mining_citation_network.Rmd
itle: “Mining the citation network” |
utput: rmarkdown::html_vignette |
ignette: > |
% |
% |
% |
Article metadata was extracted from Elsevier API using each article’s DOI. The metadata contains the citing literature and the author-defined keywords. The citing literature is used to build a citation network. Keywords are used to augment the labels of country location derived from the human-reading. This is done by looking for regular expression of the name of the 23 target countries.
The following chunk extracts the cited and citing literature from the webscrapped metadata. This chunk is not run as it requires some licensed information.
citingFile <- "citing_df.Rds"
if (!file.exists(citingFile)){
# get wosFullResult
wosFullResult <- readRDS("./data/wosFullResult.Rds")
# get source id
source_ids <- lapply(wosFullResult, get_sourceID)
# get citing cited
citing_dfs <- lapply(wosFullResult, get_citing)
citing_dfs <- na.omit(citing_dfs)
citing_df <- do.call(rbind, citing_dfs)
citing_df <- citing_df[which(citing_df$cited %in% source_ids), ]
saveRDS(citing_df, citingFile)
saveRDS(source_ids, "source_ids.Rds")
} else {
citing_df <- readRDS(citingFile)
}