public static void Main(string[] args)
{
// тренировка модели
NER_Training ner_training_obj = new NER_Training();
ner_training_obj.process_corpus();
ner_training_obj.Compute_frequencies();
ner_training_obj.Print_The_Table();
// использование натренированной модели
NER_testing ner_testing_obj = new NER_testing();
ner_testing_obj.Read_Input();
string[] x = ner_testing_obj.input_text.Split(' ');
List <string> words = new List <string>();
foreach (string item in x)
{
words.Add(item);
}
List <string> tmp_words = new List <string>();
foreach (string word in words)
{
tmp_words.Clear();
foreach (string item in words)
{
tmp_words.Add(item);
}
tmp_words.Remove(word);
if (ner_testing_obj.is_Named_Entity(word, tmp_words.ToArray(), ner_training_obj))
{
ner_testing_obj.output_text += word + "<NE>" + " ";
}
else
{
ner_testing_obj.output_text += word + " ";
};
}
ner_testing_obj.output_text.Trim();
Console.WriteLine(ner_testing_obj.output_text);
Console.Read();
}
public bool is_Named_Entity(string word_to_be_checked, string[] context_words, NER_Training table)
{
float score = 0;
foreach (string word in context_words)
{
if (table.factors.Contains(word))
{
score += table.frequencies[table.factors.IndexOf(word)];
}
}
if (score > GlobalVars.THRESHOLD_SCORE)
{
return(true);
}
else
{
return(false);
}
}