public override void Compute() { TermDocumentMatrix sourceArtifacts = SmoothingFilter.Compute(new TermDocumentMatrix((TLArtifactsCollection)Workspace.Load("SourceArtifacts"))); TermDocumentMatrix targetArtifacts = SmoothingFilter.Compute(new TermDocumentMatrix((TLArtifactsCollection)Workspace.Load("TargetArtifacts"))); TLSimilarityMatrix sims = VSM.Compute(sourceArtifacts, targetArtifacts); Workspace.Store("Similarities", sims); }
public override void Compute() { TLArtifactsCollection sourceArtifacts = (TLArtifactsCollection)Workspace.Load("SourceArtifacts"); TLArtifactsCollection targetArtifacts = (TLArtifactsCollection)Workspace.Load("TargetArtifacts"); TLSimilarityMatrix sims = VSM.Compute(sourceArtifacts, targetArtifacts); Workspace.Store("Similarities", sims); }
public void ComputeTest() { string data = @"../../Data/SimpleCorpus."; TLArtifactsCollection source = Artifacts.Import(data + "input.source.txt"); TLArtifactsCollection target = Artifacts.Import(data + "input.target.txt"); TLSimilarityMatrix testsims = VSM.Compute(source, target); TLSimilarityMatrix realsims = Similarities.Import(data + "output.VSM.txt"); Assert.AreEqual(testsims.Count, realsims.Count); TLLinksList testlinks = testsims.AllLinks; TLLinksList reallinks = realsims.AllLinks; testlinks.Sort(); reallinks.Sort(); for (int i = 0; i < reallinks.Count; i++) { Assert.AreEqual(testlinks[i].SourceArtifactId, reallinks[i].SourceArtifactId); Assert.AreEqual(testlinks[i].TargetArtifactId, reallinks[i].TargetArtifactId); Assert.AreEqual(testlinks[i].Score, reallinks[i].Score, 0.000000001); } }
public void BooleanQueriesAndTFIDFCorpusTest() { string inputData = Settings.Default.SimpleCorpusDir; string outputData = Path.Combine(inputData, "VSM"); TLArtifactsCollection source = Artifacts.ImportFile(Path.Combine(inputData, "source.txt")); TLArtifactsCollection target = Artifacts.ImportFile(Path.Combine(inputData, "target.txt")); TLSimilarityMatrix testsims = VSM.Compute(source, target, VSMWeightEnum.BooleanQueriesAndTFIDFCorpus); TLSimilarityMatrix realsims = Similarities.Import(Path.Combine(outputData, "output.txt")); Assert.AreEqual(testsims.Count, realsims.Count); TLLinksList testlinks = testsims.AllLinks; TLLinksList reallinks = realsims.AllLinks; testlinks.Sort(); reallinks.Sort(); for (int i = 0; i < reallinks.Count; i++) { Assert.AreEqual(testlinks[i].SourceArtifactId, reallinks[i].SourceArtifactId); Assert.AreEqual(testlinks[i].TargetArtifactId, reallinks[i].TargetArtifactId); Assert.AreEqual(testlinks[i].Score, reallinks[i].Score, Settings.Default.DoublePrecision); } }