Word Frequencies
March 10, 2009
Write a program that takes a filename and a parameter n and prints the n most common words in the file, and the count of their occurrences, in descending order.
- Comment
- Reblog
-
Subscribe
Subscribed
Already have a WordPress.com account? Log in now.
Programming Praxis 
perl -ne ‘while (/([a-z]+)/gi) {$words{$1}++} END{ map { print “$_ $words{$_}\n”} sort {$words{$b} $words{$a}} keys %words}’
A straightforward (though probably not very efficient) Haskell solution:
import System.Environment
import Control.Arrow
import Data.List
import Data.Ord
main = do [n, fileName] <- getArgs
content String -> [(String, Int)]
findMostCommonWords n = take n . reverse . sortBy (comparing snd) . map (head &&& length) . group . sort . words
Sigh… gotta love forms that don’t escape html characters. Let’s try that again.
import System.Environment
import Control.Arrow
import Data.List
import Data.Ord
main = do [n, fileName] <- getArgs
content <- readFile fileName
mapM_ print $ findMostCommonWords (read n) content
findMostCommonWords :: Int -> String -> [(String, Int)]
findMostCommonWords n = take n . reverse . sortBy (comparing snd) . map (head &&& length) . group . sort . words
Interesting challenge, here’s my solution in python.
http://pastebin.com/f1ac76000
Action CountWords = (filename, top) =>
{
foreach (var kv in
File.ReadAllText(filename)
.Split()
.GroupBy(w => w,
(w, c) => new
{ Word = w, Count = c.Count() })
.OrderByDescending(a => a.Count)
.Take(top))
Console.WriteLine(kv.Word + ” – ” + kv.Count);
};
http://pastebin.com/fbf06089
[…] Dictionaries are a common data type, which we have used in several exercises (Mark V. Shaney, Word Frequencies, Dodgson’s Doublets, Anagrams). Hash tables are often used as the underlying implementation […]
import qualified Data.Map as M import Data.List import Data.Char import Data.Function import System.Environment isWord :: String -> Bool isWord s = length s > 2 buildFrequencyMap :: String -> M.Map String Int buildFrequencyMap content = foldl' insertWord M.empty mots where insertWord m word = M.insertWith' (+) word 1 m mots = filter isWord $ map (map toLower . takeWhile isLetter) (words content) sortByFrequency :: M.Map String Int -> [(String, Int)] sortByFrequency m = reverse $ sortBy (compare `on` snd) (M.toList m) displayWords :: Int -> [(String, Int)] -> IO () displayWords n frequencies = mapM_ (putStrLn . format) (take n frequencies) where format (w, x) = (show x) ++ " -> " ++ w main = do args <- getArgs case args of [n, f] -> displayWords (read n) . sortByFrequency . buildFrequencyMap =<< readFile f _ -> error "Deux arguments requis"Here it is in ruby (commented so I don’t have to do it elsewhere) …
# Write a program that takes a filename and a parameter n and prints the n most # common words in the file, and the count of their occurrences, in descending # order. # Require for the command line options processor. require 'getoptlong' # Set up the command line options opts = GetoptLong.new( ["--number", "-n", GetoptLong::REQUIRED_ARGUMENT], ["--verbose", "-v", GetoptLong::NO_ARGUMENT] ) # Set the default values for the options number = 10 $verbose = false # Parse the command line options. If we find one we don't recognize # an exception will be thrown and we'll rescue with a message. begin opts.each do | opt, arg| case opt when "--number" number = arg.to_i when "--verbose" $verbose = true end end rescue puts "Illegal command line option." exit end # Create the word frequency hash. word_freq = Hash.new(0) # Loop through the remaining arguments which we'll assume are # file names. ARGV.each do |file_name| File.open(file_name) do | file | # Loop through each line of the file while line = file.gets # Split on non-words or digits. This will throw out punctuation and # numbers. If numbers are to be included, remove the \d from the regex. words = line.split(/[\W\d]/) words.each do |word| # For some reason we're getting empty strings (probably a bad regex above) so # just toss them out here. word_freq[word] += 1 if word != "" end end end end # Print out the n most frequent words. First we'll sort which normally # will sort on the key, we'll pass a block so that we can sort on the # value. Then we'll reverse to get the largest values first, and finally # we'll pull out the first n. word_freq.sort{|a,b| a[1]<=>b[1]}.reverse[0, number].each do |v| puts "#{v[0]} #{v[1]}" endClojure library has an handy frequencies function.
(defn word-frequencies [filepath n] (let [wrds (.split (.toLowerCase (slurp filepath)) "[^a-zA-Z]+")] (take n (reverse (sort-by second (frequencies wrds))))))Hello guys,
Check my solution in Python. I went a bit further I cleaned the source from punctuation characters.
It brings more relevant output.
from operator import itemgetter from string import punctuation def countWords(filename, n): all_words = (word.strip(punctuation).lower() for line in open(filename) for word in line.split()) words = {} for word in all_words: words[word] = words.get(word, 0) + 1 #sort by number and slice sort_words = sorted(words.iteritems(), key=itemgetter(1), reverse=True)[:int(n)] for index, words in enumerate(sort_words): print "%d. %s - %d" % (index + 1, words[0], words[1]) if __name__ == "__main__": from optparse import OptionParser parser = OptionParser() parser.add_option("-f", "--file", dest="filename", help="Select a FILE", metavar="FILE") parser.add_option("-n", "--num", help="Number of words to display") (options, args) = parser.parse_args() countWords(options.filename, options.num)https://github.com/ftt/programming-praxis/blob/master/20090310-word-frequencies/word-frequencies.py
#! /usr/bin/env python def word_frequencies(filename): g = open(filename, "r") f = g.read() g.close() words = {} f = ''.join((c.lower() if c in "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ" else " ") for c in f) for w in f.split(): if w == '': continue if w in words: words[w] += 1 else: words[w] = 1 return words if __name__ == "__main__": from sys import argv words = word_frequencies(argv[1]) words = [(w, words[w]) for w in words] words.sort(key=lambda w: w[1]) words.reverse() for i in xrange(min(len(words), int(argv[2]))): print words[i][0], words[i][1]# In Ruby
text = File.read(ARGV[0])
n = ARGV[1].to_i
puts text
.scan(/\w+/)
.group_by(&:itself)
.map { |k, v| [k, v.count] }
.sort_by { |_, v| -v }
.take(n)
.map { |k, v| “#{k}: #{v}” }