Cntd. QTL eval.

SmartDataAnalytics · Sep 23, 2016 · 1c3ed56 · 1c3ed56
1 parent 1f7968f
commit 1c3ed56
Show file tree

Hide file tree

Showing 6 changed files with 270 additions and 50 deletions.
diff --git a/components-core/pom.xml b/components-core/pom.xml
@@ -454,6 +454,21 @@
 			<groupId>org.apache.commons</groupId>
 			<artifactId>commons-math3</artifactId>
 		</dependency>
+
+		<!-- https://mvnrepository.com/artifact/org.jgrapht/jgrapht-ext -->
+		<dependency>
+			<groupId>org.jgrapht</groupId>
+			<artifactId>jgrapht-ext</artifactId>
+			<version>0.9.2</version>
+		</dependency>
+
+		<!-- https://mvnrepository.com/artifact/org.tinyjee.jgraphx/jgraphx -->
+		<dependency>
+			<groupId>org.tinyjee.jgraphx</groupId>
+			<artifactId>jgraphx</artifactId>
+			<version>3.4.1.3</version>
+		</dependency>
+
 		<!--
                 <dependency>
                     <groupId>org.apache.commons</groupId>

diff --git a/components-core/src/main/java/org/dllearner/utilities/QueryUtils.java b/components-core/src/main/java/org/dllearner/utilities/QueryUtils.java
@@ -21,6 +21,15 @@
 import com.google.common.collect.HashMultimap;
 import com.google.common.collect.Multimap;
 import com.google.common.collect.Sets;
+import com.mxgraph.layout.hierarchical.mxHierarchicalLayout;
+import com.mxgraph.layout.mxCircleLayout;
+import com.mxgraph.layout.mxOrganicLayout;
+import com.mxgraph.swing.mxGraphComponent;
+import com.mxgraph.util.mxCellRenderer;
+import com.mxgraph.util.mxConstants;
+import com.mxgraph.util.png.mxPngEncodeParam;
+import com.mxgraph.util.png.mxPngImageEncoder;
+import com.mxgraph.view.mxStylesheet;
 import org.apache.jena.graph.Node;
 import org.apache.jena.graph.Triple;
 import org.apache.jena.query.*;
@@ -31,9 +40,22 @@
 import org.apache.jena.vocabulary.RDF;
 import org.aksw.jena_sparql_api.core.QueryExecutionFactory;
 import org.apache.commons.collections15.ListUtils;
+import org.jgraph.JGraph;
+import org.jgrapht.DirectedGraph;
+import org.jgrapht.ext.JGraphModelAdapter;
+import org.jgrapht.ext.JGraphXAdapter;
+import org.jgrapht.graph.DefaultDirectedGraph;
+import org.jgrapht.graph.DefaultEdge;
 import org.slf4j.Logger;
 import org.slf4j.LoggerFactory;
 
+import javax.imageio.ImageIO;
+import javax.swing.*;
+import java.awt.*;
+import java.awt.image.BufferedImage;
+import java.io.File;
+import java.io.FileOutputStream;
+import java.io.IOException;
 import java.util.*;
 import java.util.Map.Entry;
 
@@ -667,6 +689,77 @@ public void filterOutGeneralTypes(QueryExecutionFactory qef, Query query){
 			triplePatterns.removeAll(triplesPatterns2Remove);
 		}
 	}
+
+	public static DirectedGraph<Node, LabeledEdge> asJGraphT(Query query) {
+		QueryUtils utils = new QueryUtils();
+
+		Set<Triple> tps = utils.extractTriplePattern(query);
+
+		DirectedGraph<Node, LabeledEdge> g = new DefaultDirectedGraph<Node, LabeledEdge>(LabeledEdge.class);
+
+		tps.forEach(tp -> {
+			g.addVertex(tp.getSubject());
+			g.addVertex(tp.getObject());
+			g.addEdge(tp.getSubject(), tp.getObject(), new LabeledEdge(tp.getSubject(), tp.getObject(), tp.getPredicate()));
+		});
+
+		return g;
+	}
+
+	public static void exportAsGraph(Query query, File file) {
+		DirectedGraph<Node, LabeledEdge> g = asJGraphT(query);
+		System.out.println(g.edgeSet().size());
+
+		JGraphXAdapter adapter = new JGraphXAdapter(g);
+
+
+		// positioning via jgraphx layouts
+		mxHierarchicalLayout layout = new mxHierarchicalLayout(adapter);
+		layout.execute(adapter.getDefaultParent());
+
+		Map<String, Object> edgeStyle = new HashMap<String, Object>();
+//edgeStyle.put(mxConstants.STYLE_EDGE, mxConstants.EDGESTYLE_ORTHOGONAL);
+		edgeStyle.put(mxConstants.STYLE_SHAPE,    mxConstants.SHAPE_CONNECTOR);
+		edgeStyle.put(mxConstants.STYLE_ENDARROW, mxConstants.ARROW_CLASSIC);
+		edgeStyle.put(mxConstants.STYLE_STROKECOLOR, "#000000");
+		edgeStyle.put(mxConstants.STYLE_FONTCOLOR, "#000000");
+		edgeStyle.put(mxConstants.STYLE_LABEL_BACKGROUNDCOLOR, "#ffffff");
+
+		Map<String, Object> nodeStyle = new HashMap<>();
+		nodeStyle.put(mxConstants.STYLE_SHAPE,    mxConstants.SHAPE_ELLIPSE);
+		nodeStyle.put(mxConstants.STYLE_VERTICAL_ALIGN, mxConstants.ALIGN_MIDDLE);
+
+		mxStylesheet stylesheet = new mxStylesheet();
+		stylesheet.setDefaultEdgeStyle(edgeStyle);
+		stylesheet.setDefaultVertexStyle(nodeStyle);
+
+		adapter.setStylesheet(stylesheet);
+
+//		JFrame frame = new JFrame();
+//		frame.getContentPane().add(new mxGraphComponent(adapter));
+//		frame.pack();
+//		frame.setDefaultCloseOperation(WindowConstants.DISPOSE_ON_CLOSE);
+//		frame.setVisible(true);
+
+
+
+		BufferedImage image = mxCellRenderer.createBufferedImage(adapter, null, 1, Color.WHITE, true, null);
+		mxPngEncodeParam param = mxPngEncodeParam.getDefaultEncodeParam(image);
+
+
+		try {
+			FileOutputStream outputStream = new FileOutputStream(file);
+			mxPngImageEncoder encoder = new mxPngImageEncoder(outputStream, param);
+			if (image != null) {
+				encoder.encode(image);
+			}
+			outputStream.close();
+//			ImageIO.write(image, "PNG", file);
+		} catch (IOException e) {
+			e.printStackTrace();
+		}
+
+	}
 
 	private Set<Node> getSuperClasses(QueryExecutionFactory qef, Node cls){
 		Set<Node> superClasses = new HashSet<>();
@@ -778,6 +871,27 @@ public int getFilterCount() {
 	public ElementGroup getElementGroup(Triple triple){
 		return triple2Parent.get(triple);
 	}
+
+	static class LabeledEdge extends DefaultEdge {
+		private final Node s;
+		private final Node t;
+		private final Node edge;
+
+		public LabeledEdge(Node s, Node t, Node edge) {
+			this.s = s;
+			this.t = t;
+			this.edge = edge;
+		}
+
+		public Node getEdge() {
+			return edge;
+		}
+
+		@Override
+		public String toString() {
+			return edge.toString();
+		}
+	}
 
 	public static void main(String[] args) throws Exception {
 		Query q = QueryFactory.create(

diff --git a/...test/java/org/dllearner/algorithms/qtl/experiments/BenchmarkDescriptionGeneratorHTML.java b/...test/java/org/dllearner/algorithms/qtl/experiments/BenchmarkDescriptionGeneratorHTML.java
@@ -34,6 +34,7 @@
 import org.apache.jena.query.QueryFactory;
 import org.apache.jena.query.ResultSet;
 import org.apache.jena.sparql.core.Var;
+import org.dllearner.utilities.QueryUtils;
 
 /**
  * @author Lorenz Buehmann
@@ -98,7 +99,10 @@ public class BenchmarkDescriptionGeneratorHTML {
 			+ "table {\n" + 
 			"    border-collapse: separate;\n" + 
 			"    border-spacing: 0 5px;\n" + 
-			"}\n" + 
+			"}\n" +
+					"table th {\n" +
+					"    width: auto !important;\n" +
+					"}" +
 			"\n" + 
 			"thead th {\n" + 
 			"    background-color: #006DCC;\n" + 
@@ -165,37 +169,53 @@ public void generateBenchmarkDescription(File benchmarkQueriesFile, File htmlOut
 		html += "<thead><tr>"
 				+ "<th data-sortable=\"true\" data-valign='middle'>ID</th>"
 				+ "<th data-sortable=\"true\" data-valign='middle'>Query</th>"
+				+ "<th data-sortable=\"true\" data-valign='middle'>Query Type</th>"
 				+ "<th data-align=\"right\" data-sortable=\"true\" data-valign='middle'>Depth</th>"
 				+ "<th data-align=\"right\" data-sortable=\"true\" data-valign='middle'>#Instances</th>"
 				+ "</tr></thead>\n";
 
 		html += "<tbody>\n";
 		int id = 1;
 		for (Query query : queries) {
+			query.getPrefixMapping().removeNsPrefix("owl");
+			query.getPrefixMapping().removeNsPrefix("rdfs");
+			query.getPrefixMapping().removeNsPrefix("foaf");
+			query.getPrefixMapping().removeNsPrefix("rdf");
+
+			if(query.toString().contains("http://dbpedia.org/ontology/")) {
+				query.getPrefixMapping().setNsPrefix("dbo", "http://dbpedia.org/ontology/");
+			}
+			if(query.toString().contains("http://dbpedia.org/property/")) {
+				query.getPrefixMapping().setNsPrefix("dbp", "http://dbpedia.org/property/");
+			}
+			if(query.toString().contains("http://xmlns.com/foaf/0.1/")) {
+				query.getPrefixMapping().setNsPrefix("foaf", "http://xmlns.com/foaf/0.1/");
+			}
+			QueryUtils.exportAsGraph(query, new File("/tmp/test.png"));
+
+
+			query.setBaseURI("http://dbpedia.org/resource/");
 			html += "<tr>\n";
 
 			// 1. column: ID
 			html += "<td>" + id++ + "</td>\n";
 
 			// 2. column: SPARQL query
 			html += "<td><pre>" + query.toString().replace("<", "&lt;").replace(">", "&gt;") + "</pre></td>\n";
+
+			// 3. column: SPARQL query type
+			html += "<td>" + SPARQLUtils.getQueryType(query) + "</td>\n";
 
-			// 3. column: depth
+			// 4. column: depth
 			int depth = org.dllearner.utilities.QueryUtils.getSubjectObjectJoinDepth(query, var) + 1;
 			html += "<td class='number'>" + depth + "</td>\n";
 
-			// 4. column: #instances
-			int nrOfInstances = 0;
-			QueryExecution qe = qef.createQueryExecution(query);
-			ResultSet rs = qe.execSelect();
-			while(rs.hasNext()) {
-				rs.next();
-				nrOfInstances++;
-			}
-			qe.close();
+			// 5. column: #instances
+			int nrOfInstances = SPARQLUtils.getResult(qef, query).size();
 			html += "<td class='number'>" + nrOfInstances + "</td>\n";
 
 			html += "</tr>\n";
+			break;
 		}
 		html += "</tbody>\n";
 		html += "</table>\n";
@@ -210,14 +230,16 @@ public void generateBenchmarkDescription(File benchmarkQueriesFile, File htmlOut
 	}
 
 	public static void main(String[] args) throws Exception{
-		if(args.length < 4) {
+		if(args.length < 3) {
 			System.out.println("Usage: BenchmarkDescriptionGeneratorHTML <source> <target> <endpointURL> <defaultGraphURI>");
 			System.exit(0);
 		}
 		File source = new File(args[0]);
 		File target = new File(args[1]);
 		String endpointURL = args[2];
-		String defaultGraph = args[3];
+		String defaultGraph = null;
+		if(args.length == 4)
+			defaultGraph = args[3];
 
 		QueryExecutionFactory qef = new QueryExecutionFactoryHttp(endpointURL, defaultGraph);
 		qef = new QueryExecutionFactoryPaginated(qef);

diff --git a/...onents-core/src/test/java/org/dllearner/algorithms/qtl/experiments/EvaluationDataset.java b/...onents-core/src/test/java/org/dllearner/algorithms/qtl/experiments/EvaluationDataset.java
@@ -18,9 +18,11 @@
  */
 package org.dllearner.algorithms.qtl.experiments;
 
+import com.google.common.base.Joiner;
 import com.jamonapi.Monitor;
 import com.jamonapi.MonitorFactory;
 import org.apache.commons.math3.stat.descriptive.DescriptiveStatistics;
+import org.apache.jena.graph.Triple;
 import org.apache.jena.query.QueryFactory;
 import org.apache.jena.rdf.model.Model;
 import org.apache.jena.rdf.model.Statement;
@@ -29,10 +31,15 @@
 import org.dllearner.kb.SparqlEndpointKS;
 import org.dllearner.kb.sparql.ConciseBoundedDescriptionGenerator;
 import org.dllearner.kb.sparql.SymmetricConciseBoundedDescriptionGeneratorImpl;
+import org.dllearner.utilities.QueryUtils;
 
+import java.io.File;
+import java.io.IOException;
+import java.nio.file.Files;
 import java.util.ArrayList;
 import java.util.Collections;
 import java.util.List;
+import java.util.Set;
 import java.util.function.Predicate;
 import java.util.stream.Collectors;
 
@@ -88,48 +95,62 @@ public List<Predicate<Statement>> getQueryTreeFilters() {
 		return queryTreeFilters;
 	}
 
+	/**
+	 * Writes the SPARQL queries line-wise to file.
+	 *
+	 * @param file the file
+	 */
+	public void saveToDisk(File file) throws IOException {
+		Files.write(file.toPath(), sparqlQueries.stream().map(q -> q.replace("\n", " ")).collect(Collectors.toList()));
+	}
+
 	public void analyze() {
 		ConciseBoundedDescriptionGenerator cbdGen = new SymmetricConciseBoundedDescriptionGeneratorImpl(ks.getQueryExecutionFactory());
 
-		String s = sparqlQueries.stream().map(QueryFactory::create).map(q -> {
+		String tsv = sparqlQueries.stream().map(QueryFactory::create).map(q -> {
 			StringBuilder sb = new StringBuilder();
-			sb.append(q.toString().replace("\n", " ")).append("\t");
+			sb.append(q.toString().replace("\n", " "));
 			try {
 				// get query result
 				List<String> result = SPARQLUtils.getResult(ks.getQueryExecutionFactory(), q);
-				sb.append(result.size());
-
-				// check CBD sizes and time
-				Monitor mon = MonitorFactory.getTimeMonitor("CBD");
-				mon.reset();
-				DescriptiveStatistics sizeStats = new DescriptiveStatistics();
-				result.stream()
-						.map(r -> {
-							System.out.println(r);
-							mon.start();
-							Model cbd = cbdGen.getConciseBoundedDescription(r, 2);
-							mon.stop();
-							return cbd;
-						})
-						.map(Model::size)
-						.forEach(sizeStats::addValue);
-
-				// show min., max. and avg. size
-				sb.append("\t").append(sizeStats.getMin());
-				sb.append("\t").append(sizeStats.getMax());
-				sb.append("\t").append(sizeStats.getMean());
-
-				// show min., max. and avg. CBD time
-				sb.append("\t").append(mon.getTotal());
-				sb.append("\t").append(mon.getMin());
-				sb.append("\t").append(mon.getMax());
-				sb.append("\t").append(mon.getAvg());
+				sb.append("\t").append(result.size());
+
+				// query type
+				SPARQLUtils.QueryType queryType = SPARQLUtils.getQueryType(q);
+				sb.append("\t").append(queryType.name());
+
+
+//				// check CBD sizes and time
+//				Monitor mon = MonitorFactory.getTimeMonitor("CBD");
+//				mon.reset();
+//				DescriptiveStatistics sizeStats = new DescriptiveStatistics();
+//				result.stream()
+//						.map(r -> {
+//							System.out.println(r);
+//							mon.start();
+//							Model cbd = cbdGen.getConciseBoundedDescription(r, 2);
+//							mon.stop();
+//							return cbd;
+//						})
+//						.map(Model::size)
+//						.forEach(sizeStats::addValue);
+//
+//				// show min., max. and avg. size
+//				sb.append("\t").append(sizeStats.getMin());
+//				sb.append("\t").append(sizeStats.getMax());
+//				sb.append("\t").append(sizeStats.getMean());
+//
+//				// show min., max. and avg. CBD time
+//				sb.append("\t").append(mon.getTotal());
+//				sb.append("\t").append(mon.getMin());
+//				sb.append("\t").append(mon.getMax());
+//				sb.append("\t").append(mon.getAvg());
 			} catch (Exception e) {
 				e.printStackTrace();
 			}
 			return sb;
 		}).collect(Collectors.joining("\n"));
 
-		System.out.println(s);
+		System.out.println(tsv);
 	}
 }