Datasets:
ArXiv:
License:
| /* | |
| * To change this template, choose Tools | Templates | |
| * and open the template in the editor. | |
| */ | |
| package tool; | |
| import tool.sieves.AffixationSieve; | |
| import tool.sieves.CompoundPhraseSieve; | |
| import tool.sieves.DiseaseModifierSynonymsSieve; | |
| import tool.sieves.HyphenationSieve; | |
| import tool.sieves.PartialMatchNCBISieve; | |
| import tool.sieves.PartialMatchSieve; | |
| import tool.sieves.PrepositionalTransformSieve; | |
| import tool.sieves.Sieve; | |
| import tool.sieves.SimpleNameSieve; | |
| import tool.sieves.StemmingSieve; | |
| import tool.sieves.SymbolReplacementSieve; | |
| import tool.util.Concept; | |
| import tool.util.Terminology; | |
| /** | |
| * | |
| * @author | |
| */ | |
| public class PassAllSievesNormalizer { | |
| public static int maxSieveLevel; | |
| public static boolean pass(Concept concept, int currentSieveLevel) { | |
| //System.out.println(concept.getCui()); | |
| if (!concept.getCui().equals("")) { | |
| concept.setAlternateCuis(Sieve.getAlternateCuis(concept.getCui())); | |
| concept.setNormalizingSieveLevel(currentSieveLevel-1); | |
| System.out.println("Normalizing sieve is " + concept.getNormalizingSieve()); | |
| //Terminology.storeNormalizedConcept(concept); | |
| return false; | |
| } | |
| if (currentSieveLevel > maxSieveLevel) | |
| { | |
| return false; | |
| } | |
| return true; | |
| } | |
| //behavior of this class | |
| public static void applyPassAllSieves(Concept concept) { | |
| int currentSieveLevel = 1; | |
| // This is the sieve for exact match | |
| //match with names in training data | |
| //Sieve 1 | |
| concept.setCui(Sieve.exactMatchSieve(concept.getName())); | |
| if (!pass(concept, ++currentSieveLevel)) | |
| return; | |
| // This is the beginning of the second type of rules | |
| //Sieve 2 | |
| concept.setCui(Sieve.exactMatchSieve(concept.getNameExpansion())); | |
| if (!pass(concept, ++currentSieveLevel)) | |
| //Sieve 3 | |
| concept.setCui(PrepositionalTransformSieve.apply(concept)); | |
| System.out.println("Applied prepositional transforms"); | |
| if (!pass(concept, ++currentSieveLevel)) | |
| //Sieve 4 | |
| concept.setCui(SymbolReplacementSieve.apply(concept)); | |
| System.out.println("Applied Symbol replacement transforms"); | |
| if (!pass(concept, ++currentSieveLevel)) | |
| //Sieve 5 | |
| concept.setCui(HyphenationSieve.apply(concept)); | |
| System.out.println("Applied hyphenation transforms"); | |
| if (!pass(concept, ++currentSieveLevel)) | |
| //Sieve 6 | |
| concept.setCui(AffixationSieve.apply(concept)); | |
| System.out.println("Applied affixation transforms"); | |
| if (!pass(concept, ++currentSieveLevel)) | |
| //Sieve 7 | |
| concept.setCui(DiseaseModifierSynonymsSieve.apply(concept)); | |
| System.out.println("Applied disease modifier transforms"); | |
| if (!pass(concept, ++currentSieveLevel)) | |
| //Sieve 8 | |
| concept.setCui(StemmingSieve.apply(concept)); | |
| System.out.println("Applied stemming transforms"); | |
| if (!pass(concept, ++currentSieveLevel)) | |
| //Sieve 9 | |
| concept.setCui(Main.test_data_dir.toString().contains("ncbi") ? CompoundPhraseSieve.applyNCBI(concept.getName()) : CompoundPhraseSieve.apply(concept.getName())); | |
| System.out.println(concept.getNamesKnowledgeBase()); | |
| if (!pass(concept, ++currentSieveLevel)) { | |
| return; | |
| } | |
| //Sieve 10 - this is the sieve of the partial match | |
| concept.setCui(SimpleNameSieve.apply(concept)); | |
| pass(concept, ++currentSieveLevel); | |
| --currentSieveLevel; | |
| if (!concept.getCui().equals("")) | |
| return; | |
| //Sieve 10 | |
| concept.setCui(Main.test_data_dir.toString().contains("ncbi") ? PartialMatchNCBISieve.apply(concept) : PartialMatchSieve.apply(concept)); | |
| pass(concept, ++currentSieveLevel); | |
| System.out.println("xxxxxx Concept not normalized xxxxxxx"); | |
| } | |
| } | |