/// <summary>
/// Calculates the Dice Coefficient for two given strings.
/// </summary>
/// <remarks>This is a similarity metric based on bigrams that is calculated as follows:
/// * D is Dice coefficient
/// * SB is Shared Bigrams
/// * TBg1 is total number of bigrams in Qgram1
/// * TBg2 is total number of bigrams in Qgram2
/// * D = (2SB)/(TBg1+TBg2)
/// A good Dice Coefficient value would be a value greater than 0.33</remarks>
/// <param name="pStr1">First string to compare</param>
/// <param name="pStr2">Second string to compare</param>
/// <returns>Dice Coefficient</returns>
internal static double DiceCoefficient(String pStr1, String pStr2)
{
// faulty input parameters
if (string.IsNullOrEmpty(pStr1) || string.IsNullOrEmpty(pStr1))
{
return(0.0);
}
var bigram1 = Bigram.Parse(pStr1);
var bigram2 = Bigram.Parse(pStr2);
// calculate number of shared bigrams
int sharedBigrams = 0;
foreach (var s1 in bigram1)
{
foreach (var s2 in bigram2)
{
if (s1.Equals(s2))
{
sharedBigrams++;
}
}
}
// calculate dice coefficient
double dice = Convert.ToDouble(sharedBigrams * 2) / Convert.ToDouble(bigram1.Length + bigram2.Length);
return(dice);
}
/// <summary>
/// Calculates the Dice Coefficient for two given strings.
/// </summary>
/// <remarks>This is a similarity metric based on bigrams that is calculated as follows:
/// * D is Dice coefficient
/// * SB is Shared Bigrams
/// * TBg1 is total number of bigrams in Qgram1
/// * TBg2 is total number of bigrams in Qgram2
/// * D = (2SB)/(TBg1+TBg2)
/// A good Dice Coefficient value would be a value greater than 0.33</remarks>
/// <param name="pStr1">First string to compare</param>
/// <param name="pStr2">Second string to compare</param>
/// <returns>Dice Coefficient</returns>
internal static double DiceCoefficient(string pStr1, string pStr2)
{
// faulty input parameters
if (string.IsNullOrEmpty(pStr1) || string.IsNullOrEmpty(pStr1))
{
return(0.0);
}
var bigram1 = Bigram.Parse(pStr1);
var bigram2 = Bigram.Parse(pStr2);
// calculate number of shared bigrams
var sharedBigrams = bigram1.Sum(s1 => bigram2.Count(s1.Equals));
// calculate dice coefficient
return(Convert.ToDouble(sharedBigrams * 2) / Convert.ToDouble(bigram1.Length + bigram2.Length));
}