From af7abeaf22669a531d1b8fabfef9d8ca83daf65e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?J=C3=A9r=C3=B4me=20Euzenat?= <Jerome.Euzenat@inria.fr>
Date: Sun, 11 Jan 2009 12:50:58 +0000
Subject: [PATCH] - Java 1.5 style for - used ConcatenedIterator iterable -
implemented qq extractor for individuals - printMatrix completed - changed
extractqqNaive into extractqqgreedy
---
.../exmo/align/impl/DistanceAlignment.java | 158 ++++++++++--------
1 file changed, 91 insertions(+), 67 deletions(-)
diff --git a/src/fr/inrialpes/exmo/align/impl/DistanceAlignment.java b/src/fr/inrialpes/exmo/align/impl/DistanceAlignment.java
index cf8def51..e997a3f2 100644
--- a/src/fr/inrialpes/exmo/align/impl/DistanceAlignment.java
+++ b/src/fr/inrialpes/exmo/align/impl/DistanceAlignment.java
@@ -1,7 +1,7 @@
/*
* $Id$
*
- * Copyright (C) INRIA, 2003-2008
+ * Copyright (C) INRIA, 2003-2009
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU Lesser General Public License as published by
@@ -101,12 +101,20 @@ public class DistanceAlignment extends ObjectAlignment implements AlignmentProce
System.out.println("\\documentclass{article}\n");
System.out.println("\\usepackage{graphics}\n");
System.out.println("\\begin{document}\n");
- System.out.println("\\begin{figure}");
+ System.out.println("\\begin{table}");
sim.printClassSimilarityMatrix("tex");
+ System.out.println("\\caption{Class distance with measure "+(String)params.getParameter("stringFunction")+"}");
+ System.out.println("\\end{table}");
System.out.println();
+ System.out.println("\\begin{table}");
sim.printPropertySimilarityMatrix("tex");
- System.out.println("\\caption{Class distance with measure "+(String)params.getParameter("stringFunction")+"}");
- System.out.println("\\end{figure}");
+ System.out.println("\\caption{Property distance with measure "+(String)params.getParameter("stringFunction")+"}");
+ System.out.println("\\end{table}");
+ System.out.println();
+ System.out.println("\\begin{table}");
+ sim.printIndividualSimilarityMatrix("tex");
+ System.out.println("\\caption{Individual distance with measure "+(String)params.getParameter("stringFunction")+"}");
+ System.out.println("\\end{table}");
System.out.println("\n\\end{document}");
}
@@ -145,6 +153,9 @@ public class DistanceAlignment extends ObjectAlignment implements AlignmentProce
else throw new AlignmentException("Unknown alignment type: "+type);
}
+ // JE: It is now certainly possible to virtualise extraction has it has
+ // been done for printing matrix in MatrixMeasure (todo)
+
/**
* Extract the alignment of a ?* type
* Complexity: O(n^2)
@@ -159,15 +170,13 @@ public class DistanceAlignment extends ObjectAlignment implements AlignmentProce
ConcatenatedIterator pit1 = new
ConcatenatedIterator(ontology1().getObjectProperties().iterator(),
ontology1().getDataProperties().iterator());
- for ( ; pit1.hasNext(); ) {
- Object prop1 = pit1.next();
+ for( Object prop1 : pit1 ){
found = false; max = threshold; val = 0;
Object prop2 = null;
ConcatenatedIterator pit2 = new
ConcatenatedIterator(ontology2().getObjectProperties().iterator(),
ontology2().getDataProperties().iterator());
- for ( ; pit2.hasNext(); ) {
- Object current = pit2.next();
+ for ( Object current : pit2 ){
val = 1 - sim.getPropertySimilarity(prop1,current);
if ( val > max) {
found = true; max = val; prop2 = current;
@@ -176,12 +185,10 @@ public class DistanceAlignment extends ObjectAlignment implements AlignmentProce
if ( found ) addAlignCell(prop1,prop2, "=", max);
}
// Extract for classes
- for (Iterator it1 = ontology1().getClasses().iterator(); it1.hasNext(); ) {
- Object class1 = it1.next();
+ for ( Object class1 : ontology1().getClasses() ) {
found = false; max = threshold; val = 0;
Object class2 = null;
- for (Iterator it2 = ontology2().getClasses().iterator(); it2.hasNext(); ) {
- Object current = it2.next();
+ for ( Object current : ontology2().getClasses() ) {
val = 1 - sim.getClassSimilarity(class1,current);
if (val > max) {
found = true; max = val; class2 = current;
@@ -191,13 +198,11 @@ public class DistanceAlignment extends ObjectAlignment implements AlignmentProce
}
// Extract for individuals
if ( params.getParameter("noinst") == null ){
- for (Iterator it1 = ontology1().getIndividuals().iterator(); it1.hasNext();) {
- Object ind1 = it1.next();
+ for ( Object ind1 : ontology1().getIndividuals() ) {
if ( ontology1().getEntityURI( ind1 ) != null ) {
found = false; max = threshold; val = 0;
Object ind2 = null;
- for (Iterator it2 = ontology2().getIndividuals().iterator(); it2.hasNext(); ) {
- Object current = it2.next();
+ for ( Object current : ontology2().getIndividuals() ) {
if ( ontology2().getEntityURI( current ) != null ) {
val = 1 - sim.getIndividualSimilarity( ind1, current );
if (val > max) {
@@ -230,12 +235,12 @@ public class DistanceAlignment extends ObjectAlignment implements AlignmentProce
Object[] class1 = new Object[nbclasses1];
Object[] class2 = new Object[nbclasses2];
int i = 0;
- for (Iterator it1 = ontology1().getClasses().iterator(); it1.hasNext(); i++) {
- class1[i] = it1.next();
+ for ( Object ob : ontology1().getClasses() ) {
+ class1[i++] = ob;
}
int j = 0;
- for (Iterator it2 = ontology2().getClasses().iterator(); it2.hasNext(); j++) {
- class2[j] = it2.next();
+ for ( Object ob : ontology2().getClasses() ) {
+ class2[j++] = ob;
}
for( i = 0; i < nbclasses1; i++ ){
for( j = 0; j < nbclasses2; j++ ){
@@ -267,16 +272,12 @@ public class DistanceAlignment extends ObjectAlignment implements AlignmentProce
ConcatenatedIterator pit1 = new
ConcatenatedIterator(ontology1().getObjectProperties().iterator(),
ontology1().getDataProperties().iterator());
- for (; pit1.hasNext(); i++) {
- prop1[i] = pit1.next();
- }
+ for ( Object ob: pit1 ) prop1[i++] = ob;
int j = 0;
ConcatenatedIterator pit2 = new
ConcatenatedIterator(ontology2().getObjectProperties().iterator(),
ontology2().getDataProperties().iterator());
- for (; pit2.hasNext(); j++) {
- prop2[j] = pit2.next();
- }
+ for ( Object ob: pit2 ) prop2[j++] = ob;
for( i = 0; i < nbprop1; i++ ){
for( j = 0; j < nbprop2; j++ ){
matrix[i][j] = 1 - sim.getPropertySimilarity(prop1[i],prop2[j]);
@@ -296,41 +297,70 @@ public class DistanceAlignment extends ObjectAlignment implements AlignmentProce
}
}
} catch (AlignmentException alex) { alex.printStackTrace(); }
+ // For individuals
+ if ( params.getParameter("noinst") == null ){
+ try{
+ // Create individual lists
+ Object[] ind1 = new Object[ontology1().nbIndividuals()];
+ Object[] ind2 = new Object[ontology2().nbIndividuals()];
+ int nbind1 = 0;
+ int nbind2 = 0;
+ for( Object ob : ontology2().getIndividuals() ){
+ // We suppress anonymous individuals... this is not legitimate
+ if ( ontology2().getEntityURI( ob ) != null ) {
+ ind2[nbind2++] = ob;
+ }
+ }
+ for( Object ob : ontology1().getIndividuals() ){
+ // We suppress anonymous individuals... this is not legitimate
+ if ( ontology1().getEntityURI( ob ) != null ) {
+ ind1[nbind1++] = ob;
+ }
+ }
+ double[][] matrix = new double[nbind1][nbind2];
+ int i, j;
+ for( i=0; i < nbind1; i++ ){
+ for( j=0; j < nbind2; j++ ){
+ matrix[i][j] = 1 - sim.getIndividualSimilarity(ind1[i],ind2[j]);
+ }
+ }
+ // Pass it to the algorithm
+ int[][] result = HungarianAlgorithm.hgAlgorithm( matrix, "max" );
+ // Extract the result
+ for( i=0; i < result.length ; i++ ){
+ // The matrix has been destroyed
+ double val = 1 - sim.getIndividualSimilarity(ind1[result[i][0]],ind2[result[i][1]]);
+ // JE: here using strict-> is a very good idea.
+ // it means that alignments with 0. similarity
+ // will be excluded from the best match.
+ if( val > threshold ){
+ addCell( new ObjectCell( (String)null, ind1[result[i][0]], ind2[result[i][1]], BasicRelation.createRelation("="), val ) );
+ }
+ }
+ } catch (AlignmentException alex) { alex.printStackTrace(); }
+ }
return((Alignment)this);
}
/**
- * Naive algorithm:
+ * Greedy algorithm:
* 1) dump the part of the matrix distance above threshold in a sorted set
* 2) traverse the sorted set and each time a correspondence involving two
* entities that have no correspondence is encountered, add it to the
* alignment.
* Complexity: O(n^2.logn)
- * Pitfall: no global optimality is warranted
+ * Pitfall: no global optimality is warranted, nor stable marriage
* for instance if there is the following matrix:
* (a,a')=1., (a,b')=.9, (b,a')=.9, (b,b')=.1
* This algorithm will select the first and last correspondances of
* overall similarity 1.1, while the optimum is the second solution
* with overall of 1.8.
*/
- public Alignment extractqqNaive( double threshold, Parameters params) {
- Object ent1=null, ent2=null;
+ public Alignment extractqqgreedy( double threshold, Parameters params) {
double val = 0;
//TreeSet could be replaced by something else
//The comparator must always tell that things are different!
- /*SortedSet cellSet = new TreeSet(
- new Comparator() {
- public int compare( Object o1, Object o2 )
- throws ClassCastException{
- if ( o1 instanceof Cell
- && o2 instanceof Cell ) {
- if ( ((Cell)o1).getStrength() > ((Cell)o2).getStrength() ){
- return -1;
- } else { return 1; }
- } else {
- throw new ClassCastException();
- }}});*/
- SortedSet<Cell> cellSet = new TreeSet<Cell>(
+ SortedSet<Cell> cellSet = new TreeSet<Cell>(
new Comparator<Cell>() {
public int compare( Cell o1, Cell o2 )
throws ClassCastException{
@@ -359,45 +389,40 @@ public class DistanceAlignment extends ObjectAlignment implements AlignmentProce
}
}
);
-
try {
// Get all the matrix above threshold in the SortedSet
// Plus a map from the objects to the cells
// O(n^2.log n)
+ // for classes
+ for ( Object ent1: ontology1().getClasses() ) {
+ for ( Object ent2: ontology2().getClasses() ) {
+ val = 1 - sim.getClassSimilarity( ent1, ent2 );
+ if ( val > threshold ){
+ cellSet.add( new ObjectCell( (String)null, ent1, ent2, BasicRelation.createRelation("="), val ) );
+ }
+ }
+ }
+ // for properties
ConcatenatedIterator pit1 = new
ConcatenatedIterator(ontology1().getObjectProperties().iterator(),
ontology1().getDataProperties().iterator());
- for (; pit1.hasNext(); ) {
- ent1 = pit1.next();
+ for ( Object ent1: pit1 ) {
ConcatenatedIterator pit2 = new
ConcatenatedIterator(ontology2().getObjectProperties().iterator(),
ontology2().getDataProperties().iterator());
- for (; pit2.hasNext(); ) {
- ent2 = pit2.next();
+ for ( Object ent2: pit2 ) {
val = 1 - sim.getPropertySimilarity( ent1, ent2 );
if ( val > threshold ){
cellSet.add( new ObjectCell( (String)null, ent1, ent2, BasicRelation.createRelation("="), val ) );
}
}
}
- for (Iterator it1 = ontology1().getClasses().iterator(); it1.hasNext(); ) {
- ent1 = it1.next();
- for (Iterator it2 = ontology2().getClasses().iterator(); it2.hasNext(); ) {
- ent2 = it2.next();
- val = 1 - sim.getClassSimilarity( ent1, ent2 );
- if ( val > threshold ){
- cellSet.add( new ObjectCell( (String)null, ent1, ent2, BasicRelation.createRelation("="), val ) );
- }
- }
- }
- // Extract for individuals
+ // for individuals
if ( params.getParameter("noinst") == null ){
- for (Iterator it1 = ontology1().getIndividuals().iterator(); it1.hasNext();) {
- ent1 = it1.next();
+ for( Object ent1: ontology1().getIndividuals() ) {
if ( ontology1().getEntityURI( ent1 ) != null ) {
- for (Iterator it2 = ontology2().getIndividuals().iterator(); it2.hasNext(); ) {
- ent2 = it2.next();
+ for( Object ent2: ontology2().getIndividuals() ) {
if ( ontology2().getEntityURI( ent2 ) != null ) {
val = 1 - sim.getIndividualSimilarity( ent1, ent2 );
if ( val > threshold ){
@@ -410,10 +435,9 @@ public class DistanceAlignment extends ObjectAlignment implements AlignmentProce
}
// O(n^2)
- for( Iterator it = cellSet.iterator(); it.hasNext(); ){
- Cell cell = (Cell)it.next();
- ent1 = cell.getObject1();
- ent2 = cell.getObject2();
+ for( Cell cell : cellSet ){
+ Object ent1 = cell.getObject1();
+ Object ent2 = cell.getObject2();
if ( (getAlignCells1( ent1 ) == null) && (getAlignCells2( ent2 ) == null) ){
// The cell is directly added!
addCell( cell );
--
GitLab