public void GetTopNLinks()
{
TLLinksList list = TLSimilarityMatrixUtil.GetTopNLinks(sims, 4);
#if Verbose
Console.WriteLine("TLSimilarityMatrixUtilTest.GetTopNLinks()");
for (int i = 0; i < list.Count; i++)
{
Console.WriteLine("{0}\t{1}\t{2}",
list[i].SourceArtifactId,
list[i].TargetArtifactId,
list[i].Score
);
}
#endif
Assert.AreEqual(4, list.Count);
TLLinksList expected = new TLLinksList();
expected.Add(new TLSingleLink("A", "B*", 10));
expected.Add(new TLSingleLink("A", "E", 9));
expected.Add(new TLSingleLink("A", "F", 8));
expected.Add(new TLSingleLink("A", "C*", 7));
for (int i = 0; i < expected.Count; i++)
{
Assert.AreEqual(expected[i], list[i]);
}
}
/// <summary>
/// Computes cosine similarities between two TermDocumentMatrices.
/// Cosine similarity is defined as (dot product) / (length * length)
/// </summary>
/// <param name="m1">Binary document matrix</param>
/// <param name="m2">tf-idf weighted document matrix</param>
/// <returns>Similarity matrix</returns>
public static TLSimilarityMatrix ComputeCosine(TermDocumentMatrix m1, TermDocumentMatrix m2)
{
TLSimilarityMatrix sims = new TLSimilarityMatrix();
List <TermDocumentMatrix> matrices = TermDocumentMatrix.Equalize(m1, m2);
for (int i = 0; i < m1.NumDocs; i++)
{
TLLinksList links = new TLLinksList();
for (int j = 0; j < m2.NumDocs; j++)
{
double lengthProduct = ComputeLength(matrices[0].GetDocument(i)) * ComputeLength(matrices[1].GetDocument(j));
if (lengthProduct == 0.0)
{
links.Add(new TLSingleLink(m1.GetDocumentName(i), m2.GetDocumentName(j), 0.0));
}
else
{
links.Add(new TLSingleLink(m1.GetDocumentName(i), m2.GetDocumentName(j), ComputeDotProduct(matrices[0].GetDocument(i), matrices[1].GetDocument(j)) / lengthProduct));
}
}
links.Sort();
foreach (TLSingleLink link in links)
{
sims.AddLink(link.SourceArtifactId, link.TargetArtifactId, link.Score);
}
}
return(sims);
}
/// <summary>
/// Computes cosine similarities between two TermDocumentMatrices.
/// Cosine similarity is defined as (dot product) / (length * length)
/// </summary>
/// <param name="m1">Binary document matrix</param>
/// <param name="m2">tf-idf weighted document matrix</param>
/// <returns>Similarity matrix</returns>
public static TLSimilarityMatrix ComputeCosine(TermDocumentMatrix m1, TermDocumentMatrix m2)
{
TLSimilarityMatrix sims = new TLSimilarityMatrix();
List<TermDocumentMatrix> matrices = TermDocumentMatrix.Equalize(m1, m2);
for (int i = 0; i < m1.NumDocs; i++)
{
TLLinksList links = new TLLinksList();
for (int j = 0; j < m2.NumDocs; j++)
{
double lengthProduct = ComputeLength(matrices[0].GetDocument(i)) * ComputeLength(matrices[1].GetDocument(j));
if (lengthProduct == 0.0)
{
links.Add(new TLSingleLink(m1.GetDocumentName(i), m2.GetDocumentName(j), 0.0));
}
else
{
links.Add(new TLSingleLink(m1.GetDocumentName(i), m2.GetDocumentName(j), ComputeDotProduct(matrices[0].GetDocument(i), matrices[1].GetDocument(j)) / lengthProduct));
}
}
links.Sort();
foreach (TLSingleLink link in links)
{
sims.AddLink(link.SourceArtifactId, link.TargetArtifactId, link.Score);
}
}
return sims;
}
public void GetLinksAtRecall100()
{
TLLinksList list = TLSimilarityMatrixUtil.GetLinksAtRecall(sims, oracle, 1.0);
#if Verbose
Console.WriteLine("TLSimilarityMatrixUtilTest.GetLinksAtRecall100()");
for (int i = 0; i < list.Count; i++)
{
Console.WriteLine("{0}\t{1}\t{2}",
list[i].SourceArtifactId,
list[i].TargetArtifactId,
list[i].Score
);
}
#endif
Assert.AreEqual(9, list.Count);
TLLinksList expected = new TLLinksList();
expected.Add(new TLSingleLink("A", "B*", 10));
expected.Add(new TLSingleLink("A", "E", 9));
expected.Add(new TLSingleLink("A", "F", 8));
expected.Add(new TLSingleLink("A", "C*", 7));
expected.Add(new TLSingleLink("A", "G", 6));
expected.Add(new TLSingleLink("A", "H", 5));
expected.Add(new TLSingleLink("A", "I", 4));
expected.Add(new TLSingleLink("A", "J", 3));
expected.Add(new TLSingleLink("A", "D*", 2));
for (int i = 0; i < expected.Count; i++)
{
Assert.AreEqual(expected[i], list[i]);
}
}
/// <summary>
/// Returns links for the desired recall level.
/// </summary>
/// <param name="matrix">Candidate matrix</param>
/// <param name="answerMatrix">Answer matrix</param>
/// <param name="level">Desired recall level</param>
/// <returns>List of links at desired recall</returns>
public static TLLinksList GetLinksAtRecall(TLSimilarityMatrix matrix, TLSimilarityMatrix answerMatrix, double level)
{
if (level <= 0.0 || level > 1.0)
{
throw new DevelopmentKitException("Recall level must be between 0 and 1.");
}
double totalCorrect = answerMatrix.Count * level;
int numCorrect = 0;
TLLinksList links = matrix.AllLinks;
links.Sort();
TLLinksList newLinks = new TLLinksList();
while (links.Count > 0 && numCorrect < totalCorrect)
{
TLSingleLink link = links[0];
if (answerMatrix.IsLinkAboveThreshold(link.SourceArtifactId, link.TargetArtifactId))
{
numCorrect++;
}
newLinks.Add(link);
links.RemoveAt(0);
}
return(newLinks);
}
public void GetLinksAboveThresholdDefault()
{
sims.Threshold = 4;
TLLinksList list = TLSimilarityMatrixUtil.GetLinksAboveThreshold(sims);
list.Sort();
#if Verbose
Console.WriteLine("TLSimilarityMatrixUtilTest.GetLinksAboveThresholdDefault()");
for (int i = 0; i < list.Count; i++)
{
Console.WriteLine("{0}\t{1}\t{2}",
list[i].SourceArtifactId,
list[i].TargetArtifactId,
list[i].Score
);
}
#endif
Assert.AreEqual(6, list.Count);
TLLinksList expected = new TLLinksList();
expected.Add(new TLSingleLink("A", "B*", 10));
expected.Add(new TLSingleLink("A", "E", 9));
expected.Add(new TLSingleLink("A", "F", 8));
expected.Add(new TLSingleLink("A", "C*", 7));
expected.Add(new TLSingleLink("A", "G", 6));
expected.Add(new TLSingleLink("A", "H", 5));
for (int i = 0; i < expected.Count; i++)
{
Assert.AreEqual(expected[i], list[i]);
}
}
public void GetLinksAboveThresholdDefault()
{
sims.Threshold = 4;
TLLinksList list = TLSimilarityMatrixUtil.GetLinksAboveThreshold(sims);
list.Sort();
#if Verbose
Console.WriteLine("TLSimilarityMatrixUtilTest.GetLinksAboveThresholdDefault()");
for (int i = 0; i < list.Count; i++)
{
Console.WriteLine("{0}\t{1}\t{2}",
list[i].SourceArtifactId,
list[i].TargetArtifactId,
list[i].Score
);
}
#endif
Assert.AreEqual(6, list.Count);
TLLinksList expected = new TLLinksList();
expected.Add(new TLSingleLink("A", "B*", 10));
expected.Add(new TLSingleLink("A", "E", 9));
expected.Add(new TLSingleLink("A", "F", 8));
expected.Add(new TLSingleLink("A", "C*", 7));
expected.Add(new TLSingleLink("A", "G", 6));
expected.Add(new TLSingleLink("A", "H", 5));
for (int i = 0; i < expected.Count; i++)
{
Assert.AreEqual(expected[i], list[i]);
}
}
/// <summary>
/// Returns all links above the given threshold.
/// </summary>
/// <param name="matrix">Matrix</param>
/// <param name="threshold">Score threshold</param>
/// <returns>List of links above threshold</returns>
public static TLLinksList GetLinksAboveThreshold(TLSimilarityMatrix matrix, double threshold)
{
TLLinksList links = new TLLinksList();
foreach (TLSingleLink link in matrix.AllLinks)
{
if (link.Score > threshold)
{
links.Add(link);
}
}
return(links);
}
/// <summary>
/// Computes cosine similarities between a set of boolean document vectors and a tfidf weighted corpus
/// </summary>
/// <param name="ids">Boolean document vectors</param>
/// <param name="tfidf">tf-idf weighted document vectors</param>
/// <returns>Similarity matrix</returns>
private static TLSimilarityMatrix ComputeSimilarities(TermDocumentMatrix ids, TermDocumentMatrix tfidf)
{
TLSimilarityMatrix sims = new TLSimilarityMatrix();
List <TermDocumentMatrix> matrices = TermDocumentMatrix.Equalize(ids, tfidf);
for (int i = 0; i < ids.NumDocs; i++)
{
TLLinksList links = new TLLinksList();
for (int j = 0; j < tfidf.NumDocs; j++)
{
double product = 0.0;
double asquared = 0.0;
double bsquared = 0.0;
for (int k = 0; k < matrices[0].NumTerms; k++)
{
double a = matrices[0][i, k];
double b = matrices[1][j, k];
product += (a * b);
asquared += Math.Pow(a, 2);
bsquared += Math.Pow(b, 2);
}
double cross = Math.Sqrt(asquared) * Math.Sqrt(bsquared);
if (cross == 0.0)
{
links.Add(new TLSingleLink(ids.GetDocumentName(i), tfidf.GetDocumentName(j), 0.0));
}
else
{
links.Add(new TLSingleLink(ids.GetDocumentName(i), tfidf.GetDocumentName(j), product / cross));
}
}
links.Sort();
foreach (TLSingleLink link in links)
{
sims.AddLink(link.SourceArtifactId, link.TargetArtifactId, link.Score);
}
}
return(sims);
}
public static TLSimilarityMatrix Compute(TLSimilarityMatrix matrix, TLSimilarityMatrix relationships)
{
// create pseudo matrix for easy lookup
// Dictionary<sourceID, Dictionary<targetID, score>>
Dictionary <string, Dictionary <string, double> > storage = new Dictionary <string, Dictionary <string, double> >();
foreach (TLSingleLink link in matrix.AllLinks)
{
if (!storage.ContainsKey(link.SourceArtifactId))
{
storage.Add(link.SourceArtifactId, new Dictionary <string, double>());
}
storage[link.SourceArtifactId].Add(link.TargetArtifactId, link.Score);
}
#if UseDelta
// compute delta
double delta = SharedUtils.ComputeDelta(matrix);
#endif
// iterate over every (source, target) pair
TLLinksList links = matrix.AllLinks;
links.Sort();
foreach (TLSingleLink link in links)
{
// get the set of target artifacts related to link.TargetArtifactId
// then update the value of (link.SourceArtifactId, relatedArtifact) by delta
foreach (string relatedArtifact in relationships.GetSetOfTargetArtifactIdsAboveThresholdForSourceArtifact(link.TargetArtifactId))
{
#if UseDelta
storage[link.SourceArtifactId][relatedArtifact] += storage[link.SourceArtifactId][relatedArtifact] * delta;
#else
storage[link.SourceArtifactId][relatedArtifact] += storage[link.SourceArtifactId][relatedArtifact] * 0.1;
#endif
}
}
// build new matrix
TLLinksList newLinks = new TLLinksList();
foreach (string source in storage.Keys)
{
foreach (string target in storage[source].Keys)
{
newLinks.Add(new TLSingleLink(source, target, storage[source][target]));
}
}
newLinks.Sort();
TLSimilarityMatrix newMatrix = new TLSimilarityMatrix();
foreach (TLSingleLink link in newLinks)
{
newMatrix.AddLink(link.SourceArtifactId, link.TargetArtifactId, link.Score);
}
return(newMatrix);
}
public static TLSimilarityMatrix Compute(TLSimilarityMatrix matrix, TLSimilarityMatrix relationships)
{
// create pseudo matrix for easy lookup
// Dictionary<sourceID, Dictionary<targetID, score>>
Dictionary<string, Dictionary<string, double>> storage = new Dictionary<string, Dictionary<string, double>>();
foreach (TLSingleLink link in matrix.AllLinks)
{
if (!storage.ContainsKey(link.SourceArtifactId))
{
storage.Add(link.SourceArtifactId, new Dictionary<string, double>());
}
storage[link.SourceArtifactId].Add(link.TargetArtifactId, link.Score);
}
#if UseDelta
// compute delta
double delta = SharedUtils.ComputeDelta(matrix);
#endif
// iterate over every (source, target) pair
TLLinksList links = matrix.AllLinks;
links.Sort();
foreach (TLSingleLink link in links)
{
// get the set of target artifacts related to link.TargetArtifactId
// then update the value of (link.SourceArtifactId, relatedArtifact) by delta
foreach (string relatedArtifact in relationships.GetSetOfTargetArtifactIdsAboveThresholdForSourceArtifact(link.TargetArtifactId))
{
#if UseDelta
storage[link.SourceArtifactId][relatedArtifact] += storage[link.SourceArtifactId][relatedArtifact] * delta;
#else
storage[link.SourceArtifactId][relatedArtifact] += storage[link.SourceArtifactId][relatedArtifact] * 0.1;
#endif
}
}
// build new matrix
TLLinksList newLinks = new TLLinksList();
foreach (string source in storage.Keys)
{
foreach (string target in storage[source].Keys)
{
newLinks.Add(new TLSingleLink(source, target, storage[source][target]));
}
}
newLinks.Sort();
TLSimilarityMatrix newMatrix = new TLSimilarityMatrix();
foreach (TLSingleLink link in newLinks)
{
newMatrix.AddLink(link.SourceArtifactId, link.TargetArtifactId, link.Score);
}
return newMatrix;
}
/// <summary>
/// Removes a percentage of links from the bottom of the list.
/// </summary>
/// <param name="links">Ranklist</param>
/// <param name="percent">Percentage to remove</param>
/// <returns>Trimmed ranklist</returns>
public static TLLinksList RemoveBottomPercentage(TLLinksList links, double percent)
{
if (percent <= 0.0 || percent >= 1.0)
{
throw new DevelopmentKitException("Percentage level must be between 0 and 1.");
}
TLLinksList remaining = new TLLinksList();
links.Sort();
int endIndex = Convert.ToInt32(Math.Floor(links.Count * (1 - percent))) - 1;
for (int i = 0; i < endIndex; i++)
{
TLSingleLink link = links[i];
remaining.Add(new TLSingleLink(link.SourceArtifactId, link.TargetArtifactId, link.Score));
}
return(remaining);
}
public static TLLinksList GetLinksAtRecall(TLSimilarityMatrix sims, TLSimilarityMatrix oracle, RecallLevel level)
{
double totalCorrect = oracle.Count * RecallLevelUtil.RecallValue(level);
int numCorrect = 0;
TLLinksList list = new TLLinksList();
TLLinksList links = sims.AllLinks;
links.Sort();
while (links.Count > 0 && numCorrect < totalCorrect)
{
TLSingleLink link = links[0];
if (oracle.IsLinkAboveThreshold(link.SourceArtifactId, link.TargetArtifactId))
{
numCorrect++;
}
list.Add(link);
links.RemoveAt(0);
}
return list;
}
public static TLLinksList GetLinksAtRecall(TLSimilarityMatrix sims, TLSimilarityMatrix oracle, RecallLevel level)
{
double totalCorrect = oracle.Count * RecallLevelUtil.RecallValue(level);
int numCorrect = 0;
TLLinksList list = new TLLinksList();
TLLinksList links = sims.AllLinks;
links.Sort();
while (links.Count > 0 && numCorrect < totalCorrect)
{
TLSingleLink link = links[0];
if (oracle.IsLinkAboveThreshold(link.SourceArtifactId, link.TargetArtifactId))
{
numCorrect++;
}
list.Add(link);
links.RemoveAt(0);
}
return(list);
}
/// <summary>
/// Computes Jensen-Shannon divergence on two TermDocumentMatrices
/// </summary>
/// <param name="source">Source artifacts collection</param>
/// <param name="target">Target artifacts collection</param>
/// <returns>Similarity matrix</returns>
public static TLSimilarityMatrix Compute(TermDocumentMatrix source, TermDocumentMatrix target)
{
List<TermDocumentMatrix> matrices = TermDocumentMatrix.Equalize(source, target);
TLSimilarityMatrix sims = new TLSimilarityMatrix();
for (int i = 0; i < matrices[0].NumDocs; i++)
{
TLLinksList list = new TLLinksList();
for (int j = 0; j < matrices[1].NumDocs; j++)
{
list.Add(new TLSingleLink(matrices[0].GetDocumentName(i), matrices[1].GetDocumentName(j),
DocumentSimilarity(matrices[0].GetDocument(i), matrices[1].GetDocument(j))));
}
list.Sort();
foreach (TLSingleLink link in list)
{
sims.AddLink(link.SourceArtifactId, link.TargetArtifactId, link.Score);
}
}
return sims;
}
/// <summary>
/// Computes Jensen-Shannon divergence on two TermDocumentMatrices
/// </summary>
/// <param name="source">Source artifacts collection</param>
/// <param name="target">Target artifacts collection</param>
/// <returns>Similarity matrix</returns>
public static TLSimilarityMatrix Compute(TermDocumentMatrix source, TermDocumentMatrix target)
{
List <TermDocumentMatrix> matrices = TermDocumentMatrix.Equalize(source, target);
TLSimilarityMatrix sims = new TLSimilarityMatrix();
for (int i = 0; i < matrices[0].NumDocs; i++)
{
TLLinksList list = new TLLinksList();
for (int j = 0; j < matrices[1].NumDocs; j++)
{
list.Add(new TLSingleLink(matrices[0].GetDocumentName(i), matrices[1].GetDocumentName(j),
DocumentSimilarity(matrices[0].GetDocument(i), matrices[1].GetDocument(j))));
}
list.Sort();
foreach (TLSingleLink link in list)
{
sims.AddLink(link.SourceArtifactId, link.TargetArtifactId, link.Score);
}
}
return(sims);
}
/// <summary>
/// Returns the top N scoring links in a matrix.
/// </summary>
/// <param name="matrix">Matrix</param>
/// <param name="topN">Number of links to return</param>
/// <returns>List of top N links</returns>
public static TLLinksList GetTopNLinks(TLSimilarityMatrix matrix, int topN)
{
if (matrix.AllLinks.Count < topN)
{
throw new DevelopmentKitException("Matrix only has " + matrix.AllLinks.Count + " links (" + topN + " requested).");
}
if (topN < 1)
{
throw new DevelopmentKitException("topN must be greater than 0.");
}
TLLinksList links = matrix.AllLinks;
links.Sort();
TLLinksList newLinks = new TLLinksList();
for (int i = 0; i < topN; i++)
{
newLinks.Add(links[i]);
}
return(newLinks);
}
/// <summary>
/// Computes cosine similarities between a set of boolean document vectors and a tfidf weighted corpus
/// </summary>
/// <param name="ids">Boolean document vectors</param>
/// <param name="tfidf">tf-idf weighted document vectors</param>
/// <returns>Similarity matrix</returns>
private static TLSimilarityMatrix ComputeSimilarities(TermDocumentMatrix ids, TermDocumentMatrix tfidf)
{
TLSimilarityMatrix sims = new TLSimilarityMatrix();
List<TermDocumentMatrix> matrices = TermDocumentMatrix.Equalize(ids, tfidf);
for (int i = 0; i < ids.NumDocs; i++)
{
TLLinksList links = new TLLinksList();
for (int j = 0; j < tfidf.NumDocs; j++)
{
double product = 0.0;
double asquared = 0.0;
double bsquared = 0.0;
for (int k = 0; k < matrices[0].NumTerms; k++)
{
double a = matrices[0][i, k];
double b = matrices[1][j, k];
product += (a * b);
asquared += Math.Pow(a, 2);
bsquared += Math.Pow(b, 2);
}
double cross = Math.Sqrt(asquared) * Math.Sqrt(bsquared);
if (cross == 0.0)
{
links.Add(new TLSingleLink(ids.GetDocumentName(i), tfidf.GetDocumentName(j), 0.0));
}
else
{
links.Add(new TLSingleLink(ids.GetDocumentName(i), tfidf.GetDocumentName(j), product / cross));
}
}
links.Sort();
foreach (TLSingleLink link in links)
{
sims.AddLink(link.SourceArtifactId, link.TargetArtifactId, link.Score);
}
}
return sims;
}
public void GetLinksAtRecall100()
{
TLLinksList list = TLSimilarityMatrixUtil.GetLinksAtRecall(sims, oracle, 1.0);
#if Verbose
Console.WriteLine("TLSimilarityMatrixUtilTest.GetLinksAtRecall100()");
for (int i = 0; i < list.Count; i++)
{
Console.WriteLine("{0}\t{1}\t{2}",
list[i].SourceArtifactId,
list[i].TargetArtifactId,
list[i].Score
);
}
#endif
Assert.AreEqual(9, list.Count);
TLLinksList expected = new TLLinksList();
expected.Add(new TLSingleLink("A", "B*", 10));
expected.Add(new TLSingleLink("A", "E", 9));
expected.Add(new TLSingleLink("A", "F", 8));
expected.Add(new TLSingleLink("A", "C*", 7));
expected.Add(new TLSingleLink("A", "G", 6));
expected.Add(new TLSingleLink("A", "H", 5));
expected.Add(new TLSingleLink("A", "I", 4));
expected.Add(new TLSingleLink("A", "J", 3));
expected.Add(new TLSingleLink("A", "D*", 2));
for (int i = 0; i < expected.Count; i++)
{
Assert.AreEqual(expected[i], list[i]);
}
}
/// <summary>
/// Get relevant links for source artifacts with score larger than threshold.
/// </summary>
/// <param name="sourceArtifactId">Id of source artifact for which the set of relevant/retrieved links is requested</param>
/// <returns>Hashset of target artifacts ids that are retrieved or relevant to the given source artifact (depends on usage).</returns>
public TLLinksList GetLinksAboveThresholdForSourceArtifact(string sourceArtifactId)
{
TLLinksList linksForSourceArtifact;
if (CacheOfLinksPerSourceArtifacts.TryGetValue(sourceArtifactId, out linksForSourceArtifact) == false)
{
linksForSourceArtifact = new TLLinksList();
Dictionary<string, double> links;
if (m_matrix.TryGetValue(sourceArtifactId, out links))
{
foreach (string targetArtifactId in links.Keys)
{
if (links[targetArtifactId] > Threshold)
{
linksForSourceArtifact.Add(new TLSingleLink(sourceArtifactId, targetArtifactId, links[targetArtifactId]));
}
}
}
CacheOfLinksPerSourceArtifacts.Add(sourceArtifactId, linksForSourceArtifact);
}
return linksForSourceArtifact; //return empty set
}
/// <summary>
/// Removes a percentage of links from the top of the list.
/// </summary>
/// <param name="links">Ranklist</param>
/// <param name="percent">Percentage to remove</param>
/// <returns>Trimmed ranklist</returns>
public static TLLinksList RemoveTopPercentage(TLLinksList links, double percent)
{
if (percent <= 0.0 || percent >= 1.0)
{
throw new DevelopmentKitException("Percentage level must be between 0 and 1.");
}
TLLinksList remaining = new TLLinksList();
links.Sort();
int startIndex = Convert.ToInt32(Math.Ceiling(links.Count * percent)) - 1;
for (int i = startIndex; i < links.Count; i++)
{
TLSingleLink link = links[i];
remaining.Add(new TLSingleLink(link.SourceArtifactId, link.TargetArtifactId, link.Score));
}
return remaining;
}
/// <summary>
/// Returns the top N scoring links in a matrix.
/// </summary>
/// <param name="matrix">Matrix</param>
/// <param name="topN">Number of links to return</param>
/// <returns>List of top N links</returns>
public static TLLinksList GetTopNLinks(TLSimilarityMatrix matrix, int topN)
{
if (matrix.AllLinks.Count < topN)
{
throw new DevelopmentKitException("Matrix only has " + matrix.AllLinks.Count + " links (" + topN + " requested).");
}
if (topN < 1)
{
throw new DevelopmentKitException("topN must be greater than 0.");
}
TLLinksList links = matrix.AllLinks;
links.Sort();
TLLinksList newLinks = new TLLinksList();
for (int i = 0; i < topN; i++)
{
newLinks.Add(links[i]);
}
return newLinks;
}
/// <summary>
/// Returns links for the desired recall level.
/// </summary>
/// <param name="matrix">Candidate matrix</param>
/// <param name="answerMatrix">Answer matrix</param>
/// <param name="level">Desired recall level</param>
/// <returns>List of links at desired recall</returns>
public static TLLinksList GetLinksAtRecall(TLSimilarityMatrix matrix, TLSimilarityMatrix answerMatrix, double level)
{
if (level <= 0.0 || level > 1.0)
{
throw new DevelopmentKitException("Recall level must be between 0 and 1.");
}
double totalCorrect = answerMatrix.Count * level;
int numCorrect = 0;
TLLinksList links = matrix.AllLinks;
links.Sort();
TLLinksList newLinks = new TLLinksList();
while (links.Count > 0 && numCorrect < totalCorrect)
{
TLSingleLink link = links[0];
if (answerMatrix.IsLinkAboveThreshold(link.SourceArtifactId, link.TargetArtifactId))
{
numCorrect++;
}
newLinks.Add(link);
links.RemoveAt(0);
}
return newLinks;
}
/// <summary>
/// Returns all links above the given threshold.
/// </summary>
/// <param name="matrix">Matrix</param>
/// <param name="threshold">Score threshold</param>
/// <returns>List of links above threshold</returns>
public static TLLinksList GetLinksAboveThreshold(TLSimilarityMatrix matrix, double threshold)
{
TLLinksList links = new TLLinksList();
foreach (TLSingleLink link in matrix.AllLinks)
{
if (link.Score > threshold)
links.Add(link);
}
return links;
}
public void GetTopNLinks()
{
TLLinksList list = TLSimilarityMatrixUtil.GetTopNLinks(sims, 4);
#if Verbose
Console.WriteLine("TLSimilarityMatrixUtilTest.GetTopNLinks()");
for (int i = 0; i < list.Count; i++)
{
Console.WriteLine("{0}\t{1}\t{2}",
list[i].SourceArtifactId,
list[i].TargetArtifactId,
list[i].Score
);
}
#endif
Assert.AreEqual(4, list.Count);
TLLinksList expected = new TLLinksList();
expected.Add(new TLSingleLink("A", "B*", 10));
expected.Add(new TLSingleLink("A", "E", 9));
expected.Add(new TLSingleLink("A", "F", 8));
expected.Add(new TLSingleLink("A", "C*", 7));
for (int i = 0; i < expected.Count; i++)
{
Assert.AreEqual(expected[i], list[i]);
}
}
