/// <summary>
/// Computes cosine similarities between documents via the Vector Space Model.
/// </summary>
/// <param name="source">Source artifacts</param>
/// <param name="target">Target artifacts</param>
/// <param name="weight">Weighting scheme</param>
/// <returns>Similarity matrix</returns>
public static TLSimilarityMatrix Compute(TLArtifactsCollection source, TLArtifactsCollection target, VSMWeightEnum weight)
{
switch (weight)
{
case VSMWeightEnum.TFIDF:
return(SimilarityUtil.ComputeCosine(WeightUtil.ComputeTFIDF(new TermDocumentMatrix(source, target)), source.Keys, target.Keys));
case VSMWeightEnum.BooleanQueriesAndTFIDFCorpus:
return(SimilarityUtil.ComputeCosine(WeightUtil.ComputeBinaryTF(new TermDocumentMatrix(source)), WeightUtil.ComputeTFIDF(new TermDocumentMatrix(target))));
case VSMWeightEnum.NoWeight:
return(SimilarityUtil.ComputeCosine(new TermDocumentMatrix(source, target), source.Keys, target.Keys));
default:
throw new NotImplementedException("Unknown weighting scheme \"" + weight + "\"");
}
}
/// <summary>
/// Smoothing filter from ICPC'11 paper "Improving IR-based Traceability Recovery Using Smoothing Filters"
/// </summary>
/// <param name="matrix">Term-by-document matrix</param>
/// <returns>Smoothed artifacts</returns>
public static TermDocumentMatrix Compute(TermDocumentMatrix matrix)
{
double[] avg = WeightUtil.ComputeAverageVector(matrix);
if (avg.Length != matrix.NumTerms)
{
throw new ArgumentException("Average vector does not have the correct number of terms.");
}
for (int i = 0; i < matrix.NumDocs; i++)
{
for (int j = 0; j < matrix.NumTerms; j++)
{
matrix[i, j] -= avg[j];
if (matrix[i, j] < 0.0)
{
matrix[i, j] = 0.0;
}
}
}
return(matrix);
}
/// <summary>
/// Smoothing filter from ICPC'11 paper "Improving IR-based Traceability Recovery Using Smoothing Filters"
/// </summary>
/// <param name="matrix">Term-by-document matrix</param>
/// <param name="IDs">Collection of document ids to smooth.</param>
/// <returns>Smoothed artifacts</returns>
public static TermDocumentMatrix Compute(TermDocumentMatrix matrix, IEnumerable <string> IDs)
{
double[] avg = WeightUtil.ComputeAverageVector(matrix, IDs);
if (avg.Length != matrix.NumTerms)
{
throw new ArgumentException("Average vector does not have the correct number of terms.");
}
foreach (string docID in IDs)
{
int i = matrix.GetDocumentIndex(docID);
for (int j = 0; j < matrix.NumTerms; j++)
{
matrix[i, j] -= avg[j];
if (matrix[i, j] < 0.0)
{
matrix[i, j] = 0.0;
}
}
}
return(matrix);
}
