R Notebook
Mapping COVID-19 Research Publications: A Bibliometric and Topic Modeling Approach
Macitas, G.I., Panoy, A., Cayot, J.G., & Javiero, F.Y.
Division of Physical Science and Mathematics, College of Arts and Sciences, UP Visayas
Description
This study aims to analyze COVID-19 research articles from the Philippines, utilizing a combination of bibliometric analysis and topic modeling. The dataset, obtained from Scopus, includes various metadata fields such as authors, titles, publication year, source title, volume, issue, page information, citation count, affiliations, abstract, keywords, funding details, and document type.
Bibliometric Analysis
The bibliometric analysis in this study aims to provide a comprehensive overview of the publication landscape for COVID-19 research articles in the Philippines. By analyzing metadata from research articles published between 2020 and 2024, we can identify trends, patterns, and key contributors in this rapidly evolving field.
# Import libraries
library(tidyverse)## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ dplyr 1.1.4 ✔ readr 2.1.5
## ✔ forcats 1.0.0 ✔ stringr 1.5.1
## ✔ ggplot2 3.5.1 ✔ tibble 3.2.1
## ✔ lubridate 1.9.3 ✔ tidyr 1.3.1
## ✔ purrr 1.0.2
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()
## ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errorslibrary(dplyr)
library(ggplot2)
library(tidyr)
library(scales)##
## Attaching package: 'scales'
##
## The following object is masked from 'package:purrr':
##
## discard
##
## The following object is masked from 'package:readr':
##
## col_factorlibrary(tidytext)
library(wordcloud)## Loading required package: RColorBrewerlibrary(tm)## Loading required package: NLP
##
## Attaching package: 'NLP'
##
## The following object is masked from 'package:ggplot2':
##
## annotatelibrary(SnowballC)
library(topicmodels)
library(NLP)
library(ldatuning)
library(tm)
library(reshape2)##
## Attaching package: 'reshape2'
##
## The following object is masked from 'package:tidyr':
##
## smiths# Import dataset
data <- read.csv("ph_covid_scopus.csv")
colnames(data)## [1] "Authors" "Author.full.names"
## [3] "Author.s..ID" "Title"
## [5] "Year" "Source.title"
## [7] "Volume" "Issue"
## [9] "Art..No." "Page.start"
## [11] "Page.end" "Page.count"
## [13] "Cited.by" "Link"
## [15] "Affiliations" "Authors.with.affiliations"
## [17] "Abstract" "Author.Keywords"
## [19] "Funding.Details" "Funding.Texts"
## [21] "Publisher" "Document.Type"
## [23] "Publication.Stage" "Open.Access"
## [25] "Source" "EID"# Check dimensions
dim(data)## [1] 3025 26# Check the data
glimpse(data)## Rows: 3,025
## Columns: 26
## $ Authors <chr> "Bandojo A.J.P.; Uy C.; Tanedo M.D.", "Latug…
## $ Author.full.names <chr> "Bandojo, Agnes Jocelyn P. (58958384900); Uy…
## $ Author.s..ID <chr> "58958384900; 13406148500; 58958987400", "58…
## $ Title <chr> "Antecedents of Intent to Reuse Learning Man…
## $ Year <int> 2024, 2024, 2024, 2024, 2024, 2024, 2024, 20…
## $ Source.title <chr> "Review of Integrative Business and Economic…
## $ Volume <chr> "13", "23", "", "19", "106", "61", "16", "28…
## $ Issue <chr> "3", "1", "", "03-Mar", "", "3", "4", "1", "…
## $ Art..No. <chr> "", "", "", "e0280144", "104441", "", "1686"…
## $ Page.start <chr> "82", "287", "", "", "", "1069", "", "", "81…
## $ Page.end <chr> "99", "294", "", "", "", "1089", "", "", "99…
## $ Page.count <int> 17, 7, NA, NA, NA, 20, NA, NA, 18, 6, NA, NA…
## $ Cited.by <int> 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0,…
## $ Link <chr> "https://www.scopus.com/inward/record.uri?ei…
## $ Affiliations <chr> "Research Center for Social Sciences and Edu…
## $ Authors.with.affiliations <chr> "Bandojo A.J.P., Research Center for Social …
## $ Abstract <chr> "Amidst the global challenges posed by COVID…
## $ Author.Keywords <chr> "intent to reuse; learning management system…
## $ Funding.Details <chr> "", "", "", "Japan Network for Public Intere…
## $ Funding.Texts <chr> "", "", "", "This work was supported by the …
## $ Publisher <chr> "GMP Press & Printing Co.,", "Technoscience …
## $ Document.Type <chr> "Article", "Article", "Article", "Article", …
## $ Publication.Stage <chr> "Final", "Final", "Article in press", "Final…
## $ Open.Access <chr> "", "All Open Access; Gold Open Access", "",…
## $ Source <chr> "Scopus", "Scopus", "Scopus", "Scopus", "Sco…
## $ EID <chr> "2-s2.0-85188843791", "2-s2.0-85187177428", …# Check structure of variables
str(data)## 'data.frame': 3025 obs. of 26 variables:
## $ Authors : chr "Bandojo A.J.P.; Uy C.; Tanedo M.D." "Latugan P.; Carabacan J.J.; Bonicillo G.; Cayog J.; Eyawa M.Q.; Cairel M.T.; Ngohayon J.M." "Aldaba F.T.; Geronimo J.A.U." "Luu M.N.; Imoto A.; Matsuo Y.; Huy N.T.; Qarawi A.; Alhady S.T.M.; Van Truong L.; Yoshino R.; Duc N.T.M.; Tabei"| __truncated__ ...
## $ Author.full.names : chr "Bandojo, Agnes Jocelyn P. (58958384900); Uy, Chin (13406148500); Tanedo, Marivic D. (58958987400)" "Latugan, P. (58926381500); Carabacan, J.J. (58926603900); Bonicillo, G. (58926818900); Cayog, J. (58925955500);"| __truncated__ "Aldaba, Fernando T. (6508320932); Geronimo, John Anthony U. (58938742700)" "Luu, Mai Ngoc (57226109824); Imoto, Atsuko (56216340000); Matsuo, Yoshimi (58946820000); Huy, Nguyen Tien (5720"| __truncated__ ...
## $ Author.s..ID : chr "58958384900; 13406148500; 58958987400" "58926381500; 58926603900; 58926818900; 58925955500; 58926604000; 58926819000; 58070998200" "6508320932; 58938742700" "57226109824; 56216340000; 58946820000; 57209480434; 57224096129; 57224078312; 57226146472; 58946859000; 5878558"| __truncated__ ...
## $ Title : chr "Antecedents of Intent to Reuse Learning Management System in the New Normal: A Structural Equation Analysis" "Analysis and Characterization of Municipal Solid Wastes Generated in Ifugao State University Potia Campus: A Ba"| __truncated__ "Shock responsiveness of the Philippine social protection system: Three case studies" "Anxiety and its risk factors among non-Japanese residents living in Japan undergoing COVID-19 situation: A cros"| __truncated__ ...
## $ Year : int 2024 2024 2024 2024 2024 2024 2024 2024 2024 2024 ...
## $ Source.title : chr "Review of Integrative Business and Economics Research" "Nature Environment and Pollution Technology" "Natural Resources Forum" "PLoS ONE" ...
## $ Volume : chr "13" "23" "" "19" ...
## $ Issue : chr "3" "1" "" "03-Mar" ...
## $ Art..No. : chr "" "" "" "e0280144" ...
## $ Page.start : chr "82" "287" "" "" ...
## $ Page.end : chr "99" "294" "" "" ...
## $ Page.count : int 17 7 NA NA NA 20 NA NA 18 6 ...
## $ Cited.by : int 0 0 0 0 0 0 1 0 0 0 ...
## $ Link : chr "https://www.scopus.com/inward/record.uri?eid=2-s2.0-85188843791&partnerID=40&md5=67e7c153950d764d158a393d08399e6a" "https://www.scopus.com/inward/record.uri?eid=2-s2.0-85187177428&doi=10.46488%2fNEPT.2024.v23i01.024&partnerID=4"| __truncated__ "https://www.scopus.com/inward/record.uri?eid=2-s2.0-85187655432&doi=10.1111%2f1477-8947.12445&partnerID=40&md5="| __truncated__ "https://www.scopus.com/inward/record.uri?eid=2-s2.0-85188200865&doi=10.1371%2fjournal.pone.0280144&partnerID=40"| __truncated__ ...
## $ Affiliations : chr "Research Center for Social Sciences and Education, College of Commerce and Business Administration, University "| __truncated__ "Ifugao State University, Potia Campus, Potia, Alfonso Lista, Ifugao, Philippines" "Economics Department, Ateneo de Manila University, Quezon City, Philippines; College of Social Work and Communi"| __truncated__ "Department of Internal Medicine, University of Medicine and Pharmacy at Ho Chi Minh City, Ho Chi Minh City, Vie"| __truncated__ ...
## $ Authors.with.affiliations: chr "Bandojo A.J.P., Research Center for Social Sciences and Education, College of Commerce and Business Administrat"| __truncated__ "Latugan P., Ifugao State University, Potia Campus, Potia, Alfonso Lista, Ifugao, Philippines; Carabacan J.J., I"| __truncated__ "Aldaba F.T., Economics Department, Ateneo de Manila University, Quezon City, Philippines; Geronimo J.A.U., Coll"| __truncated__ "Luu M.N., Department of Internal Medicine, University of Medicine and Pharmacy at Ho Chi Minh City, Ho Chi Minh"| __truncated__ ...
## $ Abstract : chr "Amidst the global challenges posed by COVID-19, educational institutions worldwide, including the Philippines, "| __truncated__ "The end of the COVID-19 pandemic resulted in the total return of students and employees in Ifugao State Univers"| __truncated__ "Philippine households face large-scale risks, like natural disasters and inflation, which can threaten their fi"| __truncated__ "Introduction In the context of collective efforts taken in Japan to control the spread of COVID-19, the state o"| __truncated__ ...
## $ Author.Keywords : chr "intent to reuse; learning management system; new normal; structural equation analysis" "Municipal solid waste; Waste composition; Waste management; Waste volume" "disaster management; managing covariate shocks; safety nets; shock responsive social protection; social assistance" "" ...
## $ Funding.Details : chr "" "" "" "Japan Network for Public Interest Activities" ...
## $ Funding.Texts : chr "" "" "" "This work was supported by the dormant deposits of Japan Network for Public Interest Activities (JANPIA), Japan"| __truncated__ ...
## $ Publisher : chr "GMP Press & Printing Co.," "Technoscience Publications" "John Wiley and Sons Inc" "Public Library of Science" ...
## $ Document.Type : chr "Article" "Article" "Article" "Article" ...
## $ Publication.Stage : chr "Final" "Final" "Article in press" "Final" ...
## $ Open.Access : chr "" "All Open Access; Gold Open Access" "" "All Open Access; Gold Open Access" ...
## $ Source : chr "Scopus" "Scopus" "Scopus" "Scopus" ...
## $ EID : chr "2-s2.0-85188843791" "2-s2.0-85187177428" "2-s2.0-85187655432" "2-s2.0-85188200865" ...Number of Publications Per Year
# Sum number of publication by year
yearly_count <- data %>%
group_by(Year) %>%
summarize(Total_Publication = n())
yearly_count# Plot the number of publications per year
ggplot(yearly_count, aes(x = Year, y = Total_Publication)) +
geom_line(color = "blue") + # Line plot
geom_point(color = "blue") + # Points on the line plot
labs(title = "Yearly Publication Trend",
x = "Year",
y= "Total Number of Publications") +
theme_minimal()
Document-Type Distribution
# Display the count of unique document types
doc_type <- data %>%
count(Document.Type, sort = TRUE)
print(doc_type)## Document.Type n
## 1 Article 2013
## 2 Conference paper 474
## 3 Letter 205
## 4 Review 152
## 5 Book chapter 77
## 6 Note 49
## 7 Editorial 23
## 8 Erratum 13
## 9 Book 8
## 10 Short survey 5
## 11 Data paper 4
## 12 Retracted 2# Plot
ggplot(doc_type, aes(x=n, y=reorder(Document.Type,n), fill = n)) +
geom_bar(stat = "identity")+
scale_fill_gradient(low = "#89CFF0", high = "darkblue") +
labs(title = "Distribution of Document Types",
x = "Total Number of Publication",
y = "Document Type") +
theme_minimal() +
guides(fill = "none")
Most Prolific Author
# Count number of publications associated with authors
author_count <- data %>%
separate_rows(Authors, sep = "; ") %>%
count(Authors, sort = TRUE)
head(author_count)# Create a plot of the top 10 authors
top_authors <- head(author_count,10)
ggplot(top_authors, aes(x = n, y = reorder(Authors,n), fill=n)) +
geom_bar(stat = "identity") +
scale_fill_gradient(low = "#89CFF0", high = "darkblue") +
labs(title = "Top 10 Most Prolific Authors",
x = "Publication Count",
y = "Author") +
theme_minimal() +
guides(fill = FALSE)## Warning: The `<scale>` argument of `guides()` cannot be `FALSE`. Use "none" instead as
## of ggplot2 3.3.4.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.
Top Publishing Journals
# Count publications per journal
journal_count <- data %>%
count(Source.title, sort=TRUE)
head(journal_count, 10)# Plot publications data
top_journal <- head(journal_count,10)
recoded <- top_journal %>%
mutate(Source.title = recode(Source.title,
"2022 IEEE 14th International Conference on Humanoid, Nanotechnology, Information Technology, Communication and Control, Environment, and Management, HNICEM 2022" = "2022 IEEE International Conference on HNICEM",
"2021 IEEE 13th International Conference on Humanoid, Nanotechnology, Information Technology, Communication and Control, Environment, and Management, HNICEM 2021" = "2021 IEEE International Conference on HNICEM"))
ggplot(recoded, aes(x = n, y = reorder(Source.title, n), fill = n)) +
geom_bar(stat = "identity") +
scale_fill_gradient(low = "#89CFF0", high = "darkblue") +
scale_y_discrete(labels = function(x) str_wrap(x, width = 50)) +
scale_x_continuous(labels = scales::comma) +
labs(title = "Top 10 Publishing Journals",
x = "Publication Count",
y = "Journals") +
theme_minimal() +
guides(fill = FALSE)
Citations
# Citations per year
cite_count <- data %>%
group_by(Year) %>%
summarize(Citations = sum(Cited.by))
cite_count# Plot citations
ggplot(cite_count, aes(x = Year, y = Citations)) +
geom_line(color = "blue") + # Line plot
geom_point(color = "blue") + # Points on the line plot
labs(title = "Yearly Citations",
x = "Year",
y= "Number of Citations") +
theme_minimal()
# Average Citation per Year
avg_cite_count <- data %>%
group_by(Year) %>%
summarize("Median Citation Per Article" = median(Cited.by, na.rm = TRUE))
avg_cite_count# Most cited paper
most_cited_papers <- data %>%
arrange(desc(Cited.by)) %>%
select(c(Title, Cited.by))
head(most_cited_papers, 10)# Most cited journal
most_cited_journals <- data %>%
group_by(Source.title) %>%
summarize(Cited.by = n()) %>%
arrange(desc(Cited.by))
head(most_cited_journals, 10)Publications by Affiliations
# Count number of publications associated with different affiliations
affil_count <- data %>%
separate_rows(Affiliations, sep = "; ") %>%
count(Affiliations, sort = TRUE)
head(affil_count)# Create a dataframe with the sample data
df <- data.frame(Authors.with.affiliations = data$Authors.with.affiliations, stringsAsFactors = FALSE)
# Split the Authors.with.affiliations column into multiple rows based on semicolons
df <- df %>%
separate_rows(Authors.with.affiliations, sep = ";") %>%
mutate(Affiliation = trimws(Authors.with.affiliations))
# Extract country from affiliation using regular expression
df <- df %>%
mutate(Country = sub('.*\\b([A-Za-z]+)$', '\\1', Affiliation))
head(df)country <- data.frame(Country = df$Country, stringsAsFactors = TRUE)
head(country, 10)country_count <- country %>%
count(Country, sort = TRUE) %>%
filter(!grepl("\\.", Country))
head(country_count,10)top_country <- head(country_count, 10)
country_recoded <- top_country %>%
mutate(Country = recode(Country, "States" = "United States", "Kingdom" = "United Kingdom"))
ggplot(country_recoded, aes(x = n, y = reorder(Country, n), fill = n)) +
geom_bar(stat = "identity") +
scale_fill_gradient(low = "#89CFF0", high = "darkblue") +
scale_y_discrete(labels = function(x) str_wrap(x, width = 50)) +
scale_x_continuous(labels = scales::comma) +
labs(title = "Countries With Most Researchers",
x = "Number of Researchers",
y = "Country") +
theme_minimal() +
guides(fill = FALSE)
Most Common Topic
# Most common words based on the titles
# Create custom stopwords
custom_stopwords <- tribble(
~word, ~lexicon,
"covid", "covid",
"19", "19",
"pandemic", "pandemic",
"analysis", "analysis",
"based", "based",
"impact","impact",
"philippines", "philippines",
"2", "2",
"study", "study",
"filipino", "filipino",
"philippine", "philippine",
"approach", "approach",
"cov","cov",
"sars", "sars",
"disease", "disease",
"factors", "factors",
"health","health",
"results", "results",
"data", "data",
"patients","patients",
"2020", "2020",
"2021","2021",
"2022","2022",
"2023","2023",
"2024","2024",
"significant","significant",
"research","research",
"findings","findings",
"related","related",
"author","author",
"participants","participants"
)
stop_words2 <- stop_words %>%
bind_rows(custom_stopwords)# Top words in the title
tidy_title <- data %>%
unnest_tokens(word, Title) %>%
anti_join(stop_words2) %>%
count(word, sort=TRUE)## Joining with `by = join_by(word)`head(tidy_title,20)# Word cloud
wordcloud(words = tidy_title$word, freq = tidy_title$n, min.freq = 5,
max.words = 100, random.order = FALSE, rot.per = 0.35,
colors = brewer.pal(8, "Dark2"))
# Most common word in abstracts
tidy_abstract <- data %>%
unnest_tokens(word, Abstract) %>%
anti_join(stop_words2) %>%
count(word, sort=TRUE)## Joining with `by = join_by(word)`head(tidy_abstract, 20)# Wordcloud
wordcloud(words = tidy_abstract$word, freq = tidy_abstract$n, min.freq = 5,
max.words = 100, random.order = FALSE, rot.per = 0.35,
colors = brewer.pal(8, "Dark2"))
Topic Modeling
# Import libraries
library(textstem)## Loading required package: koRpus.lang.en## Loading required package: koRpus## Loading required package: sylly## For information on available language packages for 'koRpus', run
##
## available.koRpus.lang()
##
## and see ?install.koRpus.lang()##
## Attaching package: 'koRpus'## The following object is masked from 'package:tm':
##
## readTagged## The following object is masked from 'package:readr':
##
## tokenize# Tidy data
data$ID <- c(1:dim(data)[1])
abstract_tidy <- data %>%
unnest_tokens(word, Abstract) %>%
anti_join(stop_words2)## Joining with `by = join_by(word)`# Clean and lemmatize
abstract_tidy$word <- removeNumbers(abstract_tidy$word)
abstract_tidy$word <- removePunctuation(abstract_tidy$word)
abstract_tidy$word <- lemmatize_words(abstract_tidy$word)# Create document-term matrix
abstract_dtm <- abstract_tidy %>%
count(ID, word) %>%
cast_dtm(ID, word, n) %>%
as.matrix()
dim(abstract_dtm)## [1] 3025 15660#n_topics = 2:50
#start_time <- Sys.time()
#lda_ldatuning_result <- FindTopicsNumber(
# abstract_dtm, topics = n_topics,
# metrics = c("Griffiths2004","CaoJuan2009", "Arun2010", "Deveaud2014"),
# method = "Gibbs", control = list(seed = 42), mc.cores = 4L, verbose = TRUE)
#end_time <- Sys.time()#print(end_time-start_time)lda_ldatuning_result <- readRDS("lda_ldatuning_result.rds")# Plot result
FindTopicsNumber_plot(lda_ldatuning_result)
# Define the range of candidate numbers of topics
# n_topics <- c(2,3,4,5,6,14,17,20,25,35,45,50)
# Number of folds for cross-validation
# k <- 5
# Create folds
# set.seed(42) # For reproducibility
# folds <- cut(seq(1, nrow(abstract_dtm)), breaks = k, labels = FALSE)
# Initialize a list to store perplexity results
# perplexity_results <- list()
#start_perp_time <- Sys.time()
# Loop over each number of topics
# for (num_topics in n_topics) {
# fold_perplexities <- numeric(k)
# Perform k-fold cross-validation
# for (i in 1:k) {
# Split data into training and validation sets
# validation_indices <- which(folds == i, arr.ind = TRUE)
# validation_set <- abstract_dtm[validation_indices, ]
# training_set <- abstract_dtm[-validation_indices, ]
#
# # Train LDA model on training set
# lda_model <- LDA(training_set, k = num_topics, method = "Gibbs", control = list(seed = 42))
#
# # Compute perplexity on validation set
# fold_perplexities[i] <- perplexity(lda_model, newdata = validation_set)}
#
# # Average perplexity over all folds for this number of topics
# avg_perplexity <- mean(fold_perplexities)
# perplexity_results[[as.character(num_topics)]] <- avg_perplexity
# cat("Number of topics:", num_topics, "Average Perplexity:", avg_perplexity, "\n")}
#
# end_perp_time <- Sys.time()
#
# # Convert results to a data frame
# perplexity_df <- data.frame(
# num_topics = n_topics,
# avg_perplexity = unlist(perplexity_results))
#
# # Print the results
# print(perplexity_df)# time elapsed
# print(end_perp_time-start_perp_time)It takes 2.23 hours to finish.
perplexity_df <- readRDS("D:\\UPV\\THIRD YEAR\\Stat 197\\Final Project\\perplexity_df")# Plot perplexity
ggplot(perplexity_df, aes(x = num_topics, y = avg_perplexity)) +
geom_line() +
geom_point() +
labs(title = "Perplexity vs. Number of Topics",
x = "Number of Topics",
y = "Average Perplexity") +
theme_minimal()
# Train based on 6-topic model
#num_topics <- 6 # Replace with the optimal number of topics
#lda_model <- LDA(abstract_dtm, k = num_topics, method = "Gibbs", control = list(seed = 42))
#summary(lda_model)lda_model <- readRDS("final_lda_model.rds")# Remove empty terms
term_names <- colnames(abstract_dtm)
# Identify empty or whitespace terms
empty_terms <- which(term_names == "" | grepl("^\\s+$", term_names))
# Remove empty or whitespace terms
abstract_dtm <- abstract_dtm[, -empty_terms]
# Train again the lda model aboveabstract_topics <- tidy(lda_model, matrix = "beta")
abstract_topics# View Words by Topic
# Find beta (word-topic probability)
word_probs <- abstract_topics %>%
group_by(topic) %>%
top_n(15, beta) %>%
ungroup() %>%
mutate(term2 = fct_reorder(term, beta))
word_probs# Plot top 15 words per topic
ggplot(word_probs, aes(term2, beta, fill = as.factor(topic))) +
geom_col(show.legend = FALSE) +
facet_wrap( ~ topic, scales = "free") +
coord_flip() +
xlab("Term") +
ylab("Beta")
# Classify Reviews by Topic
# Use gamma (topic-document probability)
abstract_gamma <- tidy(lda_model, matrix = "gamma")
abstract_gamma$document <- as.numeric(abstract_gamma$document)
abstract_gamma <- as.data.frame(abstract_gamma)
head(abstract_gamma)# Obtain documents per topic and sort by gamma
topic_1 <- abstract_gamma[abstract_gamma$topic == 1,]
topic_1 <- topic_1[order(topic_1$document),]
topic_2 <- abstract_gamma[abstract_gamma$topic == 2,]
topic_2 <- topic_2[order(topic_2$document),]
topic_3 <- abstract_gamma[abstract_gamma$topic == 3,]
topic_3 <- topic_3[order(topic_3$document),]
topic_4 <- abstract_gamma[abstract_gamma$topic == 4,]
topic_4 <- topic_4[order(topic_4$document),]
topic_5 <- abstract_gamma[abstract_gamma$topic == 5,]
topic_5 <- topic_5[order(topic_5$document),]
topic_6 <- abstract_gamma[abstract_gamma$topic == 6,]
topic_6 <- topic_6[order(topic_6$document),]# Classify gamma values
gam1 <- as.numeric(topic_1$gamma > quantile(abstract_gamma$gamma, 0.7, 5))
gam2 <- as.numeric(topic_2$gamma > quantile(abstract_gamma$gamma, 0.7, 5))
gam3 <- as.numeric(topic_3$gamma > quantile(abstract_gamma$gamma, 0.7, 5))
gam4 <- as.numeric(topic_4$gamma > quantile(abstract_gamma$gamma, 0.7, 5))
gam5 <- as.numeric(topic_5$gamma > quantile(abstract_gamma$gamma, 0.7, 5))
gam6 <- as.numeric(topic_6$gamma > quantile(abstract_gamma$gamma, 0.7, 5))# Create dataframe with gamma per document
text_topic <- data.frame(ID = topic_1$document, topic1 = gam1, topic2 = gam2, topic3 = gam3, topic4 = gam4, topic5 = gam5, topic6 = gam6)
dim(text_topic)## [1] 3025 7# Match results with original dataframe
abstract_match <- merge(data, text_topic, by= "ID", all.x = TRUE)
head(abstract_match)# Count Number of Reviews that fall on each topic
# Initialize an empty list to store counts
topic_counts_list <- list()
# Loop over each topic column and count occurrences
for (i in 1:6) {
topic <- paste0("topic", i)
counts <- abstract_match %>%
count(!!sym(topic)) %>%
rename(topic = !!sym(topic), count = n)
# Add topic column identifier
counts$topic_col <- topic
# Append to list
topic_counts_list[[i]] <- counts
}
# Combine all counts into a single data frame
all_topic_counts <- bind_rows(topic_counts_list)
# Remove odd rows
true_topic_counts <- all_topic_counts[seq(2, nrow(all_topic_counts), by = 2), 2:3]
rownames(true_topic_counts) <- NULL
true_topic_countsggplot(true_topic_counts, aes(x = count, y = topic_col, fill = count)) +
geom_bar(stat = "identity") +
scale_fill_gradient(low = "pink", high = "maroon") +
scale_y_discrete(labels = function(x) str_wrap(x, width = 50)) +
scale_x_continuous(labels = scales::comma) +
labs(title = "Number of Abstracts Classified Per Topic",
x = "Number of Abstracts",
y = "Topics") +
theme_minimal() +
guides(fill = FALSE)
# Find top documents per topic
k <- 1
dat <- abstract_gamma[abstract_gamma$topic == k,]
top_abstract <- dat[order(dat$gamma, decreasing = TRUE),]
head(top_abstract)data[data$ID == top_abstract[1,1], ]$Abstract## [1] "Aim: To examine the moderating effect of burnout on the relationship between nurses' competence, professional competence and professional values. Background: There is a preponderance of burnout studies of nurses before and during the COVID-19 pandemic. However, little is known about burnout's moderating influence on nurses' professionalism, competence and values during the ongoing pandemic. Methods: This study used a correlational, cross-sectional design and convenience sampling to recruit 809 Saudi nurses employed in four government hospitals. Four self-report instruments were used to collect data from August 2021 to March 2022. Spearman Rho, structural equation modelling and multi-group moderation analysis were used to analyse the data. Results: Most of the participants had high burnout. There was a moderate to high correlation between the dimensions of professional competence, nurse professional values, burnout and nurse competence. Both professional competence and nurse professional values significantly affected nurse competence. Nurse professional values also had a direct effect on professional competence. Professional competence and nurse professional values on nurse competence may vary among those with low, moderate and high levels of burnout, suggesting the moderating effect of burnout. Conclusions: The positive impacts of nurse professional values and professional competence on nursing competence were validated in this study, as was the evidence of burnout's moderating effect on the studied connections. To avoid burnout, it is necessary to increase organizational knowledge and support for nurses, their working conditions and the environment through strategies that promote well-being and empowerment. Implications for nursing management: Nurse managers can implement policies to help nurses minimize burnout and improve their professional values and competence. Management policies may be considered include increased staffing, opportunities to improve technical abilities through education and vocational training programs, and increased diversity of job assignments. © 2022 John Wiley & Sons Ltd."# Check second top abstract
data[data$ID == top_abstract[2,1], ]$Abstract## [1] "Aim: To determine the parsimonious model of the interrelationships of personal resilience, social support, loneliness and quality of life (QoL) and to identify the mediating effect of loneliness among nursing students amidst the pandemic. Background: The coronavirus 2019 (COVID-19) pandemic led to control measures which increased loneliness among students due to disrupted social connections and sudden shift to virtual learning. During these changes, resilience and social support can significantly influence how learners deal with pandemic-related stresses thus, minimizing loneliness and enhancing QoL. Although studies posited the influence of resilience and social support with loneliness and QoL, these were conducted prior the COVID-19 pandemic wherein loneliness and feelings of isolation play a crucial part. Design: Correlational, theory testing using covariance-based structural equation modeling Methods: A total of 550 nursing students from a comprehensive university in Manila, Philippines were recruited from September to October 2021 and answered four standardized, validated scales. Results: A good and parsimonious model (x2/df = 2.84, RMSEA = 0.058, GFI = 0.999, CFI = 0.999, PNFI = 0.048) highlighted the mediating effect of loneliness between social support, personal resilience and QoL. While personal resilience positively influenced the physical and psychological domains of QoL, social support positively affected the social relationships and environmental domains. Loneliness was a strong, negative predictor of the psychological and social domains of QoL and had a moderate, negative effect on the physical domain. Personal resilience also mediated the influence of social support on loneliness and QoL. Conclusion: Social support and personal resilience positively affected QoL, while loneliness had a negative effect. Through the mediation of loneliness, the effects of social support on QoL decreases. However, the mediation of resilience further decreases loneliness and improves QoL. The presented model assists nurse educators and administrators in developing strategies to enhance social support, resilience and QoL among students while mitigating the negative effects loneliness during the pandemic. Tweetable abstract: Loneliness and Resilience are mediators of student nurses’ quality of life during COVID-19 pandemic. Social support is the common predictor. © 2022 Elsevier Ltd"# Check third top abstract
data[data$ID == top_abstract[3,1], ]$Abstract## [1] "Aims: To investigate the mediating role of moral resilience and moral courage in the association between moral distress and moral injury. Background: There is a preponderance of nursing literature about moral distress, moral resilience, moral courage and moral injury. However, examining moral resilience and moral courage as mediators remain underreported during the COVID-19 pandemic and in the context of a developing nation. Design: Correlational, cross-sectional design compliant with the STROBE guidelines. Methods: A convenience sample of nurses (n = 412) from the Philippines were recruited using social media platforms (e.g., Facebook, Messenger, Twitter). Four self-report and validated scales (8-item Moral-Distress Appraisal Scale, 21-item Nurses’ Moral Courage Scale, 17-item Rushton Moral Resilience Scale and 10-item Moral Injury Symptom Scale: Healthcare Professionals Version) were used to collect data from January to July 2022. Pearson's r, bivariate analysis and multistage regression analyses were used for data analysis. Results: This study afforded a model that depicted the interrelationships of moral distress, moral resilience, moral courage and moral injury. Moral distress has a negative impact on moral resilience and moral courage while positively affecting moral injury. Moral resilience positively influences moral courage while having a negative impact on moral injury. Moral courage has an indirect impact on moral injury. Finally, moral resilience and moral courage demonstrated a mediating effect between moral distress and moral injury. Conclusions: Healthcare organizations, policymakers and nurse managers should include policies and programs that include improving approaches to modifying workplace conditions and evaluating nurses' moral resilience and courage. Nurse managers need to advocate ethics education and professionalism. Nurses must practice self-care strategies to strengthen morally resilient and courageous therapeutic practices. © 2023 Elsevier Ltd"Topic 1 is mostly about behavioral effects of COVID-19.
# Find top documents per topic
k <- 2
dat <- abstract_gamma[abstract_gamma$topic == k,]
top_abstract <- dat[order(dat$gamma, decreasing = TRUE),]
data[data$ID == top_abstract[1,1], ]$Abstract## [1] "The inclusion of physical education in the curriculum in the new normal, and with the mandate on the adoption of flexible learning at the tertiary level caused by the Covid-19 pandemic forced physical educators to do quaranteaching or teaching in the quarantine. This study described the lived experiences of physical education (PE) teachers in quaranteaching utilizing Edmund Husserl’s phenomenological qualitative research design. A total of seven PE teachers from two universities participated in this study which achieved data saturation. Anchored to Colaizzi’s phenomenological approach and guided by different philosophies, data were analyzed and presented in two prong clustering. The results revealed that tertiary PE teachers experienced difficulties and yet learnt to innovate in terms of teaching the course, assessing, monitoring, and giving feedback to students, dealing with the internet and technology, and handling physical and cognitive situations. It is recommended that tertiary PE teachers undergo pedagogical and technological trainings and workshops especially in the utilization of different learning platforms to help them improve their pedagogy and in the innovative use of digital tools and technologies to teach PE in the new normal. digitaliness Purpose - The sudden shift from face-to-face to online learning compelled educators including physical educators to teach PE in the midst of quarantine, with no face-to-face interaction. This study described the experiences of tertiary PE teachers in teaching PE in the quarantine. Methodology - The study utilized the descriptive phenomenological qualitative research design by Edmund Husserl. The participants were seven PE teachers from two state universities in Cebu, Philippines, selected through purposive snowball sampling. Data gathered were analyzed through Colaizzi’s approach of data analysis and presented in two prong clustering. Findings - The study revealed that the experiences of tertiary PE teachers in teaching PE during quarantine can be described under two classifications. They experienced difficulties and yet learnt to innovate in terms of teaching the course, assessing, monitoring, and giving feedback to students, using the internet and technology, and handling physical and cognitive situations. Significance - The study has provided a better understanding of PE teachers by revealing their challenges, struggles, needs, strengths and weaknesses in quaranteaching. Results of the study could serve as a basis for implementing trainings, workshops and webinars related to pedagogy and technology enhanced learning to assist PE teachers in teaching in the new normal. © 2022. Malaysian Journal of Learning and Instruction. All Rights Reserved."data[data$ID == top_abstract[2,1], ]$Abstract## [1] "The shift to distance teaching and learning during the COVID-19 pandemic brought about a real challenge for both instructors and students. To face these difficulties in teaching undergraduate Chemistry courses at the University of Santo Tomas, a blended learning strategy in the context of teaching and learning of Physical Chemistry 1 and Analytical Chemistry for Chemical Engineering students were employed. Here, we present an online strategy that facilitated the transition from traditional face-to-face learning to full online instruction. This is a five-component blended learning strategy referred to as Discover, Learn, Practice, Collaborate and Assess (DLPCA). In DLPCA, the asynchronous part of the teaching was achieved through broadcast of pre-recorded lecture videos on YouTube to allow students to study and progress with learning at their own pace. The synchronous part of the teaching was conducted using video conferencing platforms, such as Zoom or Google Meet. The DLPCA strategy was presented and discussed to the students prior to its implementation. The analysis of the teaching and learning experience based on three indicators (i) student's learning experience, (ii) student's academic performance and (iii) instructor observations showed that DLPCA had a positive impact on students and instructors. The identified challenges were stability of internet connection and instructor's familiarity with readily available internet-based teaching tools, such as video conferencing software. Instructors must also find means to improve their interaction with students and maintain student interest and engagement during online classes. The survey also indicated that most of the students are satisfied with the DLCPA strategy. Hence, this strategy is considered a manageable and effective alternative that can be adapted to full online instruction to other undergraduate Chemistry lecture courses. Overall, the findings and insights in this study will add valuable resources for further hybrid instruction in the post-COVID-19 time in higher education. © 2021 Institution of Chemical Engineers"data[data$ID == top_abstract[3,1], ]$Abstract## [1] "This paper reports on a phenomenographic study investigating former grade 12 STEM students’ conceptions of online learning during COVID-19 pandemic. Purposive sampling was employed in selecting participants who were enrolled in a private school in Laguna, Philippines that formally implemented a fully online learning setup for SY 2020-2021. By using semi-structured interviews, results conveyed an outcome space comprised of nine (9) core categories of description: 1) online learning marred with unfavorable sentiments, 2) online learning plagued by technical barriers, 3) online learning accentuated the use of online tools, 4) online learning provided aberrational learning experiences, 5) online learning hampered by the learning environment, 6) online learning adversely affected students’ well-being, 7) online learning reaffirmed the importance of social interactions, 8) online learning exhibited unsatisfactory perceptions to teaching, and 9) online learning exhibited dissatisfaction towards institutional support. Within each core category, a variety of subcategories fulfilled the anatomy of experience and describing the complexity of the relation between STEM students and online learning during COVID-19 pandemic. Analysis of all core categories of description revealed a generally unfavorable and unenthusiastic view of online learning during COVID-19 Pandemic, with ‘online learning provided aberrational learning experiences’ as the most prominent core category of description, emphasizing how the deviation from the normal setup adversely affected their learning experiences. Findings revealed in this study can be beneficial in formulating future policy changes and bases of interventions to improve the conduct of online learning and provide a more effective pandemic response in the educational system. © 2022, Duzce University, Faculty of Education. All rights reserved."data[data$ID == top_abstract[4,1], ]$Abstract## [1] "Online learning keeps on growing and being explored during this pandemic. Most of the tertiary institutions here in the Philippines have shifted from face-to-face to online instructions in all courses including physics subjects. The conduct of physics classes through virtual classrooms has its advantages and presents some challenges that are why it is important to understand the perspective and experiences of students. This study aimed to describe the experiences of students in learning physics in a virtual classroom, specifically their experiences in preparation, learning management, opportunities, and challenges they have encountered. There were four themes that emerged in this study: Technological Preparations, Self-Preparations, Challenges in Online Learning, and Leaning Management and Opportunities. It was realized in this study that students were prepared to take online classes both in terms of their technological and personal aspect. Despite their preparations, they encountered challenges such as network issues, time constraints, and distractions in their learning environment at home. However, students were optimistic, so, they saw opportunities in these challenges to improve their technological skills, time management, and maximized online resources for better understanding of physics concepts. It is recommended for future research to explore the experiences of teachers in teaching online physics classes. © 2021 International Council of Associations for Science Education (ICASE). All rights reserved."data[data$ID == top_abstract[5,1], ]$Abstract## [1] "Introduction: An evaluation of the online medical course was conducted to assess student readiness, engagement, and satisfaction at the San Beda University College of Medicine in Manila during the COVID-19 pandemic. Methodology: A convergent mixed methods approach was done with a quantitative online survey and a qualitative thematic analysis of focus group discussions (FGD) with medical students. A total of 440 students participated in the survey while 20 students participated in the FGDs. Results: The medical students were sufficiently equipped with computers and internet connections that allowed them to access the online medical course from their homes. The 5 themes identified during the study that were relevant to education were: Student readiness for online learning, Learning Management System (LMS) and internet connectivity, teaching and learning activities, the value of engagements, and teaching effectiveness of the faculty. The combined quantitative and qualitative analysis revealed vital issues that affect student learning. This included the need for students to interact with fellow students and to be engaged with their faculty. The issues that affect teaching included the need for continuing faculty training and management skills in delivering the full online course. Conclusion: The success of online education rests heavily on the interactions of the students, the teachers, and the knowledge. Student interactions, managerial and skills training for the faculty, and providing students with a mix of synchronous and asynchronous activities are the most effective means to ensure the effective delivery of online medical courses. © 2023 TAPS. All rights reserved."Topic 2 is more likely about education and learning during the COVID-19 pandemic.
# Find top documents per topic
k <- 3
dat <- abstract_gamma[abstract_gamma$topic == k,]
top_abstract <- dat[order(dat$gamma, decreasing = TRUE),]
data[data$ID == top_abstract[1,1], ]$Abstract## [1] "Purpose: This paper aims to offer Sitio San Roque, an informal settlement in the Philippines as a case study to explore long-term \"forgetful\" urban development planning in the Philippines, and the renewed visibility of the urban poor under COVID-19 lockdown. It connects scholarship on informality to issues of housing and political rights in Metro Manila to further investigate how vulnerable communities in the Global South are faring in the pandemic. Design/methodology/approach: This is an exploration of Sitio San Roque, an informal settler community in Metropolitan \"Metro\" Manila, Philippines. This paper refers to recent journalistic reports pertaining to the community's ongoing evictions and arrests while under Metro Manila's \"enhanced community quarantine.\" Furthermore, it converses with literature from disciplines including health-care policy, urban studies and recent studies on COVID-19 and vulnerable communities to critically discuss the plight of the urban poor in the pandemic-stricken Metro Manila. Findings: The urban poor and members of informal communities such as Sitio San Roque are especially vulnerable to contracting COVID-19 because of precarious livelihoods and housing instability. The creation of informality in Metro Manila can be traced to political tensions, economic agendas and development planning since the time of Marcos' administration and also to global restructuring during the 1990s. However, also important to note is that under Metro Manila's lockdown, informal settlers are further disenfranchised and stigmatized via ongoing demolitions and evictions, as well as by processes of policing and criminalization by the state. The use of military and police personnel as a way to enforce lockdown in the metropolis further impedes on the rights of informal settlers and the urban poor. Originality/value: Recent scholarship and reports discuss the challenges for informal communities and the urban poor in navigating the COVID-19 pandemic, mainly due to their housing conditions and loss of economic stability. This paper contributes to a critical understanding of these issues by adding the dimensions of political and housing rights. It refers to the case study of members of Sitio San Roque, who have experienced continuous threats of demolitions and arrests by state police for protesting the lack of government aid under lockdown. Both military approaches of governance and housing informality work in tandem to expose the vulnerabilities of the urban poor in Metro Manila's pandemic. Finally, this paper extends on urban studies scholar Gavin Shatkin's concept of “forgetful planning” (2004) by applying his discussions to the current context. Informal settlers have long been “forgotten” by the state's development plans, but are now remembered and deemed more visible in Metro Manila's ECQ. © 2021, Emerald Publishing Limited."data[data$ID == top_abstract[2,1], ]$Abstract## [1] "This introduction presents an overview of the key concepts discussed in the subsequent chapters of this book. The book explains how pandemic policies were framed and enforced in the five cities, how marginalised populations experienced and navigated the impacts of COVID-19 and state measures. It discusses the impact of COVID19 and lockdown measures on informal workers in Karachi. The book explains the role of welfare organisations played in providing food and rations. It identifies the gaps in social security schemes and the possible way forward to reform the social security provision to informal workers. The book discusses the measures taken by the government and non-government entities in Dhaka during the pandemic. It narrates how four local governments in Metro Manila-Caloocan, Manila, Pasig and Quezon City-operated in an environment of securitised pandemic response. The book ends with policy recommendations, highlighting the importance of secure workplaces for street vendors as a key agenda in Jakarta's post-COVID recovery strategies. © 2024 The right of Redento B. Recio, Kazi Nazrul Fattah, Nausheen H. Anwar, Noman Ahmed, Iderlina Mateo-Babiano, Michele Acuto, Ian Jayson Hecita and Shiva Nouri to be identified as the authors of the editorial material, and of the authors for their individual chapters, has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved."data[data$ID == top_abstract[3,1], ]$Abstract## [1] "Recent correspondence letters to the editor of this journal pointed on the praxis of faith in the context of COVID-19 pandemic. The #ChurchinAction presents the role of religious interventions in times of COVID-19. However, a question is raised: how to promote interreligious dialogue in the context of COVID-19 pandemic? To address this question, people of different faiths/religions should come together in promoting the life and dignity of the human person, a sense of community, and participation, respecting the rights and responsibilities of each person, helping the poor and vulnerable sectors of the society, respecting the dignity of work and the rights of the workers, solidarity and caring for the whole creation. In this perspective, interreligious dialogue highlighted the importance of world religions in order to provide a spiritual intervention. The dialogue of life, dialogue of action, dialogue of theological exchange, and dialogue of religious experiences promote interreligious dialogue in the context of COVID-19 pandemic. © Crown copyright 2020."data[data$ID == top_abstract[4,1], ]$Abstract## [1] "2020 was to be a landmark year for setting targets to stop biodiversity loss and prevent dangerous climate change. However, COVID-19 has caused delays to the 15th Conference of the Parties (COP) of the UN Convention on Biological Diversity and the 26th COP of the UN Framework Convention on Climate Change. Negotiations on the Global Biodiversity Framework and the second submission of Nationally Determined Contributions under the Paris Agreement were due to take place at these COPs. There is uncertainty as to how the COVID-19 disruption will affect the negotiations, whether parties will pursue more ambitious actions or take a weaker stance on issues. Our policy analysis shows there are broad opportunities for climate and biodiversity frameworks to better respond to COVID-19, by viewing future pandemics, biodiversity loss, and climate change as interconnected problems. Importantly, there needs to be greater focus on agriculture and food systems in discussions, establishing safeguards for carbon markets, and implementing nature-based solutions in meeting the Paris Agreement goals. We can no longer delay action to address the biodiversity and climate emergencies, and accelerating sustainable recovery plans through virtual spaces may help keep discussions and momentum before the resumption of in-person negotiations. Non-technical summary:. High ambition needed at UN biodiversity and climate conferences to address pandemics, biodiversity, climate change, and health. © The Author(s), 2021."data[data$ID == top_abstract[5,1], ]$Abstract## [1] "As the world was mired in distress, some leaders saw opportunities to exploit the pandemic and further consolidate their grip on power. It is, thus, the objective of this article to discuss how, when, and why the state’s coercive apparatus has been instrumentalized by its leader amid a crisis. It will also explain how such apparatus has shaped both the aura of invincibility of the state and social order within the polity. The deployment of the Philippine National Police by President Rodrigo Duterte will be analyzed and discussed. The main argument of the article is that while the police has been given extensive powers to amplify the state’s power and assist in administrating a crisis-stricken society, they have also been instrumentalized to bolster an illiberal regime. In particular, the police were bestowed positions of authority within the state’s pandemic response apparatus, provided a broader leeway to wield violence through a contentious anti-terrorism law, and mobilized to unfairly enforce government-imposed measures. Thus, amid the COVID-19 pandemic in Duterte’s Philippines, civil liberties were violated, cultures of violence and impunity worsened, and the executive powers were consolidated. © 2021 SAGE Publications."Topic 3 is more like about policy-making.
# Find top documents per topic
k <- 4
dat <- abstract_gamma[abstract_gamma$topic == k,]
top_abstract <- dat[order(dat$gamma, decreasing = TRUE),]
data[data$ID == top_abstract[1,1], ]$Abstract## [1] "A stock is an investment wherein you purchase one of the company's stocks which by means is called a share. Most investors think that when they purchase a stock, it will go up the value. It will result also in to impact of an increase to the company's stocks as well. Because there are many stocks to check and invest in, the researcher chose the US stock market and considered in this study as a prime and one of the most powerful stock markets in the world. A so-called portfolio investment means a group of assets on which these are the chosen US companies. The goal of this study is to identify and look for a portfolio that outperforms benchmark markets (New York Stock Exchange, Dow Jones, and S&P 500) during the situation of pandemic Covid-19. The top 30 companies in the NYSE list are considered as the investment pool wherein some are tagged also as blue chips. From this, the researcher was able to develop and look for a portfolio that can outperform the markets during months between March to August 2020. Using the safety-first model to back-test, it was observed from the return statistics of the back-test period that a hopeful investor can outperform the benchmark markets during times of pandemic. Also, in these pandemic times, it is important for the investors to know the current situation and impact where their money is going and if it is advisable to invest. Also, it is obvious that when you are an investor, you should know when and where your money earned profit and have good investment options. It is important in this study to know what are the substantial risks during a pandemic and what could be the long-term strategy for your chosen stocks. For first time investors, this paper will help them assess their budget on stocks, where to invest their money, and the impact of a pandemic on their money. © 2021 ACM."# Check the second top abstract for this topic
data[data$ID == top_abstract[2,1], ]$Abstract## [1] "The use of time-series analysis to examine aviation data trends through time comes crucial in planning its future. The prophet is an additive model that fits non-linear patterns. It functions best with historical data from various seasons and time series with significant seasonal impacts. This research looked closely into the aviation data in Zamboanga Peninsula, Jolo, and Tawi-Tawi to give a clearer picture of its impact on the sector and forecast passenger and aircraft movement in the coming months to see whether the impact of the opening in the aviation industry can be sustained. The final data comprise 51 data points for flight arrivals and departures and 51 data points for passenger arrivals and departures. Data show the decline in passengers and aircrafts arriving and departing in major airports in Zamboanga Peninsula, Jolo, and Tawi-Tawi during the pandemic. However, an increasing trend was observed years after the pandemic hit the region. Findings during the training and testing phase revealed that different models attained varied results; however, there are models which attained a higher degree of accuracy as depicted in the RMSE and R2. This indicates that predicting passenger and aircraft movement using models with higher accuracy is similar to real data thus, it is viable in predicting future values. Forecasting results further show a gradually increasing trend of aircraft and passenger arrivals in the major airports in Zamboanga Peninsula, Jolo, and Tawi-Tawi despite some observed smaller forecasted values. © 2022 IEEE."# Check the third top abstract for this topic
data[data$ID == top_abstract[3,1], ]$Abstract## [1] "The Philippine stock market, just like most of its neighbors in the region, was seriously impacted by the global pandemic COVID-19. Investors remain hopeful while continuing to seek great ones in the damaged market. This paper developed a methodology for portfolio selection and optimization with the use of technical analysis, machine learning techniques, and portfolio optimization model. The combined methods of technical analysis, K-means clustering algorithm, and mean-variance portfolio optimization model will result in the development of the proposed TAKMV method. The study aims to integrate these three important analyses to identify portfolio investments. This paper uses the average annual risk and annual rate of return data for the years 2018 and 2020 to form the clusters and assessed the stocks that correspond to the investor's technical strategy such as Moving Average Convergence/Divergence (MACD) and Hybrid MACD with Arnaud Legoux Moving Average (ALMA). This paper solved the risk minimization problem on selected shares of the companies, based on the mean-variance portfolio optimization model. There are 230 and 239 companies for 2018 and 2020, respectively, listed in Philippine Stock Market, and all simulations were performed in MATLAB environment platform. Results showed that MACD strategy dominates the MACD-ALMA strategy in terms of the number of assets with a positive annual rate of return. The MACD works efficiently in the pre-COVID-19 condition while MACD-ALMA works efficiently during-COVID-19 condition, regardless of the number of assets with a positive annual rate of return. The results also show that the maximum expected portfolio return (RP) can be achieved using the MACD and MACD-ALMA in the pre-and during-COVID-19 conditions, respectively. The MACD-ALMA shows an advantage during high-risk market conditions and can also provide maximum RP. The performance of the TAKMV method was validated by applying its results and comparing it to the next year's historical price. The 2018 results were compared to 2019 data and the 2020 results were compared to 2021 data. For consistency, the comparison was applied to the same company per portfolio. Simulation results show that the MACD strategy is more effective compared to MACD-ALMA. © 2023 The Authors"data[data$ID == top_abstract[4,1], ]$Abstract## [1] "Purpose – This study attempts to establish if the markets for the two most popular cryptocurrencies in the world, Bitcoin and Ethereum, follow weak-form market efficiency across various landmarks in time. Design/Methodology/Approach – Traditional testing for establishing weak-form market efficiency rests on whether the price series exhibits a random walk process, which implies that future prices cannot be predicted. However, not all random walk series automatically imply weak-form market efficiency, since some asset price behaviors may exhibit non-constant variance. In such cases, the GARCH model can be used to test for the presence of market efficiency. Since structural breaks in the prices of both cryptocurrencies are common, tests for market efficiency were carried out using sub-temporal price windows. In both price series, the last time window coincided with the 2020 COVID-19 pandemic period. Findings – Results of the GARCH analyses showed that the volatility and persistence parameters (α and β, respectively) in the Bitcoin and Ethereum models were all statistically significant, implying that prices in their sub-temporal markets were generally weak-form inefficient. The observed market inefficiency in both cryptocurrencies can be attributed to various factors like the price manipulation of crypto whales, security issues, and increased media attention, which led to inflows of information that helped big investors beat and gain from the market by successfully predicting the trend in future prices. During the 2020 COVID-19 pandemic period, both cryptocurrencies’ prices were observed to rise significantly, similar to the case of the 2017 Bitcoin price bubble. A cointegrating regression between Bitcoin and Ethereum prices during this period, however, showed a spurious relationship. Despite the absence of a long run relationship between these two price series, the current price bubbles in the cryptocurrency markets are speculated to be tied together. Research Implications – Players in the cryptocurrency market must always be cautious in making investment decisions regarding this type of asset since the markets are generally price inefficient and risky; any idiosyncratic decision that may be triggered by a price bubble burst in one cryptocurrency market may or may not serve as a signal that the other market will do the same. Since the Bitcoin and Ethereum prices were shown to exhibit volatility spillover and persistence, investors can use this information to make informed decisions as to whether to invest in these cryptocurrencies despite the huge risks that are magnified during the COVID-19 pandemic. © 2021 International Academy of Global Business and Trade. All rights reserved."data[data$ID == top_abstract[5,1], ]$Abstract## [1] "Face masks have been widely used since the start of the COVID-19 pandemic. Facial detection and recognition technologies, such as the iPhone's Face ID, heavily rely on seeing the facial features that are now obscured due to wearing a face mask. Currently, the only way to utilize Face ID with a mask on is by having an Apple Watch as well. As such, this paper intends to find initial means of a reliable personal facial recognition system while the user is wearing a face mask without having the need for an Apple Watch. This may also be applicable to other security systems or measures. Through the use of Multi-Task Cascaded Convolutional Networks or MTCNN, a type of neural network which identifies faces and facial landmarks, and FaceNet, a deep neural network utilized for deriving features from a picture of a face, the masked face of the user could be identified and more importantly be recognized. Utilizing MTCNN, detecting the masked faces and automatically cropping them from the raw images are done. The learning phase then takes place wherein the exposed facial features are given emphasis while the masks themselves are excluded as a factor in recognition. Data in the form of images are acquired from taking multiple pictures of a certain individual's face as well as from repositories online for other people's faces. Images used are taken in various settings or modes such as different lighting levels, facial angles, head angles, colors and designs of face masks, and the presence or absence of glasses. The goal is to recognize whether it is the certain individual or not in the image. The training accuracy is 99.966% while the test accuracy is 99.921%. © 2022 IEEE."data[data$ID == top_abstract[6,1], ]$Abstract## [1] "Closed-circuit television camera (CCTV) and thermal imaging devices are used to detect febrile individuals entering establishments for Coronavirus 2019 (COVID-19) containment. Real-time tracking in post-COVID is manually checked by security personnel, which has risks of less efficiency due to human errors, as advance thermal cameras are unaffordable for some business owners. The main goal is converting an installed CCTV interfaced with infrared sensor to develop an economical thermal screening system with acoustic alarm. In this project, the colored and heatmap images transmitted from the thermal camera were processed through OpenCV. A calibration method was also performed to validate the temperature reading from the thermal camera. The project comes with graphical user interface (GUI) connected into a database, which visually tracks individuals exhibits elevated body temperature. The performance of the system shows above 95% accuracy upon conducting an inexpensive calibration check. The significance of this project is highlighting the effective mitigation of virus spread which offers safe and contactless analysis of potential individuals showing early symptoms of COVID-19. Additional features can be added for future work such as facemask detector, multiple thermal camera setup, and Login Options making the device and application exclusively for business owners. © 2022 IEEE."Topic 4 is most likely about technologies, systems and models that are helpful during COVID-19.
# Find top documents per topic
k <- 5
dat <- abstract_gamma[abstract_gamma$topic == k,]
top_abstract <- dat[order(dat$gamma, decreasing = TRUE),]
data[data$ID == top_abstract[1,1], ]$Abstract## [1] "Background: Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) variant of concern (VoC) Omicron (B.1.1.529) has rapidly spread around the world, presenting a new threat to global public human health. Due to the large number of mutations accumulated by SARS-CoV-2 Omicron, concerns have emerged over potentially reduced diagnostic accuracy of reverse-transcription polymerase chain reaction (RT-qPCR), the gold standard diagnostic test for diagnosing coronavirus disease 2019 (COVID-19). Thus, we aimed to assess the impact of the currently endemic Omicron sublineages BA.4 and BA.5 on the integrity and sensitivity of RT-qPCR assays used for coronavirus disease 2019 (COVID-19) diagnosis via in silico analysis. We employed whole genome sequencing data and evaluated the potential for false negatives or test failure due to mismatches between primers/probes and the Omicron VoC viral genome. Methods: In silico sensitivity of 12 RT-qPCR tests (containing 30 primers and probe sets) developed for detection of SARS-CoV-2 reported by the World Health Organization (WHO) or available in the literature, was assessed for specifically detecting SARS-CoV-2 Omicron BA.4 and BA.5 sublineages, obtained after removing redundancy from publicly available genomes from National Center for Biotechnology Information (NCBI) and Global Initiative on Sharing Avian Influenza Data (GISAID) databases. Mismatches between amplicon regions of SARS-CoV-2 Omicron VoC and primers and probe sets were evaluated, and clustering analysis of corresponding amplicon sequences was carried out. Results: From the 1164 representative SARS-CoV-2 Omicron VoC BA.4 sublineage genomes analyzed, a substitution in the first five nucleotides (C to T) of the amplicon's 3′-end was observed in all samples resulting in 0% sensitivity for assays HKUnivRdRp/Hel (mismatch in reverse primer) and CoremCharite N (mismatch in both forward and reverse primers). Due to a mismatch in the forward primer's 5′-end (3-nucleotide substitution, GGG to AAC), the sensitivity of the ChinaCDC N assay was at 0.69%. The 10 nucleotide mismatches in the reverse primer resulted in 0.09% sensitivity for Omicron sublineage BA.4 for Thai N assay. Of the 1926 BA.5 sublineage genomes, HKUnivRdRp/Hel assay also had 0% sensitivity. A sensitivity of 3.06% was observed for the ChinaCDC N assay because of a mismatch in the forward primer's 5′-end (3-nucleotide substitution, GGG to AAC). Similarly, due to the 10 nucleotide mismatches in the reverse primer, the Thai N assay's sensitivity was low at 0.21% for sublineage BA.5. Further, eight assays for BA.4 sublineage retained high sensitivity (more than 97%) and 9 assays for BA.5 sublineage retained more than 99% sensitivity. Conclusion: We observed four assays (HKUnivRdRp/Hel, ChinaCDC N, Thai N, CoremCharite N) that could potentially result in false negative results for SARS-CoV-2 Omicron VoCs BA.4 and BA.5 sublineages. Interestingly, CoremCharite N had 0% sensitivity for Omicron Voc BA.4 but 99.53% sensitivity for BA.5. In addition, 66.67% of the assays for BA.4 sublineage and 75% of the assays for BA.5 sublineage retained high sensitivity. Further, amplicon clustering and additional substitution analysis along with sensitivity analysis could be used for the modification and development of RT-qPCR assays for detecting SARS-CoV-2 Omicron VoC sublineages. © 2022 Wiley Periodicals LLC."data[data$ID == top_abstract[2,1], ]$Abstract## [1] "We assessed the non-inferiority of homologous boosting compared with heterologous boosting with the recombinant protein vaccine, SCB-2019, in adults previously immunized with different COVID-19 vaccines. Three equal cohorts (N ~ 420) of Philippino adults (18–80 years) previously immunized with Comirnaty, CoronaVac or Vaxzevria COVID-19 vaccines were randomized 1:1 to receive homologous or heterologous (SCB-2019) boosters. Neutralizing antibodies against prototype SARS-CoV-2 (Wuhan-Hu-1) were measured in all participants and against Delta variant and Omicron sub-lineages in subsets (30‒50 per arm) 15 days after boosting. Participants recorded solicited adverse events for 7 days and unsolicited and serious adverse events until Day 60. Prototype SARS-CoV-2 neutralizing responses on Day 15 after SCB-2019 were statistically non-inferior to homologous Vaxzevria boosters, superior to CoronaVac, but lower than homologous Comirnaty. Neutralizing responses against Delta and Omicron BA.1, BA.2, BA.4 and BA.5 variants after heterologous SCB-2019 were higher than homologous CoronaVac or Vaxzevria, but lower than homologous Comirnaty. Responses against Omicron BF.7, BQ.1.1.3, and XBB1.5 following heterologous SCB-2019 were lower than after homologous Comirnaty booster but significantly higher than after Vaxzevria booster. SCB-2019 reactogenicity was similar to CoronaVac or Vaxzevria, but lower than Comirnaty; most frequent events were mild/moderate injection site pain, headache and fatigue. No vaccine-related serious adverse events were reported. Heterologous SCB-2019 boosting was well tolerated and elicited neutralizing responses against all tested SARS-COV-2 viruses including Omicron BA.1, BA.2, BA.4, BA.5, BF.7, BQ.1.1.3, and XBB1.5 sub-lineages that were non-inferior to homologous boosting with CoronaVac or Vaxzevria, but not homologous Comirnaty booster. © 2024 Clover Biopharmaceuticals Inc. Published with license by Taylor & Francis Group, LLC."data[data$ID == top_abstract[3,1], ]$Abstract## [1] "Purpose of review: Recent outbreaks of poliomyelitis in countries that have been free of cases for decades highlight the challenges of eradicating polio in a globalized interconnected world beset with a novel viral pandemic. We provide an epidemiological update, advancements in vaccines, and amendments in public health strategy of poliomyelitis in this review. Recent findings: Last year, new cases of wild poliovirus type 1 (WPV1) were documented in regions previously documented to have eradicated WPV1 and reports of circulating vaccine-derived poliovirus type 2 (cVDPV2) and 3 (cVDPV3) in New York and Jerusalem made international headlines. Sequencing of wastewater samples from environmental surveillance revealed that the WPV1 strains were related to WPV1 lineages from endemic countries and the cVDPV2 strains from New York and Jerusalem were not only related to each other but also to environmental isolates found in London. The evidence of importation of WPV1 cases from endemic countries, and global transmission of cVDPVs justifies renewed efforts in routine vaccination programs and outbreak control measures that were interrupted by the COVID-19 pandemic. After the novel oral poliovirus vaccine type 2 (nOPV2) received emergency authorization for containment of cVDPV2 outbreaks in 2021, subsequent reduced incidence, transmission rates, and vaccine adverse events, alongside increased genetic stability of viral isolates substantiates the safety and efficacy of nOPV2. The nOPV1 and nOPV3 vaccines, against type 1 and 3 cVDPVs, and measures to increase accessibility and efficacy of inactivated poliovirus vaccine (IPV) are in development. Summary: A revised strategy utilizing more genetically stable vaccine formulations, with uninterrupted vaccination programs and continued active surveillance optimizes the prospect of global poliomyelitis eradication. Copyright © 2023 Wolters Kluwer Health, Inc. All rights reserved."data[data$ID == top_abstract[4,1], ]$Abstract## [1] "The global pandemic disease COVID-19 commands a number of host immunological responses to deadly cytokine-storm manifestations. Mitigation of incoming infectious comorbidities ushered by immunosuppression such as tuberculosis (TB) is warranted including the requirement of new drugs. Employment of computational screening methods and intricate selection of putative viral protein targets allows identification of lead mitigating compounds against COVID-19. Antitubercular natural products may offer privileged starting points as potential antagonistic prospects for SARS-CoV-2 non-structural proteins implicated in viral replication and infection mechanisms. In this study, antitubercular natural products from Philippine medicinal plants are repurposed as inhibitors of vital SARS-CoV-2 nsps such as 3CLpro (3-chymotrypsin-like protease), PLpro (papain-like protease), and RdRp (RNA-dependent RNA polymerase) using in silico methods. A total of 67 antitubercular compounds were computationally screened using molecular docking in 3CLpro, PLpro, and RdRp followed by subsequent ADMET (absorption, distribution, metabolism, excretion, and toxicity) analysis of top-binding compounds to assess their drug-likeness and pharmacokinetic properties. Based on the results of our molecular docking analysis, 10 antitubercular natural product cocktails exhibited strong binding propensities against 3CLpro, PLpro, and RdRp. Several compounds demonstrated multi-targeting inhibition to at least two vital non-structural proteins (nsps). Vobtusine lactone (1), deoxyvobtusine lactone (6), and deoxyvobtusine (7) exhibited superior target-specific binding while globospiramine (3) was found to be a potential multi-targeting alkaloid with amenable drug-like properties. The study elaborated for the first time, the computational screening of antitubercular compounds against three vital targets of SARS-CoV-2 yielding potential drug-like compounds against COVID-19 infection. A combination of excellent binding scores, dynamic stability, drug-likeness, and pharmacokinetic properties identified vobtusine lactone (1), deoxyvobtusine lactone (6), deoxyvobtusine (7), and the triple targeting globospiramine (3) as promising leads representing new anti-SARS-CoV-2 scaffolds. © 2022, Department of Science and Technology. All rights reserved."data[data$ID == top_abstract[5,1], ]$Abstract## [1] "Background: Due to the Covid-19 pandemic, a newly formulated Chinese herbal medicine, Ji Qin Yin (JQY), has emerged as a potential antiviral adjuvant. JQY contains nine medicinal herbs with established pharmacological activities but unknown mechanisms. Thus, this first-attempt study aims to provide a novel electrochemical perspective to reveal the antiviral and immunomodulatory properties of JQY for antiviral drug development. Methods: The bioenergy-stimulating characteristics and therapeutic efficacy of JQY were assessed by adopting voltammetry and Microbial Fuel Cells (MFCs) as novel bioenergy-based platforms. Phytochemical analyses correlated the total phytonutrient content with the antioxidant and electron-stimulating activities of JQY. The antiviral and immunomodulatory potential of JQY were assessed via molecular docking. Significant findings: JQY has significant bioenergy-stimulation properties with a 1.85 ± 0.19 amplification factor, implying bioenergy-steered and electron-mediated antiviral and immunomodulatory efficacy. Major flavonoids (e.g., chlorogenic acid, rosmarinic acid) identified by HPLC exhibit electron-shuttling (ES) properties due to their ortho- or para-dihydroxyl groups. These dihydroxyl groups also bound enzymes via hydrogen bonds; thus, chlorogenic acid and rosmarinic acid exhibited higher LibDock scores than the standard drugs. All test compounds passed Lipinski's rule of five with low drug toxicity risks, suggesting prospective candidates for drug development. © 2023"Topic 5 revolves around vaccines, drugs and experiments.
# Find top documents per topic
k <- 6
dat <- abstract_gamma[abstract_gamma$topic == k,]
top_abstract <- dat[order(dat$gamma, decreasing = TRUE),]
data[data$ID == top_abstract[1,1], ]$Abstract## [1] "Background: Patients admitted due to severe COVID-19 pneumonia are at high risk for malnutrition and worsening of their clinical condition. Patients with type 2 diabetes admitted for COVID-19 pneumonia have an increased risk for poor clinical outcomes. Adequate nutrition is recommended to augment a strong immune response. The American Society of Parenteral and Enteral Nutrition (ASPEN), Philippine Society of Parenteral and Enteral Nutrition (PHILSPEN), European Society of Parenteral and Enteral Nutrition (ESPEN) recommend an energy intake goal of 15-20 kcal/kg actual body weight (ABW) per day or 70-80% of caloric requirements after the acute phase of critical illness, with recommended protein intake of 1.2-2.0 g/kg ABW per day. This study aims to provide an association between calorie and protein intake with negative clinical outcomes. Methods: This is a retrospective cohort study of 55 mechanically ventilated SARS COV-2 RT-PCR positive patients admitted in the critical unit of Chinese General Hospital between April 1, 2020 to December 30, 2020. Clinical profile taken include: sex, age, height weight, BMI, comorbidities, and components of the modified SOFA score and APACHE II score. Calorie and protein intake from day 3 to day 7 of ICU admission were taken. Clinical outcome data were in-hospital mortality, number of days of ICU stay, hospital admission, vasopressor use and mechanical ventilation. Results: Majority of the patients included in the study were male, elderly, overweight, and with comorbidities such as hypertension, diabetes, and chronic kidney disease. Majority of the patients were on mechanical ventilation and on vasopressors for more than 1 week. In-hospital mortality accounted for 65.5% of cases. For patients without diabetes, calorie intake was 16.9 kcal/kg/day and protein intake was 0.72 g/kg/day, while patients with diabetes had a calorie intake of 20.2 kcal/kg/day and protein intake of 0.86 g/kg/day. Based on the 70% cutoff, patients with diabetes were noted to have been provided with more adequate protein (P-value= 0.027). Higher caloric intake was inversely associated with in-hospital mortality among patients younger than 75 years old (P-value=0.026) and among patients with diabetes (P-value=0.003). Higher calorie intake was also significantly associated with decreased duration of pressors among patients with diabetes (P-value=0.021). Higher protein intake positively associated with the number of days admitted among patients with lower modified SOFA scores (P-value=0.041) and among patients with diabetes (P-value=0.021). All other associations did not display significant results (all P-values>0.05). Conclusion: Based on this study, increased caloric intake was associated with increased survival among patients less than 75 years, but no association was found in patients 75 years and older. Among patients with lower mortality risk, increased protein intake was associated with longer duration of hospital stay, however patients with higher risk had higher in-hospital mortality regardless of protein and calorie intake. Among patients with diabetes, higher calorie intake was associated with increased survival and decreased duration of pressor requirement, while increased protein was associated with longer length of hospital stay. Treatment for COVID-19 pneumonia, however, was not established at this time and there was significant in-hospital mortality among these mechanically ventilated patients. © 2021, Philippine College of Physicians. All rights reserved."data[data$ID == top_abstract[2,1], ]$Abstract## [1] "Patients diagnosed with diabetes mellitus (DM) who are infected with severe acute respiratory syndrome coronavirus 2 (SARS-COV-2) belong to the most vulnerable patient subgroups. Emerging data has shown increased risks of severe infections, increased in ICU admissions, longer durations of admission, and increased mortality among coronavirus disease 2019 (COVID-19) patients with diabetes. We performed a subgroup analysis comparing the outcomes of patients diagnosed with DM (n = 2191) versus patients without DM (n = 8690) on our data from our study based on a nationwide, comparative, retrospective, cohort study among adult, hospitalized COVID-19 patients involving 37 hospital sites from around the Philippines. We determined distribution differences between two independent samples using Mann–Whitney U and t tests. Data on the time to onset of mortality, respiratory failure, intensive care unit (ICU) admission were used to build Kaplan–Meier curves and to compute for hazard ratios (HR). The odds ratios (OR) for longer ventilator dependence, longer ICU stay, and longer hospital stays were computed via multivariate logistic regression. Adjusted hazard ratios (aHR) and ORs (aOR) with 95% CI were calculated. We included a total of 10,881 patients with confirmed COVID-19 infection (2191 have DM while 8690 did not have DM). The median age of the DM cohort was 61, with a female to male ratio of 1:1.25 and more than 50% of the DM population were above 60 years old. The aOR for mortality was significantly higher among those in the DM group by 1.46 (95% CI 1.28–1.68; p < 0.001) as compared to the non-DM group. Similarly, the aOR for respiratory failure was also significantly higher among those in the DM group by 1.67 (95% CI 1.46–1.90). The aOR for developing severe COVID-19 at nadir was significantly higher among those in the DM group by 1.85 (95% CI 1.65–2.07; p < 0.001). The aOR for ICU admission was significantly higher among those in the DM group by 1.80 (95% CI 1.59–2.05) than those in the non-DM group. DM patients had significantly longer duration of ventilator dependence (aOR 1.33, 95% CI 1.08–1.64; p = 0.008) and longer hospital admission (aOR 1.13, 95% CI 1.01–1.26; p = 0.027). The presence of DM among COVID-19 patients significantly increased the risk of mortality, respiratory failure, duration of ventilator dependence, severe/critical COVID-19, ICU admission, and length of hospital stay. © 2021, The Author(s)."data[data$ID == top_abstract[3,1], ]$Abstract## [1] "Background: Malnutrition, as defined by the World Health Organization (WHO), includes undernutrition. In the Philippines, malnutrition is common due to several factors. The nutritional biomarkers can be used as an alternative indicator of dietary intake and nutritional status that can detect deficiencies in support to clinical management of COVID-19 patients. Apart from that, biomarkers are potentially useful for screening, clinical management, and prevention of serious complications of COVID-19 patients. Serum albumin, c-reactive protein (CRP), leukocyte count, lymphocyte count, blood urea nitrogen (BUN) to compute the nutritional prognostic indices (Prognostic nutritional index (PNI) score, BUN/Albumin ratio (BAR) and CRP/Albumin ratio (CAR). Objectives: To compare the nutritional biomarkers of patients with COVID-19 based on case severity and determine the nutritional prognostic indices and associate to patients’ clinical outcome during hospital stay. Methods: A single center, cross-sectional study was performed between June 2021 to August 2021 in a COVID-19 designated referral center in CALABARZON which comprised of 167 patients as part of the study. Clinicodemographic profile including patients' age, sex, co-morbidities, weight, height, laboratory, and serum biomarkers during the first 48 h of admission (serum albumin, leukocyte count, lymphocytes count, CRP, and BUN) were collated wherein the nutritional prognostic indices were computed and analyzed. Clinical outcomes of the patients were based on the patients’ final diagnoses (recovered, length of hospital stay (LOHS), progression of severity and mortality). Results: 167 non-critically ill COVID-19 patients were included in the analysis, of which 52.7% are admitted under the COVID-19 severe group and 47.3% for COVID-19 Mild/Moderate. Mostly are male (53.3%) with an average body mass index (BMI) of 24.26 (SD = 3.52) and have hypertension (55.1%) and diabetes (42.5%). Among the nutritional biomarker, albumin (p = 0.028; p = 0.004), total lymphocyte count (TLC) (p = 0.013; p = 0.005) and BUN (p = 0.001; p=<0.001) were shown to be significantly associated with progression of severity and mortality. Univariate logistic regression analysis showed the following nutritional prognostic score were correlated. (1.) progression of COVID-19 severity: PNI score (OR 0.928, 95% CI 0.886, 0.971, p=<0.001), and BAR value (OR 1.130, 95% CI 1.027, 1.242, p = 0.012); (2.) Mortality: PNI score (OR 0.926, 95% CI 0.878, 0.977, p = 0.005), CAR (OR 1.809, 95% CI 1.243, 2.632, p = 0.002), and BAR (OR 1.180, 95% CI 1.077, 1.292, p=<0.001). The average LOHS of COVID-19 patients was 12 days (SD = 7.72). However, it does not show any significant correlation between any nutritional biomarker, prognostic indices and LOHS. Conclusion: This study demonstrated that deranged level of nutritional biomarkers can affect patient's COVID-19 severity and associated with patient's clinical outcome. Low albumin (≤2.5 g/dL), low level of TLC (≤1500 cells/mm3), elevated BUN (≥7.1 mmol/L) are associated with patient's case severity progression and mortality while low PNI score (<42.49), high BAR value (≥2.8) and CAR value (≥2.04) provided an important nutritional prognostic information and could predict mortality which can be a useful parameter in admission, hence it is recommended to screen all COVID-19 patients to reduce mortality. © 2022 European Society for Clinical Nutrition and Metabolism"data[data$ID == top_abstract[4,1], ]$Abstract## [1] "Background: COVID-19 infection is known to cause a wide array of clinical chronic sequelae, but little is known regarding the long-term cardiac complications. We aim to report echocardiographic follow-up findings and describe the changes in left (LV) and right ventricular (RV) function that occur following acute infection. Methods: Patients enrolled in the World Alliance Societies of Echocardiography-COVID study with acute COVID-19 infection were asked to return for a follow-up transthoracic echocardiogram. Overall, 198 returned at a mean of 129 days of follow-up, of which 153 had paired baseline and follow-up images that were analyzable, including LV volumes, ejection fraction (LVEF), and longitudinal strain (LVLS). Right-sided echocardiographic parameters included RV global longitudinal strain, RV free wall strain, and RV basal diameter. Paired echocardiographic parameters at baseline and follow-up were compared for the entire cohort and for subgroups based on the baseline LV and RV function. Results: For the entire cohort, echocardiographic markers of LV and RV function at follow-up were not significantly different from baseline (all P > .05). Patients with hyperdynamic LVEF at baseline (>70%), had a significant reduction of LVEF at follow-up (74.3% ± 3.1% vs 64.4% ± 8.1%, P < .001), while patients with reduced LVEF at baseline (<50%) had a significant increase (42.5% ± 5.9% vs 49.3% ± 13.4%, P = .02), and those with normal LVEF had no change. Patients with normal LVLS (<−18%) at baseline had a significant reduction of LVLS at follow-up (−21.6% ± 2.6% vs −20.3% ± 4.0%, P = .006), while patients with impaired LVLS at baseline had a significant improvement at follow-up (−14.5% ± 2.9% vs −16.7% ± 5.2%, P < .001). Patients with abnormal RV global longitudinal strain (>−20%) at baseline had significant improvement at follow-up (−15.2% ± 3.4% vs −17.4% ± 4.9%, P = .004). Patients with abnormal RV basal diameter (>4.5 cm) at baseline had significant improvement at follow-up (4.9 ± 0.7 cm vs 4.6 ± 0.6 cm, P = .019). Conclusions: Overall, there were no significant changes over time in the LV and RV function of patients recovering from COVID-19 infection. However, differences were observed according to baseline LV and RV function, which may reflect recovery from the acute myocardial injury occurring in the acutely ill. Left ventricular and RV function tends to improve in those with impaired baseline function, while it tends to decrease in those with hyperdynamic LV or normal RV function. © 2021 American Society of Echocardiography"data[data$ID == top_abstract[5,1], ]$Abstract## [1] "Aim: The aim of this study was to assess the recovery rates of diagnostic cardiac procedure volumes in the Oceania Region, midway through the coronavirus disease 2019 (COVID-19) pandemic. Methods: A survey was performed comparing procedure volumes between March 2019 (pre-pandemic), April 2020 (during first wave of COVID-19 pandemic), and April 2021 (1 year into the COVID-19 pandemic). A total of 31 health care facilities within Oceania that perform cardiac diagnostic procedures were surveyed, including a mixture of metropolitan and regional, hospital and outpatient, public and private sites, as well as teaching and non-teaching hospitals. A comparison was made with 549 centres in 96 countries in the rest of the world (RoW) outside of Oceania. The total number and median percentage change in procedure volume were measured between the three timepoints, compared by test type and by facility. Results: A total of 11,902 cardiac diagnostic procedures were performed in Oceania in April 2021 as compared with 11,835 pre-pandemic in March 2019 and 5,986 in April 2020; whereas, in the RoW, 499,079 procedures were performed in April 2021 compared with 497,615 pre-pandemic in March 2019 and 179,014 in April 2020. There was no significant difference in the median recovery rates for total procedure volumes between Oceania (−6%) and the RoW (−3%) (p=0.81). While there was no statistically significant difference in percentage recovery been functional ischaemia testing and anatomical coronary testing in Oceania as compared with the RoW, there was, however, a suggestion of poorer recovery in anatomical coronary testing in Oceania as compared with the RoW (CT coronary angiography -16% in Oceania vs −1% in RoW, and invasive coronary angiography −20% in Oceania vs −9% in RoW). There was no statistically significant difference in recovery rates in procedure volume between metropolitan vs regional (p=0.44), public vs private (p=0.92), hospital vs outpatient (p=0.79), or teaching vs non-teaching centres (p=0.73). Conclusions: Total cardiology procedure volumes in Oceania normalised 1 year post-pandemic compared to pre-pandemic levels, with no significant difference compared with the RoW and between the different types of health care facilities. © 2024 Australian and New Zealand Society of Cardiac and Thoracic Surgeons (ANZSCTS) and the Cardiac Society of Australia and New Zealand (CSANZ)"Topic 6 is most likely comorbities associated with COVID-19..
colnames(abstract_match)## [1] "ID" "Authors"
## [3] "Author.full.names" "Author.s..ID"
## [5] "Title" "Year"
## [7] "Source.title" "Volume"
## [9] "Issue" "Art..No."
## [11] "Page.start" "Page.end"
## [13] "Page.count" "Cited.by"
## [15] "Link" "Affiliations"
## [17] "Authors.with.affiliations" "Abstract"
## [19] "Author.Keywords" "Funding.Details"
## [21] "Funding.Texts" "Publisher"
## [23] "Document.Type" "Publication.Stage"
## [25] "Open.Access" "Source"
## [27] "EID" "topic1"
## [29] "topic2" "topic3"
## [31] "topic4" "topic5"
## [33] "topic6"# Distribution of Topics Across Journals
# Ensure the topic columns are treated as numeric
abstract_match$topic1 <- as.numeric(abstract_match$topic1)
abstract_match$topic2 <- as.numeric(abstract_match$topic2)
abstract_match$topic3 <- as.numeric(abstract_match$topic3)
abstract_match$topic4 <- as.numeric(abstract_match$topic4)
abstract_match$topic5 <- as.numeric(abstract_match$topic5)
abstract_match$topic6 <- as.numeric(abstract_match$topic6)
topic_summary <- abstract_match %>%
group_by(Source.title) %>%
summarise(
topic1_count = sum(topic1),
topic2_count = sum(topic2),
topic3_count = sum(topic3),
topic4_count = sum(topic4),
topic5_count = sum(topic5),
topic6_count = sum(topic6)
)
head(topic_summary)# Count the number of publications per journal
journal_counts <- abstract_match %>%
group_by(Source.title) %>%
summarise(publication_count = n()) %>%
arrange(desc(publication_count))
# Select the top 10 journals
top_journals <- journal_counts %>%
top_n(10, wt = publication_count) %>%
pull(Source.title)
# Filter the original data frame to include only the top 10 journals
top_journals_data <- abstract_match %>%
filter(Source.title %in% top_journals)
# Summarize topic counts by journal for the top 10 journals
topic_summary_top <- top_journals_data %>%
group_by(Source.title) %>%
summarise(
topic1_count = sum(as.numeric(topic1)),
topic2_count = sum(as.numeric(topic2)),
topic3_count = sum(as.numeric(topic3)),
topic4_count = sum(as.numeric(topic4)),
topic5_count = sum(as.numeric(topic5)),
topic6_count = sum(as.numeric(topic6))
)# Melt the data for easier plotting
topic_summary_melt_top <- melt(topic_summary_top, id.vars = "Source.title",
variable.name = "Topic", value.name = "Count")
# Wrap the labels for better readability
topic_summary_melt_top$Source.title <- str_wrap(topic_summary_melt_top$Source.title, width = 80)
# Rename topics
topic_summary_melt_top$Topic <- factor(topic_summary_melt_top$Topic,
levels = c("topic1_count", "topic2_count", "topic3_count", "topic4_count", "topic5_count", "topic6_count"),
labels = c("Topic 1", "Topic 2", "Topic 3", "Topic 4", "Topic 5", "Topic 6"))
# Create a heatmap
ggplot(topic_summary_melt_top, aes(x = Topic, y = Source.title, fill = Count)) +
geom_tile(color = "white") +
scale_fill_gradient(low = "white", high = "blue") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1, size = 8),
axis.text.y = element_text(size = 8),
plot.title = element_text(hjust = 1)) +
labs(title = "Heatmap of Topic Distribution Across Top 10 Journals",
x = "Topics",
y = "Journals",
fill = "Count")
# Count the number of funders per research
abstract_match$num_funders <- lengths(strsplit(as.character(abstract_match$Funding.Details), ";"))
# Summarize the total number of funders per topic
funding_summary <- abstract_match %>%
summarise(
topic1_funding = sum(as.numeric(topic1) * num_funders),
topic2_funding = sum(as.numeric(topic2) * num_funders),
topic3_funding = sum(as.numeric(topic3) * num_funders),
topic4_funding = sum(as.numeric(topic4) * num_funders),
topic5_funding = sum(as.numeric(topic5) * num_funders),
topic6_funding = sum(as.numeric(topic6) * num_funders)
)
# Melt the data for easier plotting
funding_summary_melt <- melt(funding_summary,
variable.name = "Topic",
value.name = "Total_Num_Funders")## No id variables; using all as measure variables# Rename topics for clarity
funding_summary_melt$Topic <- factor(funding_summary_melt$Topic,
levels = c("topic1_funding", "topic2_funding", "topic3_funding", "topic4_funding", "topic5_funding", "topic6_funding"),
labels = c("Topic 1", "Topic 2", "Topic 3", "Topic 4", "Topic 5", "Topic 6"))
# Create a bar plot
ggplot(funding_summary_melt, aes(x = Topic, y = Total_Num_Funders, fill = Topic)) +
geom_bar(stat = "identity", show.legend = FALSE) +
theme_minimal() +
labs(title = "Total Number of Funders Across Topics",
x = "Topics",
y = "Total Number of Funders",
fill = "Topics") +
theme(plot.title = element_text(hjust = 0.5),
axis.text.x = element_text(angle = 45, hjust = 1))
Mostly funded researches are about vaccine and drug development studies and other comorbidities studies.
# Create a long format DataFrame to count citations for each topic
citation_data <- abstract_match %>%
select(Cited.by, starts_with("topic")) %>%
gather(key = "topic", value = "present", -Cited.by) %>%
filter(present == 1) %>%
group_by(topic) %>%
summarize(total_citations = sum(Cited.by, na.rm = TRUE))
# Rename topics for better readability
citation_data$topic <- gsub("topic", "Topic ", citation_data$topic)
# Create the plot
ggplot(citation_data, aes(x = topic, y = total_citations, fill=topic)) +
geom_bar(stat = "identity", show.legend = FALSE) +
theme_minimal() +
labs(title = "Number of Citations per Topic",
x = "Topic",
y = "Total Citations") +
theme(plot.title = element_text(hjust = 0.5),
axis.text.x = element_text(angle = 45, hjust = 1))
Most cited topics are about vaccine and drug development, followed by studies on comorbidities.
# Save the trained model to a file
saveRDS(lda_model, "final_lda_model.rds")# Save the perplexity to a file
saveRDS(perplexity_df, "perplexity_df")# Save tuning results
saveRDS(lda_ldatuning_result, "lda_ldatuning_result.rds")# Save all variables in the R environment
save(list = ls(), file = "all_variables.RData")# Save all output
sink("output.txt")
sink()