public void ComputeProbability(List <string> tokens)
{
var mleDictionary = new Dictionary <string, double>();
var lsDictionary = new Dictionary <string, double>();
var adDictionary = new Dictionary <string, double>();
for (var i = 0; i < tokens.Count; i++)
{
var mleProbability = EstimationMethods.MaximumLikelihoodEstimate(tokens[i], tokens[i + 1], tokens);
mleDictionary.Add(tokens[i] + tokens[i + 1], mleProbability);
var lsProbability = EstimationMethods.LaplaceSmoothing(tokens[i], tokens[i + 1], tokens);
lsDictionary.Add(tokens[i] + tokens[i + 1], lsProbability);
var adProbability = EstimationMethods.AbsoluteDiscounting(tokens[i], tokens[i + 1], tokens);
adDictionary.Add(tokens[i] + tokens[i + 1], adProbability);
}
}
static void Main(string[] args)
{
Console.WriteLine("Reading File...");
//var text = Preprocessing.ReadFile(@"C:\src\nlp-language-models\Homework1\train.txt");
//var corpus = Preprocessing.DumbSentenceSegmentor(text);
//var tokenizedCorpus = Preprocessing.DumbTokenizer(corpus);
//tokenizedCorpus = new List<string>(tokenizedCorpus.RemoveAll(string.IsNullOrEmpty));
var test = "A fat dog is a fat dog only when other fat dogs say he is a fat dog";
var tokens = Preprocessing.DumbTokenizer(test);
for (var i = 0; i < tokens.Count - 1; i++)
{
var mle = EstimationMethods.MaximumLikelihoodEstimate(tokens[i], tokens[i + 1], tokens);
var ls = EstimationMethods.LaplaceSmoothing(tokens[i], tokens[i + 1], tokens);
var ads = EstimationMethods.AbsoluteDiscounting(tokens[i], tokens[i + 1], tokens);
Console.WriteLine("P(" + tokens[i] + " | " + tokens[i + 1] + ")" + " = " + mle.ToString());
}
Console.ReadKey();
}