Scientometric Analysis
Sebastian Robledo
20/12/2023
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
library(readxl)
library(httr)
tf<-"C:/Core Of Science/all_data_milton_organic_coffee.xlsx"
wos_scopus <- readxl::read_excel(tf,sheet = 1)
wos <- readxl::read_excel(tf, sheet = 2)
scopus <- readxl::read_excel(tf, sheet = 3)
reference_df <- readxl::read_excel(tf,sheet = 4)
journal_df <- readxl::read_excel(tf, sheet = 5)
author_df <- readxl::read_excel(tf, sheet = 6)
TC_all <- readxl::read_excel(tf,sheet = 7)
figure_1_data <- readxl::read_excel(tf, sheet = 8)
table_2_country <- readxl::read_excel(tf, sheet = 10)
figure_2_country_wos_scopus <- readxl::read_excel(tf, sheet = 11)
figure_2_country_wos_scopus_1 <-
readxl::read_excel(tf, sheet = 12) |>
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 <- readxl::read_excel(tf, sheet = 13)
table_4_authors <- readxl::read_excel(tf, sheet = 14)
AU_CO_links <- readxl::read_excel(tf, sheet = 15)
tos <- readxl::read_excel(tf, sheet = 16)
edges_tos <- readxl::read_excel(tf, sheet = 17)
nodes_tos <- readxl::read_excel(tf, sheet = 18)
SO_edges <- readxl::read_excel(tf, sheet = 19)
SO_nodes <- readxl::read_excel(tf, sheet = 20)
AU_ego_edges <- readxl::read_excel(tf, sheet = 21)
AU_ego_nodes <- readxl::read_excel(tf, sheet = 22)Summary of WoS and Scopus
table_1 <-
tibble(wos = length(wos$AU), # Create a dataframe with the values.
scopus = length(scopus$AU),
total = length(wos_scopus$AU))
table_1 %>%
DT::datatable(class = "cell-border stripe",
rownames = F,
filter = "top",
editable = FALSE,
extensions = "Buttons",
options = list(dom = "Bfrtip",
buttons = c("copy",
"csv",
"excel",
"pdf",
"print")))wos_scopus %>%
tidyr::separate_rows(DT, sep = ";") %>%
dplyr::count(DT, sort = TRUE)%>%
dplyr::mutate(percentage = n /sum(n),
percentage = percentage * 100,
percentage = round(percentage, digits = 2)) %>%
dplyr::rename(total = n) %>%
DT::datatable(class = "cell-border stripe",
rownames = F,
filter = "top",
editable = FALSE,
extensions = "Buttons",
options = list(dom = "Bfrtip",
buttons = c("copy",
"csv",
"excel",
"pdf",
"print")))Resutls
Scientometric Analysis
3.1 Scientific Production
Figure 1a - Scopus + WoS
Combine charts using Python Matplotlib & Reticulate
library(reticulate)
# create a new environment
# conda_create("r-reticulate")
# install Matplotlib
# conda_install("r-reticulate", "matplotlib")
# import Matplotlib (it will be automatically discovered in "r-reticulate")
plt <- import("matplotlib")
np <- import("numpy")# From Double get integers
# TC y
TC_all$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)
ax.bar(wx1, wy, color='orange', label = 'WoS', alpha=0.8, width=barw)
# Y2 - Total citations
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(0,1600) ######### <-----Important--------- """"Change Y2 Coordinate"""""## (0.0, 1600.0)# plt.annotate() customize numbers for each position
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='brown', 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 0x000002571BF159A0>, <matplotlib.axis.XTick object at 0x000002571BF15BB0>, <matplotlib.axis.XTick object at 0x000002571A501C70>, <matplotlib.axis.XTick object at 0x000002571BF45EE0>, <matplotlib.axis.XTick object at 0x000002571BFB5490>, <matplotlib.axis.XTick object at 0x000002571BFB5F40>, <matplotlib.axis.XTick object at 0x000002571BFBBA30>, <matplotlib.axis.XTick object at 0x000002571BFA9970>, <matplotlib.axis.XTick object at 0x000002571BFBBDF0>, <matplotlib.axis.XTick object at 0x000002571BFC48E0>, <matplotlib.axis.XTick object at 0x000002571BFC93D0>, <matplotlib.axis.XTick object at 0x000002571BFC9E80>, <matplotlib.axis.XTick object at 0x000002571BF37E80>, <matplotlib.axis.XTick object at 0x000002571BFCF6A0>, <matplotlib.axis.XTick object at 0x000002571BFCFDF0>, <matplotlib.axis.XTick object at 0x000002571BFD6C40>, <matplotlib.axis.XTick object at 0x000002571BFD6F70>, <matplotlib.axis.XTick object at 0x000002571BFDA580>, <matplotlib.axis.XTick object at 0x000002571BFDACD0>, <matplotlib.axis.XTick object at 0x000002571BFE3B20>, <matplotlib.axis.XTick object at 0x000002571BFEA610>, <matplotlib.axis.XTick object at 0x000002571BFDA970>, <matplotlib.axis.XTick object at 0x000002571BFEACD0>, <matplotlib.axis.XTick object at 0x000002571BFEDB20>], [Text(2023.0, 0, '2023'), Text(2022.0, 0, '2022'), Text(2021.0, 0, '2021'), Text(2020.0, 0, '2020'), Text(2019.0, 0, '2019'), Text(2018.0, 0, '2018'), Text(2017.0, 0, '2017'), Text(2016.0, 0, '2016'), Text(2015.0, 0, '2015'), Text(2014.0, 0, '2014'), Text(2013.0, 0, '2013'), Text(2012.0, 0, '2012'), Text(2011.0, 0, '2011'), Text(2010.0, 0, '2010'), Text(2009.0, 0, '2009'), Text(2008.0, 0, '2008'), Text(2007.0, 0, '2007'), Text(2006.0, 0, '2006'), Text(2005.0, 0, '2005'), Text(2004.0, 0, '2004'), Text(2003.0, 0, '2003'), Text(2002.0, 0, '2002'), Text(2001.0, 0, '2001'), Text(2000.0, 0, '2000')])fig.autofmt_xdate(rotation = 70)
# The Y1 ticks depends from tpy scale limits
yticks = [0,5,10,15,20,25,30,35,40] ########## <-----Important---- Choose scale .. just specify which numbers you want
ax.set_yticks(yticks)
# Export Figure as SVG
plt.savefig("paola_13.svg")
plt.show()
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_2a <-
figure_2_country_wos_scopus_1 |>
activate(edges) |>
# tidygraph::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
Figure 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",
year = as.numeric(year)) |>
bind_rows(figure_2c_edges |>
mutate(type = "links",
year = as.numeric(year))) |>
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