From: <ku...@us...> - 2008-09-30 12:11:46
|
Revision: 1318 http://dl-learner.svn.sourceforge.net/dl-learner/?rev=1318&view=rev Author: kurzum Date: 2008-09-30 12:11:29 +0000 (Tue, 30 Sep 2008) Log Message: ----------- server commit Modified Paths: -------------- trunk/src/dl-learner/org/dllearner/scripts/SemanticBibleComparison.java trunk/src/dl-learner/org/dllearner/scripts/WikipediaCategoryCleaner.java trunk/src/dl-learner/org/dllearner/scripts/improveWikipedia/WikipediaCategoryTasks.java Modified: trunk/src/dl-learner/org/dllearner/scripts/SemanticBibleComparison.java =================================================================== --- trunk/src/dl-learner/org/dllearner/scripts/SemanticBibleComparison.java 2008-09-30 10:50:46 UTC (rev 1317) +++ trunk/src/dl-learner/org/dllearner/scripts/SemanticBibleComparison.java 2008-09-30 12:11:29 UTC (rev 1318) @@ -243,6 +243,9 @@ Files.appendFile(jamonlog, "<xmp>\n"+JamonMonitorLogger.getStringForAllSortedByLabel()); } + + + public static void conductExperiment(Experiments exp){ @@ -399,8 +402,9 @@ //defaultSettings: c.setUseHasValueConstructor(false); - c.setUseBooleanDatatypes(true); - c.setUseDoubleDatatypes(true); + c.setUseBooleanDatatypes(false); + c.setUseDoubleDatatypes(false); + if(exp.toString().contains("HASVALUE")){ c.setUseHasValueConstructor(true); Modified: trunk/src/dl-learner/org/dllearner/scripts/WikipediaCategoryCleaner.java =================================================================== --- trunk/src/dl-learner/org/dllearner/scripts/WikipediaCategoryCleaner.java 2008-09-30 10:50:46 UTC (rev 1317) +++ trunk/src/dl-learner/org/dllearner/scripts/WikipediaCategoryCleaner.java 2008-09-30 12:11:29 UTC (rev 1318) @@ -71,7 +71,7 @@ // parameters // used for developing, - private static final boolean DEVELOP = true; + private static final boolean DEVELOPSTABLESETS = true; public static final int SPARQL_RESULTSET_LIMIT = 500; @@ -99,11 +99,37 @@ logger.info("Start"); SortedSet<String> wikipediaCategories = new TreeSet<String>(); + String q = "SELECT * WHERE { ?a <http://www.w3.org/2004/02/skos/core#subject> ?cat }"; + SortedSet<String> s = sparqlTasks.queryAsSet(q, "cat"); + SortedSet<String> results = new TreeSet<String>(); + int i = 0; + for (String category : s) { + System.out.println(""+(i++)+" "+results.size()); + + String q2 = "SELECT DISTINCT ?subject WHERE { ?subject <http://www.w3.org/2004/02/skos/core#subject> <"+category+"> }"; + SortedSet<String> subj = sparqlTasks.queryAsSet(q2, "subject"); + if(40<subj.size() && subj.size()<80){ + results.add(category); + + } + + if(results.size()>200){ + for (String cat : results) { + System.out.println("cat.add(\""+cat+"\");"); + } + System.exit(0); + } + //System.out.println(subj.size() +" "+ string); + } + System.exit(0); String test = "http://dbpedia.org/resource/Category:Prime_Ministers_of_the_United_Kingdom"; wikipediaCategories.add(test); test = "http://dbpedia.org/resource/Category:Best_Actor_Academy_Award_winners"; - wikipediaCategories.add(test); + + // <http://dbpedia.org/resource/Category:Assassinated_monarchs> + // <http://dbpedia.org/resource/Category:Alabama_musicians> + // wikipediaCategories.add(test); for (String target : wikipediaCategories) { @@ -115,7 +141,66 @@ // JamonMonitorLogger.printAllSortedByLabel(); } + + private static SortedSet<String> returnCat (){ + SortedSet<String> cat = new TreeSet<String>(); + + cat.add("http://dbpedia.org/resource/Category:.NET_framework"); + cat.add("http://dbpedia.org/resource/Category:1948_songs"); + cat.add("http://dbpedia.org/resource/Category:1949_songs"); + cat.add("http://dbpedia.org/resource/Category:1951_songs"); + cat.add("http://dbpedia.org/resource/Category:1953_songs"); + cat.add("http://dbpedia.org/resource/Category:1961_songs"); + cat.add("http://dbpedia.org/resource/Category:1970s_pop_songs"); + cat.add("http://dbpedia.org/resource/Category:1991_introductions"); + cat.add("http://dbpedia.org/resource/Category:1993_introductions"); + cat.add("http://dbpedia.org/resource/Category:1995_introductions"); + cat.add("http://dbpedia.org/resource/Category:2001_television_films"); + cat.add("http://dbpedia.org/resource/Category:2008_establishments"); + cat.add("http://dbpedia.org/resource/Category:3-manifolds"); + cat.add("http://dbpedia.org/resource/Category:Al-Qaeda_activities"); + cat.add("http://dbpedia.org/resource/Category:Albums_produced_by_Teo_Macero"); + cat.add("http://dbpedia.org/resource/Category:American_accordionists"); + cat.add("http://dbpedia.org/resource/Category:American_comedy_musicians"); + cat.add("http://dbpedia.org/resource/Category:American_entertainers"); + cat.add("http://dbpedia.org/resource/Category:Apollo_asteroids"); + cat.add("http://dbpedia.org/resource/Category:Assassinated_monarchs"); + cat.add("http://dbpedia.org/resource/Category:Ayumi_Hamasaki_songs"); + cat.add("http://dbpedia.org/resource/Category:Best_Song_Academy_Award_winning_songs"); + cat.add("http://dbpedia.org/resource/Category:Books_about_film"); + cat.add("http://dbpedia.org/resource/Category:Brian_Eno_albums"); + cat.add("http://dbpedia.org/resource/Category:British_military_personnel_killed_in_action"); + cat.add("http://dbpedia.org/resource/Category:British_military_personnel_of_the_Falklands_War"); + cat.add("http://dbpedia.org/resource/Category:CENTR_members"); + cat.add("http://dbpedia.org/resource/Category:Companies_of_Finland"); + cat.add("http://dbpedia.org/resource/Category:Computing_platforms"); + cat.add("http://dbpedia.org/resource/Category:Coordination_compounds"); + cat.add("http://dbpedia.org/resource/Category:Dance-punk_musical_groups"); + cat.add("http://dbpedia.org/resource/Category:Deathgrind_musical_groups"); + cat.add("http://dbpedia.org/resource/Category:Disney_Channel_original_films"); + cat.add("http://dbpedia.org/resource/Category:Dutch_Eurovision_songs"); + cat.add("http://dbpedia.org/resource/Category:Executed_royalty"); + cat.add("http://dbpedia.org/resource/Category:Films_based_on_Stephen_King%27s_works"); + cat.add("http://dbpedia.org/resource/Category:First_Nations_history"); + cat.add("http://dbpedia.org/resource/Category:Fluorescent_dyes"); + cat.add("http://dbpedia.org/resource/Category:ForeFront_Records_albums"); + cat.add("http://dbpedia.org/resource/Category:Former_municipalities_of_Utrecht_%28province%29"); + cat.add("http://dbpedia.org/resource/Category:Fred_Astaire_songs"); + cat.add("http://dbpedia.org/resource/Category:Home_computer_magazines"); + cat.add("http://dbpedia.org/resource/Category:Honolulu_County%2C_Hawaii"); + cat.add("http://dbpedia.org/resource/Category:House_of_Hashim"); + cat.add("http://dbpedia.org/resource/Category:Hugo_Award_Winner_for_Best_Short_Story"); + cat.add("http://dbpedia.org/resource/Category:Irish_folk_songs"); + cat.add("http://dbpedia.org/resource/Category:Islands_of_Tonga"); + cat.add("http://dbpedia.org/resource/Category:James_Bond"); + cat.add("http://dbpedia.org/resource/Category:Jason_Nevins_remixes"); + cat.add("http://dbpedia.org/resource/Category:Jay-Z_songs"); + cat.add("http://dbpedia.org/resource/Category:Jo_Stafford_songs"); + + return cat; + } + private static void doit(String target) { List<EvaluatedDescription> conceptresults; SortedSet<String> currentPOSITIVEex = new TreeSet<String>(); @@ -131,7 +216,7 @@ // PHASE 1 ************* wikiTasks.makeInitialExamples(target, PERCENT_OF_SKOSSET, NEGFACTOR, - SPARQL_RESULTSET_LIMIT, DEVELOP); + SPARQL_RESULTSET_LIMIT, DEVELOPSTABLESETS); currentPOSITIVEex.addAll(wikiTasks.getPosExamples()); currentNEGATIVEex.addAll(wikiTasks.getNegExamples()); ExampleBasedROLComponent la = learn(currentPOSITIVEex, currentNEGATIVEex); Modified: trunk/src/dl-learner/org/dllearner/scripts/improveWikipedia/WikipediaCategoryTasks.java =================================================================== --- trunk/src/dl-learner/org/dllearner/scripts/improveWikipedia/WikipediaCategoryTasks.java 2008-09-30 10:50:46 UTC (rev 1317) +++ trunk/src/dl-learner/org/dllearner/scripts/improveWikipedia/WikipediaCategoryTasks.java 2008-09-30 12:11:29 UTC (rev 1318) @@ -128,7 +128,7 @@ */ public void makeInitialExamples(String targetCategory, double percentOfSKOSSet, double negFactor, int sparqlResultLimit, - boolean develop) { + boolean stable) { fullPositiveSet.clear(); // fullPosSetWithoutPosExamples.clear(); posExamples.clear(); @@ -153,7 +153,7 @@ aneg.makeNegativeExamplesFromParallelClasses(posExamples, sparqlResultLimit); - negExamples = aneg.getNegativeExamples(neglimit, develop); + negExamples = aneg.getNegativeExamples(neglimit, stable); logger.debug("POSITIVE EXAMPLES"); for (String pos : posExamples) { This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site. |