplyr::ungroup() |>
dplyr::mutate(percentage = n / max(n)) |>
select(year, percentage)figure_2c <-
figure_2c_nodes |>
mutate(type = "nodes") |>
bind_rows(figure_2c_edges |>
mutate(type = "links")) |>
ggplot(aes(x = year,
y = percentage,
color = type)) +
geom_point() +
geom_line() +
theme(legend.position = "right",
text = element_text(color = "black",
face = "bold",
family = "Times"),
plot.title = element_text(size = 25),
panel.background = element_rect(fill = "white"),
axis.text.y = element_text(size = 15,
colour = "black"),
axis.text.x = element_text(size = 15,
colour = "black",
angle = 45, vjust = 0.5
),
axis.title.x = element_text(size = 20),
axis.title.y = element_text(size = 20),
legend.text = element_text(size = "15"),
legend.title = element_blank()) +
labs(title = "Nodes and links through time",
y = "Percentage") +
scale_y_continuous(labels = scales::percent) +
scale_x_continuous(breaks = seq(2002, 2022, by = 1))
figure_2c
# library(patchwork)
# figure_2c / figure_2b | figure_2a# library(gridExtra)
# grid.arrange(figure_2c,
# figure_2b,
# figure_2a,
# ncol=3,
# widths=c(3,2, 2),
# heights=c(1,1, 1),
# layout_matrix = rbind(c(1,3, 3), c(2,3, 3)))3.3 Journal Analysis
Table 3 Most productive journals
table_3_journal |>
arrange(desc(total)) |>
DT::datatable(class = "cell-border stripe",
rownames = F,
filter = "top",
editable = FALSE,
extensions = "Buttons",
options = list(dom = "Bfrtip",
buttons = c("copy",
"csv",
"excel",
"pdf",
"print")))Figure 3 Journal Citation Network
Creating the graph object
journal_citation_graph_weighted_tbl_small <-
journal_df |>
dplyr::select(JI_main, JI_ref) |>
dplyr::group_by(JI_main, JI_ref) |>
dplyr::count() |>
dplyr::rename(weight = n) |>
as_tbl_graph(directed = FALSE) |>
# convert(to_simple) |>
activate(nodes) |>
dplyr::mutate(components = tidygraph::group_components(type = "weak")) |>
dplyr::filter(components == 1) |>
activate(nodes) |>
dplyr::mutate(degree = centrality_degree(),
community = tidygraph::group_louvain()) |>
dplyr::select(-components) |>
dplyr::filter(degree >= 1)
# activate(edges) |>
# dplyr::filter(weight != 1)
communities <-
journal_citation_graph_weighted_tbl_small |>
activate(nodes) |>
data.frame() |>
dplyr::count(community, sort = TRUE) |>
dplyr::slice(1:10) |>
dplyr::select(community) |>
dplyr::pull()
# Filtering biggest communities
journal_citation_graph_weighted_tbl_small_fig <-
journal_citation_graph_weighted_tbl_small |>
activate(nodes) |>
dplyr::filter(community %in% communities)Selecting nodes to show
jc_com_1 <-
journal_citation_graph_weighted_tbl_small_fig |>
activate(nodes) |>
dplyr::filter(community == communities[1]) |>
dplyr::mutate(degree = centrality_degree()) |>
dplyr::arrange(desc(degree)) |>
dplyr::slice(1:10) |>
data.frame() |>
dplyr::select(name)
jc_com_2 <-
journal_citation_graph_weighted_tbl_small_fig |>
activate(nodes) |>
dplyr::filter(community == communities[2]) |>
dplyr::mutate(degree = centrality_degree()) |>
dplyr::arrange(desc(degree)) |>
dplyr::slice(1:10) |>
data.frame() |>
dplyr::select(name)
jc_com_3 <-
journal_citation_graph_weighted_tbl_small_fig |>
activate(nodes) |>
dplyr::filter(community == communities[3]) |>
dplyr::mutate(degree = centrality_degree()) |>
dplyr::arrange(desc(degree)) |>
dplyr::slice(1:10) |>
data.frame() |>
dplyr::select(name)
jc_com_4 <-
journal_citation_graph_weighted_tbl_small_fig |>
activate(nodes) |>
dplyr::filter(community == communities[4]) |>
dplyr::mutate(degree = centrality_degree()) |>
dplyr::arrange(desc(degree)) |>
dplyr::slice(1:10) |>
data.frame() |>
dplyr::select(name)
jc_com_5 <-
journal_citation_graph_weighted_tbl_small_fig |>
activate(nodes) |>
dplyr::filter(community == communities[5]) |>
dplyr::mutate(degree = centrality_degree()) |>
dplyr::arrange(desc(degree)) |>
dplyr::slice(1:10) |>
data.frame() |>
dplyr::select(name)
jc_com_6 <-
journal_citation_graph_weighted_tbl_small_fig |>
activate(nodes) |>
dplyr::filter(community == communities[6]) |>
dplyr::mutate(degree = centrality_degree()) |>
dplyr::arrange(desc(degree)) |>
dplyr::slice(1:10) |>
data.frame() |>
dplyr::select(name)
jc_com_7<-
journal_citation_graph_weighted_tbl_small_fig |>
activate(nodes) |>
dplyr::filter(community == communities[7]) |>
dplyr::mutate(degree = centrality_degree()) |>
dplyr::arrange(desc(degree)) |>
dplyr::slice(1:10) |>
data.frame() |>
dplyr::select(name)
jc_com_8 <-
journal_citation_graph_weighted_tbl_small_fig |>
activate(nodes) |>
dplyr::filter(community == communities[8]) |>
dplyr::mutate(degree = centrality_degree()) |>
dplyr::arrange(desc(degree)) |>
dplyr::slice(1:10) |>
data.frame() |>
dplyr::select(name)
jc_com_9 <-
journal_citation_graph_weighted_tbl_small_fig |>
activate(nodes) |>
dplyr::filter(community == communities[9]) |>
dplyr::mutate(degree = centrality_degree()) |>
dplyr::arrange(desc(degree)) |>
dplyr::slice(1:10) |>
data.frame() |>
dplyr::select(name)
jc_com_10 <-
journal_citation_graph_weighted_tbl_small_fig |>
activate(nodes) |>
dplyr::filter(community == communities[10]) |>
dplyr::mutate(degree = centrality_degree()) |>
dplyr::arrange(desc(degree)) |>
dplyr::slice(1:10) |>
data.frame() |>
dplyr::select(name)
jc_com <-
jc_com_1 |>
bind_rows(jc_com_2,
jc_com_3,
# jc_com_4,
# jc_com_5,
# jc_com_6,
# jc_com_7,
# jc_com_8,
# jc_com_9,
# jc_com_10
)figure_3a <-
journal_citation_graph_weighted_tbl_small_fig |>
activate(nodes) |>
dplyr::filter(name %in% jc_com$name) |>
dplyr::mutate(degree = centrality_degree(),
community = factor(community)) |>
dplyr::filter(degree != 0) |>
ggraph(layout = "graphopt") +
geom_edge_link(aes(width = weight),
colour = "lightgray") +
scale_edge_width(name = "Link strength") +
geom_node_point(aes(color = community,
size = degree)) +
geom_node_text(aes(label = name), repel = TRUE) +
scale_size(name = "Degree") +
# scale_color_binned(name = "Communities") +
theme_graph()
figure_3a
Figure 3b clusters
figure_3b <-
journal_citation_graph_weighted_tbl_small |>
activate(nodes) |>
data.frame() |>
dplyr::select(community) |>
dplyr::count(community, sort = TRUE) |>
dplyr::slice(1:10) |>
ggplot(aes(x = reorder(community, n), y = n)) +
geom_point(stat = "identity") +
geom_line(group = 1) +
# geom_text(label = as.numeric(community),
# nudge_x = 0.5,
# nudge_y = 0.5,
# check_overlap = T) +
labs(title = "Communities by size",
x = "communities",
y = "Journals") +
theme(text = element_text(color = "black",
face = "bold",
family = "Times"),
plot.title = element_text(size = 25),
panel.background = element_rect(fill = "white"),
axis.text.y = element_text(size = 15,
colour = "black"),
axis.text.x = element_text(size = 15,
colour = "black"),
axis.title.x = element_text(size = 20),
axis.title.y = element_text(size = 20)
)
figure_3b
Figure 3c Longitudinal
# Create a dataframe with links
figure_3c_edges <-
journal_df |>
select(from = JI_main, to = JI_ref, PY = PY_ref) %>%
dplyr::filter(from != to) |>
tidygraph::as_tbl_graph() |>
activate(edges) |>
as_tibble() |>
dplyr::select(year = PY) |>
dplyr::count(year) |>
dplyr::filter(year >= 2002,
year <= 2022) |>
dplyr::mutate(percentage = n/max(n)) |>
dplyr::select(year, percentage)
# Create a data frame with author and year
figure_3c_nodes <- # 21 row
journal_df |>
select(from = JI_main, to = JI_ref, PY = PY_ref) %>%
dplyr::filter(from != to) |>
tidygraph::as_tbl_graph() |>
activate(edges) |>
as_tibble() |>
dplyr::select(CO = from,
year = PY) |>
bind_rows(journal_df |>
select(from = JI_main,
to = JI_ref,
PY = PY_ref) %>%
tidygraph::as_tbl_graph() |>
tidygraph::activate(edges) |>
tidygraph::as_tibble() |>
dplyr::select(CO = to,
year = PY)) |>
unique() |>
dplyr::group_by(CO) |>
dplyr::slice(which.min(year)) |>
dplyr::ungroup() |>
dplyr::select(year) |>
dplyr::group_by(year) |>
dplyr::count(year) |>
dplyr::filter(year >= 2002,
year <= 2022) |>
dplyr::ungroup() |>
dplyr::mutate(percentage = n / max(n)) |>
select(year, percentage)plotting figure 3b
figure_3c <-
figure_3c_nodes |>
mutate(type = "nodes") |>
bind_rows(figure_3c_edges |>
mutate(type = "links")) |>
ggplot(aes(x = year,
y = percentage,
color = type)) +
geom_point() +
geom_line() +
theme(legend.position = "right",
text = element_text(color = "black",
face = "bold",
family = "Times"),
plot.title = element_text(size = 25),
panel.background = element_rect(fill = "white"),
axis.text.y = element_text(size = 15,
colour = "black"),
axis.text.x = element_text(size = 15,
colour = "black",
angle = 45, vjust = 0.5
),
axis.title.x = element_text(size = 20),
axis.title.y = element_text(size = 20),
legend.text = element_text(size = "15"),
legend.title = element_blank()) +
labs(title = "Nodes and links through time",
y = "Percentage") +
scale_y_continuous(labels = scales::percent) +
scale_x_continuous(breaks = seq(2002, 2022, by = 1))
figure_3c
# library(patchwork)
# figure_3c / figure_3b | figure_3a3.4 Author Analysis
Table 4
table_4_authors |>
dplyr::select(authors_total, papers_total) %>%
DT::datatable(class = "cell-border stripe",
rownames = F,
filter = "top",
editable = FALSE,
extensions = "Buttons",
options = list(dom = "Bfrtip",
buttons = c("copy",
"csv",
"excel",
"pdf",
"print")))Creating the ASN - graph object
author_network_time <-
author_df |>
tidygraph::as_tbl_graph(directed = FALSE) |>
activate(nodes) |>
dplyr::mutate(components = tidygraph::group_components(type = "weak")) |>
dplyr::filter(components == 1) |>
dplyr::mutate(degree = centrality_degree(),
community = as.factor(group_louvain()))
author_network <-
author_df |>
dplyr::select(-PY) |>
dplyr::group_by(from, to) |>
dplyr::count() |>
dplyr::rename(weight = n) |>
tidygraph::as_tbl_graph(directed = FALSE) |>
activate(nodes) |>
# dplyr::mutate(components = tidygraph::group_components(type = "weak")) |>
# dplyr::filter(components == 1) |>
dplyr::mutate(degree = centrality_degree(),
community = as.factor(group_louvain()))Figure 4a clusters
figure_4a <-
author_network |>
activate(nodes) |>
data.frame() |>
dplyr::count(community) |>
slice(1:10) |>
ggplot(aes(x = reorder(community, n), y = n)) +
geom_point(stat = "identity") +
geom_line(group = 1) +
# geom_text(label = as.numeric(community),
# nudge_x = 0.5,
# nudge_y = 0.5,
# check_overlap = T) +
labs(title = "Communities by size",
x = "communities",
y = "Authors") +
theme(text = element_text(color = "black",
face = "bold",
family = "Times"),
plot.title = element_text(size = 25),
panel.background = element_rect(fill = "white"),
axis.text.y = element_text(size = 15,
colour = "black"),
axis.text.x = element_text(size = 15,
colour = "black"),
axis.title.x = element_text(size = 20),
axis.title.y = element_text(size = 20)
)
figure_4a
Figure 4b Longitudinal
# Create a dataframe with links
fig_1c_edges <-
author_network_time |>
activate(edges) |>
as_tibble() |>
dplyr::select(year = PY) |>
dplyr::count(year) |>
dplyr::filter(year >= 2002,
year <= 2022) |>
dplyr::mutate(percentage = n/max(n)) |>
dplyr::select(year, percentage)
# Create a data frame with author and year
fig_1c_nodes <- # 21 row
author_network_time |>
activate(edges) |>
as_tibble() |>
dplyr::select(author = from,
year = PY) |>
bind_rows(author_network_time |>
activate(edges) |>
as_tibble() |>
dplyr::select(author = to,
year = PY)) |>
unique() |>
dplyr::group_by(author) |>
dplyr::slice(which.min(year)) |>
dplyr::ungroup() |>
dplyr::select(year) |>
dplyr::group_by(year) |>
dplyr::count(year) |>
dplyr::filter(year >= 2002,
year <= 2022) |>
dplyr::ungroup() |>
dplyr::mutate(percentage = n / max(n)) |>
select(year, percentage)plotting figure 4b
figure_4b <-
fig_1c_nodes |>
mutate(type = "nodes") |>
bind_rows(fig_1c_edges |>
mutate(type = "links")) |>
ggplot(aes(x = year,
y = percentage,
color = type)) +
geom_point() +
geom_line() +
theme(legend.position = "right",
text = element_text(color = "black",
face = "bold",
family = "Times"),
plot.title = element_text(size = 25),
panel.background = element_rect(fill = "white"),
axis.text.y = element_text(size = 15,
colour = "black"),
axis.text.x = element_text(size = 15,
colour = "black",
angle = 45, vjust = 0.5
),
axis.title.x = element_text(size = 20),
axis.title.y = element_text(size = 20),
legend.text = element_text(size = "15"),
legend.title = element_blank()) +
labs(title = "Nodes and links through time",
y = "Percentage") +
scale_y_continuous(labels = scales::percent) +
scale_x_continuous(breaks = seq(2002, 2022, by = 1))
figure_4b
Filtering only the top 10 nodes with best degree in the first 6 clusters.
asn_TM_connected_1 <-
author_network |>
activate(nodes) |>
dplyr::mutate(community = as.numeric(community)) |>
# filter(community >= 6) |>
dplyr::filter(community == 1) |>
# group_by(community) |>
dplyr::mutate(degree_community = centrality_degree()) |>
dplyr::arrange(desc(degree_community)) |>
dplyr::slice(1:10)
asn_TM_connected_2 <-
author_network |>
activate(nodes) |>
dplyr::mutate(community = as.numeric(community)) |>
# filter(community >= 6) |>
dplyr::filter(community == 2) |>
# group_by(community) |>
dplyr::mutate(degree_community = centrality_degree()) |>
dplyr::arrange(desc(degree_community))|>
dplyr::slice(1:10)
asn_TM_connected_3 <-
author_network |>
activate(nodes) |>
dplyr::mutate(community = as.numeric(community)) |>
# filter(community >= 6) |>
dplyr::filter(community == 3) |>
# group_by(community) |>
dplyr::mutate(degree_community = centrality_degree()) |>
dplyr::arrange(desc(degree_community)) |>
dplyr::slice(1:10)
asn_TM_connected_4 <-
author_network |>
activate(nodes) |>
dplyr::mutate(community = as.numeric(community)) |>
# filter(community >= 6) |>
dplyr::filter(community == 4) |>
# group_by(community) |>
dplyr::mutate(degree_community = centrality_degree()) |>
dplyr::arrange(desc(degree_community)) |>
dplyr::slice(1:10)
asn_TM_connected_5 <-
author_network |>
activate(nodes) |>
dplyr::mutate(community = as.numeric(community)) |>
# filter(community >= 6) |>
dplyr::filter(community == 5) |>
# group_by(community) |>
dplyr::mutate(degree_community = centrality_degree()) |>
dplyr::arrange(desc(degree_community)) |>
dplyr::slice(1:10)
asn_TM_connected_6 <-
author_network |>
activate(nodes) |>
dplyr::mutate(community = as.numeric(community)) |>
# filter(community >= 6) |>
dplyr::filter(community == 6) |>
# group_by(community) |>
dplyr::mutate(degree_community = centrality_degree()) |>
dplyr::arrange(desc(degree_community)) |>
dplyr::slice(1:10)Saving the nodes we’re gonna show
nodes_community_1 <-
asn_TM_connected_1 |>
activate(nodes) |>
as_tibble() |>
dplyr::select(name)
nodes_community_2 <-
asn_TM_connected_2 |>
activate(nodes) |>
as_tibble() |>
dplyr::select(name)
nodes_community_3 <-
asn_TM_connected_3 |>
activate(nodes) |>
as_tibble() |>
dplyr::select(name)
nodes_community_4 <-
asn_TM_connected_4 |>
activate(nodes) |>
as_tibble() |>
dplyr::select(name)
nodes_community_5 <-
asn_TM_connected_5 |>
activate(nodes) |>
as_tibble() |>
dplyr::select(name)
nodes_community_6 <-
asn_TM_connected_6 |>
activate(nodes) |>
as_tibble() |>
dplyr::select(name)
nodes_selected_10 <-
nodes_community_1 |>
bind_rows(nodes_community_2,
nodes_community_3,
# nodes_community_4,
# nodes_community_5,
# nodes_community_6
)Filtering selected nodes
asn_selected_nodes <-
author_network |>
activate(nodes) |>
dplyr::filter(name %in% nodes_selected_10$name) |>
dplyr::mutate(degree = centrality_degree())
# dplyr::mutate(final_plot = tidygraph::group_components(type = "weak")) |>
# dplyr::filter(final_plot == 1)Figure 4c Author Network
figure_4c <-
asn_selected_nodes |>
ggraph(layout = "graphopt") +
geom_edge_link(width = 1,
colour = "lightgray") +
geom_node_point(aes(color = community,
size = degree)) +
geom_node_text(aes(label = name), repel = TRUE) +
theme_graph()
figure_4c
# library(patchwork)
# figure_4b / figure_4a | figure_4c 3.4.1 Ego top 10 authors
merging ego_networks
egos <-
tidygraph::as_tbl_graph(x = AU_ego_edges) %>%
tidygraph::activate(nodes) %>%
dplyr::left_join(AU_ego_nodes %>%
mutate(id = as.character(id)),
by = c("name" = "id")) %>%
dplyr::mutate(component = as.character(component))
egos |>
ggraph(layout = "graphopt") +
geom_edge_link(aes(width = weight),
colour = "lightgray") +
scale_edge_width(name = "Link strength") +
geom_node_point(aes(color = component,
size = degree)) +
geom_node_text(aes(label = Label), repel = TRUE) +
scale_size(name = "Degree") +
# scale_color_binned(name = "Communities") +
theme_graph()