Merge pull request #60 from JohT/feature/optimize-similarity

JohT · web-flow · commit f17588bc9280 · 2023-10-29T10:19:50.000+01:00
Optimize Similarity using in-memory mutation
diff --git a/cypher/Artifact_Dependencies/Artifacts_with_duplicate_packages.cypher b/cypher/Artifact_Dependencies/Artifacts_with_duplicate_packages.cypher
@@ -0,0 +1,12 @@
+// Artifacts with the same full qualified package name (duplicate packages). These can lead to confusion and provide access to package protected classes to another artifact that might not be intended.
+
+ MATCH (artifact:Artifact)-[:CONTAINS]->(package:Package)
+ WHERE EXISTS ((package)-[:CONTAINS]->(:Type))
+  WITH package.fqn AS packageName
+      ,replace(last(split(artifact.fileName, '/')), '.jar', '') as artifactName
+  WITH packageName
+      ,collect(DISTINCT artifactName) AS artifactNames
+      ,count(*) AS duplicates 
+ WHERE duplicates > 1
+RETURN packageName, duplicates, artifactNames
+ ORDER BY duplicates DESCENDING
diff --git a/cypher/Centrality/Centrality_9a_Hyperlink_Induced_Topic_Search_HITS_Estimate.cypher b/cypher/Centrality/Centrality_9a_Hyperlink_Induced_Topic_Search_HITS_Estimate.cypher
@@ -1,6 +1,6 @@
 // Centrality 9a Hyperlink-Induced Topic Search (HITS) Memory Estimation
 
-  CALL gds.alpha.hits.write.estimate(
+  CALL gds.hits.write.estimate(
  $dependencies_projection + '-cleaned', {
      hitsIterations: 20
     ,authProperty: $dependencies_projection_write_property + 'Authority'
diff --git a/cypher/Centrality/Centrality_9b_Hyperlink_Induced_Topic_Search_HITS_Statistics.cypher b/cypher/Centrality/Centrality_9b_Hyperlink_Induced_Topic_Search_HITS_Statistics.cypher
@@ -1,6 +1,6 @@
 // Centrality 9b Hyperlink-Induced Topic Search (HITS) Statistics
 
-  CALL gds.alpha.hits.stats(
+  CALL gds.hits.stats(
  $dependencies_projection + '-cleaned', {
     hitsIterations: 20
 })
diff --git a/cypher/Centrality/Centrality_9c_Hyperlink_Induced_Topic_Search_HITS_Mutate.cypher b/cypher/Centrality/Centrality_9c_Hyperlink_Induced_Topic_Search_HITS_Mutate.cypher
@@ -1,6 +1,6 @@
 // Centrality 9c Hyperlink-Induced Topic Search (HITS) Mutate
 
-  CALL gds.alpha.hits.mutate(
+  CALL gds.hits.mutate(
  $dependencies_projection + '-cleaned', {
      hitsIterations: 20
     ,authProperty: $dependencies_projection_write_property + 'Authority'
diff --git a/cypher/Centrality/Centrality_9d_Hyperlink_Induced_Topic_Search_HITS_Stream.cypher b/cypher/Centrality/Centrality_9d_Hyperlink_Induced_Topic_Search_HITS_Stream.cypher
@@ -1,6 +1,6 @@
 // Centrality 9d Hyperlink-Induced Topic Search (HITS) Stream
 
-  CALL gds.alpha.hits.stream(
+  CALL gds.hits.stream(
  $dependencies_projection + '-cleaned', {
     hitsIterations: 20
 })
diff --git a/cypher/Centrality/Centrality_9e_Hyperlink_Induced_Topic_Search_HITS_Write.cypher b/cypher/Centrality/Centrality_9e_Hyperlink_Induced_Topic_Search_HITS_Write.cypher
@@ -1,6 +1,6 @@
 // Centrality 9e Hyperlink-Induced Topic Search (HITS) Write
 
-  CALL gds.alpha.hits.write(
+  CALL gds.hits.write(
  $dependencies_projection + '-cleaned', {
      hitsIterations: 20
     ,authProperty: $dependencies_projection_write_property + 'Authority'
diff --git a/cypher/Community_Detection/Community_Detection_2a_Leiden_Estimate.cypher b/cypher/Community_Detection/Community_Detection_2a_Leiden_Estimate.cypher
@@ -1,6 +1,6 @@
 //Community Detection Leiden Estimate Memory
 
-CALL gds.beta.leiden.write.estimate(
+CALL gds.leiden.write.estimate(
  $dependencies_projection + '-cleaned', {
   gamma: toFloat($dependencies_leiden_gamma),
   theta: 0.001,
diff --git a/cypher/Community_Detection/Community_Detection_2b_Leiden_Statistics.cypher b/cypher/Community_Detection/Community_Detection_2b_Leiden_Statistics.cypher
@@ -1,6 +1,6 @@
 //Community Detection Leiden Statistics
 
-CALL gds.beta.leiden.stats(
+CALL gds.leiden.stats(
  $dependencies_projection + '-cleaned', {
   gamma: toFloat($dependencies_leiden_gamma),
   theta: 0.001,
diff --git a/cypher/Community_Detection/Community_Detection_2d_Leiden_Stream.cypher b/cypher/Community_Detection/Community_Detection_2d_Leiden_Stream.cypher
@@ -1,6 +1,6 @@
 //Community Detection Leiden Stream
 
-CALL gds.beta.leiden.stream(
+CALL gds.leiden.stream(
  $dependencies_projection + '-cleaned', {
   gamma: toFloat($dependencies_leiden_gamma),
   theta: 0.001,
diff --git a/cypher/Community_Detection/Community_Detection_2d_Leiden_Write_Node_Property.cypher b/cypher/Community_Detection/Community_Detection_2d_Leiden_Write_Node_Property.cypher
@@ -1,6 +1,6 @@
 //Community Detection Leiden Write property communityLeidenId
 
-CALL gds.beta.leiden.write(
+CALL gds.leiden.write(
  $dependencies_projection + '-cleaned', {
   gamma: toFloat($dependencies_leiden_gamma),
   theta: 0.001,
diff --git a/cypher/Community_Detection/Community_Detection_7d_Modularity.cypher b/cypher/Community_Detection/Community_Detection_7d_Modularity.cypher
@@ -1,6 +1,6 @@
 // Community Detection Modularity
 
-CALL gds.alpha.modularity.stream(
+CALL gds.modularity.stream(
  $dependencies_projection + '-cleaned', {
      relationshipWeightProperty: $dependencies_projection_weight_property
     ,communityProperty: $dependencies_projection_write_property
diff --git a/cypher/Community_Detection/Community_Detection_Summary.cypher b/cypher/Community_Detection/Community_Detection_Summary.cypher
@@ -0,0 +1,18 @@
+// Community Detection Summary. Variables: dependencies_projection_node ("Artifact", "Package", "Type")
+
+MATCH (codeUnit)
+WHERE (codeUnit.incomingDependencies > 0 OR codeUnit.outgoingDependencies > 0)
+  AND $dependencies_projection_node IN LABELS(codeUnit) 
+RETURN coalesce(codeUnit.fqn, codeUnit.fileName, codeUnit.signature, codeUnit.name)      AS name
+      ,coalesce(codeUnit.name, replace(last(split(codeUnit.fileName, '/')), '.jar', '')) AS shortName
+      ,codeUnit.communityLouvainId                  AS louvainId
+      ,codeUnit.communityLouvainIntermediateIds     AS louvainIntermediateIds
+      ,codeUnit.communityLeidenId                   AS leidenId
+      ,codeUnit.communityLeidenIntermediateIds      AS leidenIntermediateIds                   
+      ,codeUnit.communityLeidenIdModularity         AS leidenModularity
+      ,codeUnit.communityWeaklyConnectedComponentId AS weaklyConnectedComponentId
+      ,codeUnit.communityLabelPropagationId         AS labelPropagationId
+      ,codeUnit.communityKCoreDecompositionValue    AS kCoreDecompositionValue
+      ,codeUnit.communityMaximumKCutId              AS maximumKCutId
+      ,codeUnit.incomingDependencies                AS incomingDependencies
+      ,codeUnit.outgoingDependencies                AS outgoingDependencies
diff --git a/cypher/Dependencies_Projection/Dependencies_3c_Create_Type_Projection.cypher b/cypher/Dependencies_Projection/Dependencies_3c_Create_Type_Projection.cypher
@@ -0,0 +1,20 @@
+// Create filtered Type node projection without zero-degree nodes, external types, java types or duplicates. Variables: dependencies_projection. Requires 'Label_base_java_types', 'Label_buildin_java_types' and 'Label_resolved_duplicate_types' of 'Types' directory.
+
+  MATCH (internalType:Type&!PrimitiveType&!Void&!JavaType&!ResolvedDuplicateType&!ExternalType)
+ OPTIONAL MATCH (internalType)-[typeDependency:DEPENDS_ON]->(dependentType:Type&!PrimitiveType&!Void&!JavaType&!ResolvedDuplicateType&!ExternalType)
+  WITH internalType
+      ,typeDependency
+      ,dependentType
+      ,exists((internalType)<-[:DEPENDS_ON]-(:Type))   AS hasIncomingDependenciesInternal
+ WHERE (hasIncomingDependenciesInternal OR dependentType IS NOT NULL)
+  WITH gds.graph.project($dependencies_projection + '-cleaned', internalType, dependentType, {
+         sourceNodeProperties: internalType {.incomingDependencies, .outgoingDependencies}
+        ,targetNodeProperties: dependentType {.incomingDependencies, .outgoingDependencies}
+        ,relationshipProperties: typeDependency {.weight}
+        ,relationshipType: 'DEPENDS_ON'
+      }
+      ) AS projection
+RETURN projection.graphName         AS graphName
+      ,projection.nodeCount         AS nodeCount
+      ,projection.relationshipCount AS relationshipCount
+      ,projection.projectMillis     AS projectMillis
diff --git a/cypher/Dependencies_Projection/Dependencies_4c_Create_Undirected_Type_Projection.cypher b/cypher/Dependencies_Projection/Dependencies_4c_Create_Undirected_Type_Projection.cypher
@@ -0,0 +1,23 @@
+// Create filtered Type node projection without zero-degree nodes, external types, java types or duplicates. Variables: dependencies_projection. Requires 'Label_base_java_types', 'Label_buildin_java_types' and 'Label_resolved_duplicate_types' of 'Types' directory.
+
+  MATCH (internalType:Type&!PrimitiveType&!Void&!JavaType&!ResolvedDuplicateType&!ExternalType)
+ OPTIONAL MATCH (internalType)-[typeDependency:DEPENDS_ON]->(dependentType:Type&!PrimitiveType&!Void&!JavaType&!ResolvedDuplicateType&!ExternalType)
+  WITH internalType
+      ,typeDependency
+      ,dependentType
+      ,exists((internalType)<-[:DEPENDS_ON]-(:Type))   AS hasIncomingDependenciesInternal
+ WHERE (hasIncomingDependenciesInternal OR dependentType IS NOT NULL)
+  WITH gds.graph.project($dependencies_projection + '-cleaned', internalType, dependentType, {
+         sourceNodeProperties: internalType {.incomingDependencies, .outgoingDependencies}
+        ,targetNodeProperties: dependentType {.incomingDependencies, .outgoingDependencies}
+        ,relationshipProperties: typeDependency {.weight}
+        ,relationshipType: 'DEPENDS_ON'
+      },
+      {
+         undirectedRelationshipTypes: ['*']
+      }
+      ) AS projection
+RETURN projection.graphName         AS graphName
+      ,projection.nodeCount         AS nodeCount
+      ,projection.relationshipCount AS relationshipCount
+      ,projection.projectMillis     AS projectMillis
diff --git a/cypher/Dependencies_Projection/Dependencies_5_Create_Subgraph.cypher b/cypher/Dependencies_Projection/Dependencies_5_Create_Subgraph.cypher
@@ -1,6 +1,6 @@
 //Create filtered subgraph projection without zero-degree nodes. Variables: dependencies_projection, dependencies_projection_node
 
-CALL gds.beta.graph.project.subgraph(
+CALL gds.graph.filter(
   $dependencies_projection + '-cleaned',
   $dependencies_projection,
   'n.outgoingDependencies > 0 OR n.incomingDependencies > 0',
diff --git a/cypher/External_Dependencies/Label_external_types_and_annotations.cypher b/cypher/External_Dependencies/Label_external_types_and_annotations.cypher
@@ -1,17 +1,13 @@
-// Label external types and annotations
+// Label external types and external annotations. Requires 'Label_base_java_types', 'Label_buildin_java_types' and 'Label_resolved_duplicate_types' of 'Types' directory.
 
- MATCH (externalType:Type)
- WHERE externalType.byteCodeVersion IS NULL // byte code not available -> external dependency
-  WITH externalType
-      ,(NOT externalType.fqn CONTAINS '.')                       AS isPrimitiveType
-      ,(externalType.fqn STARTS WITH 'java.')                    AS isJavaType
-      ,exists((externalType)-[:RESOLVES_TO]->(:Type))            AS isAlsoInternalType
-      ,exists((externalType)<-[:OF_TYPE]-()<-[:ANNOTATED_BY]-()) AS isAnnotation
- WHERE isPrimitiveType    = false
-   AND isJavaType         = false
-   AND isAlsoInternalType = false
-  WITH externalType
-      ,CASE WHEN isAnnotation THEN [1] ELSE [] END AS annotated
-FOREACH (x in annotated | SET externalType:ExternalAnnotation)
-    SET externalType:ExternalType 
- RETURN labels(externalType), count(externalType) as numberOfExternalTypes
+  MATCH (type:Type&!PrimitiveType&!Void&!JavaType&!ResolvedDuplicateType)
+   WITH type
+       ,type.byteCodeVersion IS NULL                      AS isExternalType
+       ,exists((type)<-[:OF_TYPE]-()<-[:ANNOTATED_BY]-()) AS isAnnotation
+   WITH type
+       ,isExternalType
+       ,CASE WHEN isAnnotation THEN [1] ELSE [] END AS annotation
+  WHERE isExternalType
+FOREACH (x in annotation | SET type:ExternalAnnotation)
+    SET type:ExternalType 
+ RETURN labels(type), count(type) as numberOfExternalTypes
diff --git a/cypher/Overview/Words_for_Wordcloud.cypher b/cypher/Overview/Words_for_Wordcloud.cypher
@@ -2,10 +2,8 @@
 
 MATCH (package:Package)
  WITH collect(package.name) AS packageNames
-MATCH (type:Type)
-WHERE type.byteCodeVersion IS NOT NULL  // no external types 
-  AND NOT type.fqn STARTS WITH 'java.'  // no java types
-  AND type.name <> 'package-info'       // not package-info files
+MATCH (type:Type&!PrimitiveType&!Void&!JavaType&!ResolvedDuplicateType&!ExternalType)
+WHERE type.name <> 'package-info'       // not package-info files
  WITH packageNames, split(type.name, '$') AS splittedInnerClasses
 UNWIND splittedInnerClasses AS splittedInnerClass
  WITH packageNames
diff --git a/cypher/Similarity/Similarity_1a_Estimate.cypher b/cypher/Similarity/Similarity_1a_Estimate.cypher
@@ -3,6 +3,7 @@
 CALL gds.nodeSimilarity.write.estimate(
  $dependencies_projection + '-cleaned', {
      relationshipWeightProperty: $dependencies_projection_weight_property
+    ,relationshipTypes: ['DEPENDS_ON']
     ,writeRelationshipType: 'SIMILAR'
     ,writeProperty: 'score'
     ,topK: 3
diff --git a/cypher/Similarity/Similarity_1b_Statistics.cypher b/cypher/Similarity/Similarity_1b_Statistics.cypher
@@ -3,6 +3,7 @@
 CALL gds.nodeSimilarity.stats(
   $dependencies_projection + '-cleaned', {
       relationshipWeightProperty: $dependencies_projection_weight_property
+     ,relationshipTypes: ['DEPENDS_ON']
      ,topK: 3
  })
  YIELD nodesCompared
diff --git a/cypher/Similarity/Similarity_1c_Mutate.cypher b/cypher/Similarity/Similarity_1c_Mutate.cypher
@@ -0,0 +1,32 @@
+// Similarity Mutate
+
+ CALL gds.nodeSimilarity.mutate(
+  $dependencies_projection + '-cleaned', {
+      relationshipWeightProperty: $dependencies_projection_weight_property
+     ,relationshipTypes: ['DEPENDS_ON']
+     ,topK: 3
+     ,mutateRelationshipType: 'SIMILAR'
+     ,mutateProperty: 'score'
+ })
+YIELD  relationshipsWritten
+      ,nodesCompared
+      ,preProcessingMillis
+      ,computeMillis
+      ,mutateMillis
+      ,postProcessingMillis
+      ,similarityDistribution
+RETURN relationshipsWritten
+      ,nodesCompared
+      ,preProcessingMillis
+      ,computeMillis
+      ,mutateMillis
+      ,postProcessingMillis
+      ,similarityDistribution.min
+      ,similarityDistribution.mean
+      ,similarityDistribution.max
+      ,similarityDistribution.p50
+      ,similarityDistribution.p75
+      ,similarityDistribution.p90
+      ,similarityDistribution.p95
+      ,similarityDistribution.p99
+      ,similarityDistribution.p999
diff --git a/cypher/Similarity/Similarity_1d_Stream_Mutated.cypher b/cypher/Similarity/Similarity_1d_Stream_Mutated.cypher
@@ -0,0 +1,35 @@
+// Read the similarity relationship from the projection. Variables: dependencies_projection
+
+CALL gds.graph.relationshipProperty.stream(
+     $dependencies_projection + '-cleaned'
+    ,'score'
+    ,['SIMILAR']
+)
+ YIELD sourceNodeId
+      ,targetNodeId
+      ,relationshipType
+      ,propertyValue
+ WITH gds.util.asNode(sourceNodeId) AS sourceNode
+     ,gds.util.asNode(targetNodeId) AS targetNode
+     ,propertyValue                 AS similarity
+OPTIONAL MATCH (sourceArtifact:Artifact)-[:CONTAINS]->(sourceNode)
+OPTIONAL MATCH (targetArtifact:Artifact)-[:CONTAINS]->(targetNode)
+ WITH sourceNode
+     ,targetNode
+     ,replace(last(split(sourceArtifact.fileName, '/')), '.jar', '')  AS sourceArtifactName
+     ,replace(last(split(targetArtifact.fileName, '/')), '.jar', '')  AS targetArtifactName
+     ,similarity
+ WHERE (sourceNode.incomingDependencies > 0
+   OR   sourceNode.outgoingDependencies > 0)
+   AND (targetNode.incomingDependencies > 0
+   OR   targetNode.outgoingDependencies > 0)
+RETURN similarity
+      ,sourceArtifactName
+      ,coalesce(sourceNode.fqn, sourceNode.fileName, sourceNode.name) AS sourceNodeName
+      ,sourceNode.incomingDependencies                                AS sourceNodeIncomingDependencies  
+      ,sourceNode.outgoingDependencies                                AS sourceNodeOutgoingDependencies   
+      ,targetArtifactName
+      ,coalesce(targetNode.fqn, targetNode.fileName, targetNode.name) AS targetNodeName
+      ,targetNode.incomingDependencies                                AS targetNodeIncomingDependencies  
+      ,targetNode.outgoingDependencies                                AS targetNodeOutgoingDependencies  
+ORDER BY similarity DESCENDING, sourceNodeName, targetNodeName
diff --git a/cypher/Similarity/Similarity_1e_Stream.cypher b/cypher/Similarity/Similarity_1e_Stream.cypher
@@ -3,6 +3,7 @@
  CALL gds.nodeSimilarity.stream(
   $dependencies_projection + '-cleaned', {
       relationshipWeightProperty: $dependencies_projection_weight_property
+     ,relationshipTypes: ['DEPENDS_ON']
      ,topK: 3
  })
 YIELD node1, node2, similarity
diff --git a/cypher/Similarity/Similarity_1f_Delete_Relationships.cypher b/cypher/Similarity/Similarity_1f_Delete_Relationships.cypher
diff --git a/cypher/Similarity/Similarity_1g_Write_Mutated.cypher b/cypher/Similarity/Similarity_1g_Write_Mutated.cypher
@@ -0,0 +1,9 @@
+// Write the Similarity relationship from the projection into the Graph. Variables: dependencies_projection, dependencies_projection_write_property
+
+CALL gds.graph.relationship.write(
+   $dependencies_projection + '-cleaned'
+  ,'SIMILAR'
+  ,'score'
+)
+ YIELD relationshipsWritten, propertiesWritten, writeMillis
+RETURN relationshipsWritten, propertiesWritten, writeMillis
diff --git a/cypher/Similarity/Similarity_1h_Write.cypher b/cypher/Similarity/Similarity_1h_Write.cypher
@@ -3,6 +3,7 @@
 CALL gds.nodeSimilarity.write(
  $dependencies_projection + '-cleaned', {
      relationshipWeightProperty: $dependencies_projection_weight_property
+    ,relationshipTypes: ['DEPENDS_ON']
     ,writeRelationshipType: 'SIMILAR'
     ,writeProperty: 'score'
     ,topK: 3
diff --git a/cypher/Similarity/Similarity_1i_Write_Node_Properties.cypher b/cypher/Similarity/Similarity_1i_Write_Node_Properties.cypher
diff --git a/cypher/Types/Label_base_java_types.cypher b/cypher/Types/Label_base_java_types.cypher
@@ -0,0 +1,15 @@
+// Label primitive Java types and void
+
+  MATCH (type:Type)
+   WITH type
+       ,type.fqn in ['byte', 'short', 'int', 'long', 'float', 'double', 'boolean', 'char'] AS isPrimitiveType
+       ,type.fqn = 'void' AS isVoid
+   WITH type
+       ,isPrimitiveType
+       ,isVoid
+       ,CASE WHEN isPrimitiveType     THEN [1] ELSE [] END AS primitive
+       ,CASE WHEN isVoid              THEN [1] ELSE [] END AS void
+  WHERE isPrimitiveType OR isVoid
+FOREACH (x in primitive         | SET type:PrimitiveType)
+FOREACH (x in void              | SET type:Void)
+ RETURN labels(type), count(type) as numberOfBaseTypes
diff --git a/cypher/Types/Label_buildin_java_types.cypher b/cypher/Types/Label_buildin_java_types.cypher
@@ -0,0 +1,34 @@
+// Label build-in Java types
+
+  MATCH (t:Type)
+   WITH t
+       ,CASE WHEN (t.fqn STARTS WITH 'java.') THEN true
+             WHEN (t.fqn STARTS WITH 'com.sun.') THEN true
+             WHEN (t.fqn STARTS WITH 'javax.accessibility') THEN true
+             WHEN (t.fqn STARTS WITH 'javax.annotation.processing') THEN true
+             WHEN (t.fqn STARTS WITH 'javax.crypto') THEN true
+             WHEN (t.fqn STARTS WITH 'javax.lang.model') THEN true
+             WHEN (t.fqn STARTS WITH 'javax.management') THEN true
+             WHEN (t.fqn STARTS WITH 'javax.naming') THEN true
+             WHEN (t.fqn STARTS WITH 'javax.net') THEN true
+             WHEN (t.fqn STARTS WITH 'javax.print') THEN true
+             WHEN (t.fqn STARTS WITH 'javax.rmi.ssl') THEN true
+             WHEN (t.fqn STARTS WITH 'javax.script') THEN true
+             WHEN (t.fqn STARTS WITH 'javax.security.') THEN true
+             WHEN (t.fqn STARTS WITH 'javax.smartcardio') THEN true
+             WHEN (t.fqn STARTS WITH 'javax.sound.') THEN true
+             WHEN (t.fqn STARTS WITH 'javax.sql') THEN true
+             WHEN (t.fqn STARTS WITH 'javax.swing') THEN true
+             WHEN (t.fqn STARTS WITH 'javax.tools') THEN true
+             WHEN (t.fqn STARTS WITH 'javax.transaction.xa') THEN true
+             WHEN (t.fqn STARTS WITH 'javax.xml') THEN true
+             WHEN (t.fqn STARTS WITH 'jdk.') THEN true
+             WHEN (t.fqn STARTS WITH 'netscape.javascript') THEN true
+             WHEN (t.fqn STARTS WITH 'org.ietf.jgss') THEN true
+             WHEN (t.fqn STARTS WITH 'org.w3c.dom') THEN true
+             WHEN (t.fqn STARTS WITH 'org.xml.sax') THEN true    
+            ELSE false
+        END AS isJavaType                                          
+  WHERE isJavaType
+    SET t:JavaType 
+ RETURN labels(t), count(t) as numberOfJavaTypes
diff --git a/cypher/Types/Label_resolved_duplicate_types.cypher b/cypher/Types/Label_resolved_duplicate_types.cypher
@@ -0,0 +1,7 @@
+// Label resolved duplicate types.
+
+  MATCH (type:Type)
+   WITH type, exists((type)-[:RESOLVES_TO]->(:Type)) AS isResolvedDuplicate
+  WHERE isResolvedDuplicate
+    SET type:ResolvedDuplicateType
+ RETURN labels(type), count(type) as numberOfResolvedDuplicateTypes
diff --git a/cypher/Types/Remove_extended_type_labels.cypher b/cypher/Types/Remove_extended_type_labels.cypher
@@ -0,0 +1,4 @@
+// Remove external type and annotation labels
+
+  MATCH (extendedLabeledType:JavaType|PrimitiveType|Void|ResolvedDuplicateType)
+ REMOVE extendedLabeledType:JavaType:PrimitiveType:Void:ResolvedDuplicateType
diff --git a/jupyter/InternalDependencies.ipynb b/jupyter/InternalDependencies.ipynb
diff --git a/scripts/configuration/template-neo4jv4-jqassistant.yaml b/scripts/configuration/template-neo4jv4-jqassistant.yaml
diff --git a/scripts/configuration/template-neo4jv5-jqassistant.yaml b/scripts/configuration/template-neo4jv5-jqassistant.yaml
diff --git a/scripts/prepareAnalysis.sh b/scripts/prepareAnalysis.sh
diff --git a/scripts/reports/ArtifactDependenciesCsv.sh b/scripts/reports/ArtifactDependenciesCsv.sh
diff --git a/scripts/reports/CentralityCsv.sh b/scripts/reports/CentralityCsv.sh
diff --git a/scripts/reports/CommunityCsv.sh b/scripts/reports/CommunityCsv.sh
diff --git a/scripts/reports/NodeEmbeddingsCsv.sh b/scripts/reports/NodeEmbeddingsCsv.sh
diff --git a/scripts/reports/SimilarityCsv.sh b/scripts/reports/SimilarityCsv.sh
