private static double ActualBlockDistanceSimilarity(Collection <string> firstTokens, Collection <string> secondTokens)
{
TokeniserUtilities <string> _tokenUtilities = new TokeniserUtilities <string>();
Collection <string> collection = _tokenUtilities.CreateMergedList(firstTokens, secondTokens);
int num = 0;
foreach (string str in collection)
{
int num2 = 0;
int num3 = 0;
if (firstTokens.Contains(str))
{
num2++;
}
if (secondTokens.Contains(str))
{
num3++;
}
if (num2 > num3)
{
num += num2 - num3;
}
else
{
num += num3 - num2;
}
}
return(num);
}
public static double CosineSimilarity2(this string source, string target)
{
TokeniserUtilities <string> _tokenUtilities = new TokeniserUtilities <string>();
if (source != null && target != null && _tokenUtilities.CreateMergedSet(_tokeniser.Tokenize(source), _tokeniser.Tokenize(target)).Count > 0)
{
return(_tokenUtilities.CommonSetTerms() / (Math.Pow(_tokenUtilities.FirstSetTokenCount, 0.5) * Math.Pow(_tokenUtilities.SecondSetTokenCount, 0.5)));
}
return(0.0);
}
public static Collection <string> TokenizeToSet(string word)
{
TokeniserUtilities <string> _tokenUtilities = new TokeniserUtilities <string>();
if (word != null)
{
return(_tokenUtilities.CreateSet(Tokenize(word)));
}
return(null);
}
public static double OverlapCoefficientSimilarity(this string firstWord, string secondWord)
{
TokeniserUtilities <string> _tokenUtilities = new TokeniserUtilities <string>();
if (firstWord != null && secondWord != null)
{
_tokenUtilities.CreateMergedSet(_tokeniser.Tokenize(firstWord), _tokeniser.Tokenize(secondWord));
return(_tokenUtilities.CommonSetTerms() / (double)Math.Min(_tokenUtilities.FirstSetTokenCount, _tokenUtilities.SecondSetTokenCount));
}
return(DefaultMismatchScore);
}
public static double EuclideanSimilarity(this string source, string target)
{
TokeniserUtilities <int> _tokenUtilities = new TokeniserUtilities <int>();
if (source != null && target != null)
{
double unnormalisedSimilarity = source.UnnormalisedEuclideanSimilarity(target);
double num2 = Math.Sqrt(_tokenUtilities.FirstTokenCount + _tokenUtilities.SecondTokenCount);
return((num2 - unnormalisedSimilarity) / num2);
}
return(DefaultMismatchScore);
}
//! NOT WORKING
public static double MatchingCoefficientSimilarity(this string source, string target)
{
TokeniserUtilities <string> _tokenUtilities = new TokeniserUtilities <string>();
if (source != null && target != null)
{
double unnormalisedSimilarity = source.UnnormalisedJaroSimilarity(target);
int num2 = Math.Max(_tokenUtilities.FirstTokenCount, _tokenUtilities.SecondTokenCount);
return(unnormalisedSimilarity / num2);
}
return(0);
}
//private static readonly ITokeniser _tokeniser;
//private static readonly TokeniserUtilities<string> => _tokenUtilities;
//private static readonly _tokeniser => TokeniserWhitespace;
private static double ActualMatchingCoefficientSimilarity(Collection <string> firstTokens, Collection <string> secondTokens)
{
TokeniserUtilities <string> _tokenUtilities = new TokeniserUtilities <string>();
_tokenUtilities.CreateMergedList(firstTokens, secondTokens);
int num = 0;
foreach (string str in firstTokens)
{
if (secondTokens.Contains(str))
{
num++;
}
}
return(num);
}
//private readonly double _estimatedTimingConstant;
//private readonly ITokeniser _tokeniser;
//private readonly TokeniserUtilities<string> _tokenUtilities;
//public EuclideanDistance() : this(new TokeniserWhitespace())
//{
//}
//public EuclideanDistance(ITokeniser tokeniserToUse)
//{
// _estimatedTimingConstant = 7.4457137088757008E-05;
// _tokeniser = tokeniserToUse;
// _tokenUtilities = new TokeniserUtilities<string>();
//}
private static double GetActualDistance(Collection <string> firstTokens, Collection <string> secondTokens)
{
TokeniserUtilities <string> _tokenUtilities = new TokeniserUtilities <string>();
Collection <string> collection = _tokenUtilities.CreateMergedList(firstTokens, secondTokens);
int num = 0;
foreach (string str in collection)
{
int num2 = 0;
int num3 = 0;
if (firstTokens.Contains(str))
{
num2++;
}
if (secondTokens.Contains(str))
{
num3++;
}
num += (num2 - num3) * (num2 - num3);
}
return(Math.Sqrt(num));
}
