static void Main(string[] args) { var src = @"F:\ToSummarize.txt"; string str = System.IO.File.ReadAllText(src); string[] Lines = str.Split(new string[] { "\r\n", "\n", "." }, StringSplitOptions.RemoveEmptyEntries); WordHandle stopword = new WordHandle(); TFIDF tf = new TFIDF(Lines); using (System.IO.StreamWriter file = new System.IO.StreamWriter(@"F:\Summary.txt")) { foreach (var item in tf.DocResult) { file.WriteLine(Lines[item]); } } }
public string[] Partition(string input) { Regex r = new Regex("([ \\t{}():;. \n])"); input = input.ToLower(); String[] tokens = r.Split(input); ArrayList fltr = new ArrayList(); for (int i = 0; i < tokens.Length; i++) { MatchCollection mc = r.Matches(tokens[i]); if (mc.Count <= 0 && tokens[i].Trim().Length > 0 && !WordHandle.IsStopword(tokens[i])) { fltr.Add(tokens[i]); } } return(ArrLstToArr(fltr)); }