internal static EnglishWord CreateForTest(string text)
{
var word = new EnglishWord();
word.Create(text);
return(word);
}
private static List <string> GetVocabulary(string[] docs, out List <List <string> > stemmedDocs,
int vocabularyThreshold)
{
var wordCountList = new Dictionary <string, int>();
stemmedDocs = new List <List <string> >();
var docIndex = 0;
foreach (var doc in docs)
{
var stemmedDoc = new List <string>();
docIndex++;
if (docIndex % 100 == 0)
{
Console.WriteLine("Processing " + docIndex + "/" + docs.Length);
}
var parts2 = WordsProcessing(doc);
var words = new List <string>();
foreach (var part in parts2)
{
var stripped = Regex.Replace(part, "[^a-zA-Z0-9]", "");
if (StopWords.StopWordsList.Contains(stripped.ToLower()))
{
continue;
}
try
{
var english = new EnglishWord(stripped);
var stem = english.Stem;
words.Add(stem);
if (stem.Length > 0)
{
if (wordCountList.ContainsKey(stem))
{
wordCountList[stem]++;
}
else
{
wordCountList.Add(stem, 0);
}
stemmedDoc.Add(stem);
}
}
catch (Exception e)
{
Console.WriteLine(e);
throw;
}
}
stemmedDocs.Add(stemmedDoc);
}
var vocabList = wordCountList.Where(w => w.Value >= vocabularyThreshold);
return(vocabList.Select(item => item.Key).ToList());
}
