List of usage examples for opennlp.tools.parser Parse parseParse
public static Parse parseParse(String parse)
From source file:es.ehu.si.ixa.pipe.parse.Annotate.java
/** * Takes as input a list of parse strings, one for line, and annotates the * headwords/* w ww .j a v a 2s.c om*/ * * @param inputTrees * @return a list of parse trees with headwords annotated */ private String addHeadWordsToTreebank(List<String> inputTrees) { StringBuffer parsedDoc = new StringBuffer(); for (String parseSent : inputTrees) { Parse parsedSentence = Parse.parseParse(parseSent); headFinder.printHeads(parsedSentence); parsedSentence.show(parsedDoc); parsedDoc.append("\n"); } return parsedDoc.toString(); }
From source file:es.ehu.si.ixa.pipe.convert.Convert.java
/** * Reads a list of Parse trees and calls {@code getTokens} to create tokenized * oneline text.//from www .j a v a 2 s. c o m * * @param inputTrees * the list of trees in penn treebank format * @return the tokenized document the document tokens */ private String getTokensFromTree(List<String> inputTrees) { StringBuilder parsedDoc = new StringBuilder(); for (String parseSent : inputTrees) { Parse parse = Parse.parseParse(parseSent); StringBuilder sentBuilder = new StringBuilder(); getTokens(parse, sentBuilder); parsedDoc.append(sentBuilder.toString()).append("\n"); } return parsedDoc.toString(); }
From source file:es.ehu.si.ixa.pipe.convert.Convert.java
/** * Reads a list of Parse trees and calls {@code getWordType} to create POS * training data in Word_POS form//ww w . j a v a2s . com * * @param inputTrees * @return the document with Word_POS sentences */ private String getPreTerminals(List<String> inputTrees) { StringBuilder parsedDoc = new StringBuilder(); for (String parseSent : inputTrees) { Parse parse = Parse.parseParse(parseSent); StringBuilder sentBuilder = new StringBuilder(); getWordType(parse, sentBuilder); parsedDoc.append(sentBuilder.toString()).append("\n"); } return parsedDoc.toString(); }
From source file:es.ehu.si.ixa.pipe.convert.Convert.java
/** * It takes as input a semi-pruned penn treebank tree (e.g., with -NONE- * traces removed) via sed 's/-NONE-\s[\*A-Za-z0-9]*[\*]*[\-]*[A-Za-z0-9]*' * //from www . j a v a2s . c om * and prunes the empty trees remaining from the sed operation. The parseParse * function also removes function tags by default. * * @param inputTrees * @return */ // TODO add the sed regexp to this function private String normalizeParse(List<String> inputTrees) { StringBuilder parsedDoc = new StringBuilder(); for (String parseSent : inputTrees) { Parse parse = Parse.parseParse(parseSent); Parse.pruneParse(parse); StringBuffer sentBuilder = new StringBuffer(); parse.show(sentBuilder); parsedDoc.append(sentBuilder.toString()).append("\n"); } return parsedDoc.toString(); }