public static void sortAndPrintMap(Map<String, Double> map) throws FileNotFoundException {
// pr = new PrintStream(new FileOutputStream("/semplest/lluis/keywordExp/wordmap.txt"));
pr = System.out;
ValueComparator bvc = new ValueComparator(map);
TreeMap<String, Double> sorted = new TreeMap<String, Double>(bvc);
sorted.putAll(map);
for (String key : sorted.keySet()) {
pr.println(key + " : " + map.get(key));
}
System.out.println("Number of words : " + map.size());
}
public static Map<String, Double> getWordCount(Map<String, List<String>> data) {
Map<String, Double> wordMap = new HashMap<String, Double>();
for (String url : data.keySet()) {
for (String line : data.get(url)) {
line = line.toLowerCase();
line = line.replaceAll("\\p{Punct}", " ");
line = line.replaceAll("[0-9]+", "");
String[] words = line.split("\\s+");
for (String word : words) {
if (wordMap.containsKey(word)) {
wordMap.put(word, wordMap.get(word) + 1);
} else {
wordMap.put(word, new Double(0.0));
}
}
}
}
return wordMap;
}
