If I Had a Text File, I’d Hack Regexes in the Morning
Want to share your content on R-bloggers? click here if you have a blog, or here if you don't.
Yesterday the topic of academic citation counts came up, so I decided that I should write up some tools for exploring cite counts. The first thing I did was to build a cheap screenscraper in Ruby for pulling citation count information from Google scholar. You’ll see the ugly hack I produced below.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 | module CitationTools
require 'rubygems'
require 'open-uri'
def get_ten_most_cited_works_for_author(author_name)
# First, let's clean up the author's name before using it in a URL.
escaped_author_name = author_name.gsub(/s+/, '+')
# Let's create a variable we'll place the Google Scholar HTML in.
page_content = nil
# Let's figure out the right URL for Google Scholar.
url = "http://scholar.google.com/scholar?q=#{escaped_author_name}"
# Let's access that URL using open-uri and get the HTML from the page.
open(url) do |page|
page_content = page.read()
end
# Let's scan the HTML for the names of this author's works.
work_titles = page_content.scan(/<p class=g>.*?>([^<]+)(?:</a></span>)?(?:(?:<font size=-1>)|(?:s+-s+<span class=a>)|(?:s+-s+<a class=fl))/)
# Let's scan the HTML for the citation counts for each work.
cite_counts = page_content.scan(/Cited by (d+)/)
# Let's set aside an array of hashes to store all of this data.
works = []
# As long as we have the same number of titles and counts, we're good.
if work_titles.size == cite_counts.size
work_titles.each_with_index do |title, index|
works << {:title => title, :citation_count => cite_counts[index]}
end
return works
else
puts "Failed to process HTML for #{author_name}"
return nil
end
end
end |
With that in hand, I wrote a simple wrapper to pull information for a list of authors you store in a file called authors.txt from Google Scholar. The wrapper then prints a CSV file to STDOUT that can be redirected to a file for later analysis.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 | # Let's include a mix-in with some methods for parsing Google scholar data.
require 'CitationTools'
include CitationTools
# Let's pick a haphazard sample of authors.
authors = File.new('authors.txt', 'r').readlines.map {|line| line.chomp}
# Let's add a header line to our output.
puts '"Author","Work","Citations"'
# And then let's iterate over those authors.
authors.each do |author_name|
cited_work_data = get_ten_most_cited_works_for_author(author_name)
if cited_work_data.nil?
print "Skipping #{author_name}"
end
cited_work_data.each do |cited_work|
puts ""#{author_name}","#{cited_work[:title]}",#{cited_work[:citation_count]}"
end
end |
Then I coded up a simple barplot in R to give you a sense of the citation count for the first few authors that came to mind. The result is below.

Now I think the goal should be to put these tools to a good use.
R-bloggers.com offers daily e-mail updates about R news and tutorials about learning R and many other topics. Click here if you're looking to post or find an R/data-science job.
Want to share your content on R-bloggers? click here if you have a blog, or here if you don't.