Return-Path: X-Original-To: apmail-ctakes-commits-archive@www.apache.org Delivered-To: apmail-ctakes-commits-archive@www.apache.org Received: from mail.apache.org (hermes.apache.org [140.211.11.3]) by minotaur.apache.org (Postfix) with SMTP id 2F5C1108C8 for ; Thu, 6 Feb 2014 22:15:15 +0000 (UTC) Received: (qmail 19711 invoked by uid 500); 6 Feb 2014 22:15:14 -0000 Delivered-To: apmail-ctakes-commits-archive@ctakes.apache.org Received: (qmail 19674 invoked by uid 500); 6 Feb 2014 22:15:14 -0000 Mailing-List: contact commits-help@ctakes.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Reply-To: dev@ctakes.apache.org Delivered-To: mailing list commits@ctakes.apache.org Received: (qmail 19667 invoked by uid 99); 6 Feb 2014 22:15:14 -0000 Received: from nike.apache.org (HELO nike.apache.org) (192.87.106.230) by apache.org (qpsmtpd/0.29) with ESMTP; Thu, 06 Feb 2014 22:15:14 +0000 X-ASF-Spam-Status: No, hits=-2000.0 required=5.0 tests=ALL_TRUSTED X-Spam-Check-By: apache.org Received: from [140.211.11.4] (HELO eris.apache.org) (140.211.11.4) by apache.org (qpsmtpd/0.29) with ESMTP; Thu, 06 Feb 2014 22:15:11 +0000 Received: from eris.apache.org (localhost [127.0.0.1]) by eris.apache.org (Postfix) with ESMTP id CCF61238883D; Thu, 6 Feb 2014 22:14:50 +0000 (UTC) Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Subject: svn commit: r1565461 - /ctakes/trunk/ctakes-assertion/src/main/java/org/apache/ctakes/assertion/util/AssertionTreeUtils.java Date: Thu, 06 Feb 2014 22:14:50 -0000 To: commits@ctakes.apache.org From: tmill@apache.org X-Mailer: svnmailer-1.0.9 Message-Id: <20140206221450.CCF61238883D@eris.apache.org> X-Virus-Checked: Checked by ClamAV on apache.org Author: tmill Date: Thu Feb 6 22:14:50 2014 New Revision: 1565461 URL: http://svn.apache.org/r1565461 Log: CTAKES-94: Fix to add semantics to dependency tree fragment representation. Modified: ctakes/trunk/ctakes-assertion/src/main/java/org/apache/ctakes/assertion/util/AssertionTreeUtils.java Modified: ctakes/trunk/ctakes-assertion/src/main/java/org/apache/ctakes/assertion/util/AssertionTreeUtils.java URL: http://svn.apache.org/viewvc/ctakes/trunk/ctakes-assertion/src/main/java/org/apache/ctakes/assertion/util/AssertionTreeUtils.java?rev=1565461&r1=1565460&r2=1565461&view=diff ============================================================================== --- ctakes/trunk/ctakes-assertion/src/main/java/org/apache/ctakes/assertion/util/AssertionTreeUtils.java (original) +++ ctakes/trunk/ctakes-assertion/src/main/java/org/apache/ctakes/assertion/util/AssertionTreeUtils.java Thu Feb 6 22:14:50 2014 @@ -16,6 +16,27 @@ import org.apache.uima.jcas.tcas.Annotat public class AssertionTreeUtils { + public static final SimpleTree NULL_TREE = SimpleTree.fromString("(S (TOK nullparse))"); + + public static SimpleTree extractFeatureTree(JCas jcas, Annotation mention, SemanticClasses sems){ + SimpleTree tree = null; + TopTreebankNode annotationTree = AnnotationTreeUtils.getAnnotationTree(jcas, mention); + if(annotationTree != null){ + TopTreebankNode root = AnnotationTreeUtils.getTreeCopy(jcas, annotationTree); + AnnotationTreeUtils.insertAnnotationNode(jcas, root, mention, "CONCEPT"); + tree = TreeExtractor.getSimpleClone(root); + }else{ + tree = NULL_TREE; + } + + TreeExtractor.lowercaseWords(tree); + if(sems != null){ + replaceWordsWithSemanticClasses(tree, sems); + } + + return tree; + } + public static SimpleTree extractAboveLeftConceptTree(JCas jcas, Annotation mention, SemanticClasses sems){ SimpleTree tree = null; TopTreebankNode annotationTree = AnnotationTreeUtils.getAnnotationTree(jcas, mention); @@ -35,7 +56,7 @@ public class AssertionTreeUtils { tree = TreeExtractor.getSimpleClone(node); }else{ - tree = SimpleTree.fromString("(S noparse)"); + tree = NULL_TREE; } TreeExtractor.lowercaseWords(tree); @@ -145,6 +166,22 @@ public class AssertionTreeUtils { } } + public static void replaceDependencyWordsWithSemanticClasses(SimpleTree tree, + SemanticClasses sems) { + + // same but don't need to check for leaf node + for(Map.Entry> semClass : sems.entrySet()){ + if(semClass.getValue().contains(tree.cat)){ + tree.cat = "semclass_" + semClass.getKey(); + } + } + + // now iterate over children + for(SimpleTree child : tree.children){ + replaceDependencyWordsWithSemanticClasses(child, sems); + } + } + static HashMap wordMap = new HashMap(); static Random random = new Random(); public void randomizeWords(SimpleTree tree, boolean dep) { @@ -170,4 +207,5 @@ public class AssertionTreeUtils { } } + }