/// <summary>
/// Compute jaccard index: |A inter B| / |A union B|.
/// </summary>
/// <param name="s1">First string</param>
/// <param name="s2">Second string</param>
/// <returns>Similarity</returns>
public double Similarity(string s1, string s2)
{
KShingling ks = new KShingling(k);
int[] profile1 = ks.GetArrayProfile(s1);
int[] profile2 = ks.GetArrayProfile(s2);
int length = Math.Max(profile1.Length, profile2.Length);
profile1 = profile1.WithPadding(length);
profile2 = profile2.WithPadding(length);
int inter = 0;
int union = 0;
for (int i = 0; i < length; i++)
{
if (profile1[i] > 0 || profile2[i] > 0)
{
union++;
if (profile1[i] > 0 && profile2[i] > 0)
{
inter++;
}
}
}
return(1.0 * inter / union);
}
public double Similarity(string s1, string s2)
{
if (s1.Length < k || s2.Length < k)
{
return(0);
}
KShingling ks = new KShingling(k);
int[] profile1 = ks.GetArrayProfile(s1);
int[] profile2 = ks.GetArrayProfile(s2);
return(DotProduct(profile1, profile2) / (Norm(profile1) * Norm(profile2)));
}
/// <summary>
/// The distance between two strings is defined as the L1 norm of the
/// difference of their profiles (the number of occurence of each k-shingle).
/// </summary>
/// <param name="s1">The first string</param>
/// <param name="s2">The second string</param>
/// <returns></returns>
public double Distance(string s1, string s2)
{
var ks = new KShingling(k);
int[] profile1 = ks.GetArrayProfile(s1);
int[] profile2 = ks.GetArrayProfile(s2);
int length = Math.Max(profile1.Length, profile2.Length);
//profile1 = java.util.Arrays.copyOf(profile1, length);
//profile2 = java.util.Arrays.copyOf(profile2, length);
profile1 = profile1.WithPadding(length);
profile2 = profile2.WithPadding(length);
int d = 0;
for (int i = 0; i < length; i++)
{
d += Math.Abs(profile1[i] - profile2[i]);
}
return(d);
}
/// <summary>
/// Similarity is computed as 2 * |A inter B| / (|A| + |B|).
/// </summary>
/// <param name="s1">The first string</param>
/// <param name="s2">The second string</param>
/// <returns></returns>
public double Similarity(string s1, string s2)
{
KShingling ks = new KShingling(k);
int[] profile1 = ks.GetArrayProfile(s1);
int[] profile2 = ks.GetArrayProfile(s2);
int length = Math.Max(profile1.Length, profile2.Length);
//profile1 = java.util.Arrays.copyOf(profile1, length);
//profile2 = java.util.Arrays.copyOf(profile2, length);
profile1 = profile1.WithPadding(length);
profile2 = profile2.WithPadding(length);
int inter = 0;
int sum = 0;
for (int i = 0; i < length; i++)
{
if (profile1[i] > 0 && profile2[i] > 0)
{
inter++;
}
if (profile1[i] > 0)
{
sum++;
}
if (profile2[i] > 0)
{
sum++;
}
}
return(2.0 * inter / sum);
}
public StringSet(SparseBooleanVector vector, KShingling ks)
{
_vector = vector;
_ks = ks;
}
public StringProfile(SparseIntegerVector vector, KShingling ks)
{
Vector = vector;
_ks = ks;
}
