Search in sources :

Example 1 with WebDocumentFetcherFilter

use of info.ephyra.answerselection.filters.WebDocumentFetcherFilter in project lucida by claritylab.

the class OpenEphyraCorpus method initFactoidCorpus.

/**
	 * Initializes the pipeline for factoid questions, using a local corpus as a
	 * knowledge source.
	 */
protected void initFactoidCorpus() {
    // question analysis
    Ontology wordNet = new WordNet();
    // - dictionaries for term extraction
    QuestionAnalysis.clearDictionaries();
    QuestionAnalysis.addDictionary(wordNet);
    // - ontologies for term expansion
    QuestionAnalysis.clearOntologies();
    QuestionAnalysis.addOntology(wordNet);
    // query generation
    QueryGeneration.clearQueryGenerators();
    QueryGeneration.addQueryGenerator(new BagOfWordsG());
    QueryGeneration.addQueryGenerator(new BagOfTermsG());
    QueryGeneration.addQueryGenerator(new PredicateG());
    QueryGeneration.addQueryGenerator(new QuestionInterpretationG());
    QueryGeneration.addQueryGenerator(new QuestionReformulationG());
    // search
    // - knowledge miners for unstructured knowledge sources
    Search.clearKnowledgeMiners();
    for (String[] indriIndices : IndriKM.getIndriIndices()) Search.addKnowledgeMiner(new IndriKM(indriIndices, false));
    for (String[] indriServers : IndriKM.getIndriServers()) Search.addKnowledgeMiner(new IndriKM(indriServers, true));
    // - knowledge annotators for (semi-)structured knowledge sources
    Search.clearKnowledgeAnnotators();
    // answer extraction and selection
    // (the filters are applied in this order)
    AnswerSelection.clearFilters();
    // - answer extraction filters
    AnswerSelection.addFilter(new AnswerTypeFilter());
    AnswerSelection.addFilter(new AnswerPatternFilter());
    AnswerSelection.addFilter(new WebDocumentFetcherFilter());
    AnswerSelection.addFilter(new PredicateExtractionFilter());
    AnswerSelection.addFilter(new FactoidsFromPredicatesFilter());
    AnswerSelection.addFilter(new TruncationFilter());
// - answer selection filters
}
Also used : Ontology(info.ephyra.nlp.semantics.ontologies.Ontology) AnswerPatternFilter(info.ephyra.answerselection.filters.AnswerPatternFilter) PredicateExtractionFilter(info.ephyra.answerselection.filters.PredicateExtractionFilter) WebDocumentFetcherFilter(info.ephyra.answerselection.filters.WebDocumentFetcherFilter) IndriKM(info.ephyra.search.searchers.IndriKM) TruncationFilter(info.ephyra.answerselection.filters.TruncationFilter) WordNet(info.ephyra.nlp.semantics.ontologies.WordNet) BagOfWordsG(info.ephyra.querygeneration.generators.BagOfWordsG) PredicateG(info.ephyra.querygeneration.generators.PredicateG) AnswerTypeFilter(info.ephyra.answerselection.filters.AnswerTypeFilter) QuestionReformulationG(info.ephyra.querygeneration.generators.QuestionReformulationG) BagOfTermsG(info.ephyra.querygeneration.generators.BagOfTermsG) QuestionInterpretationG(info.ephyra.querygeneration.generators.QuestionInterpretationG) FactoidsFromPredicatesFilter(info.ephyra.answerselection.filters.FactoidsFromPredicatesFilter)

Example 2 with WebDocumentFetcherFilter

use of info.ephyra.answerselection.filters.WebDocumentFetcherFilter in project lucida by claritylab.

the class OpenEphyraCorpus method initFactoidWeb.

/**
	 * Initializes the pipeline for factoid questions, using the Web as a
	 * knowledge source.
	 * 
	 * @param resultsCorp results retrieved from the corpus
	 */
protected void initFactoidWeb(Result[] resultsCorp) {
    // question analysis
    Ontology wordNet = new WordNet();
    // - dictionaries for term extraction
    QuestionAnalysis.clearDictionaries();
    QuestionAnalysis.addDictionary(wordNet);
    // - ontologies for term expansion
    QuestionAnalysis.clearOntologies();
    QuestionAnalysis.addOntology(wordNet);
    // query generation
    QueryGeneration.clearQueryGenerators();
    QueryGeneration.addQueryGenerator(new BagOfWordsG());
    QueryGeneration.addQueryGenerator(new BagOfTermsG());
    QueryGeneration.addQueryGenerator(new PredicateG());
    QueryGeneration.addQueryGenerator(new QuestionInterpretationG());
    QueryGeneration.addQueryGenerator(new QuestionReformulationG());
    // search
    // - knowledge miners for unstructured knowledge sources
    Search.clearKnowledgeMiners();
    Search.addKnowledgeMiner(new BingKM());
    //		Search.addKnowledgeMiner(new GoogleKM());
    //		Search.addKnowledgeMiner(new YahooKM());
    // - knowledge annotators for (semi-)structured knowledge sources
    Search.clearKnowledgeAnnotators();
    // answer extraction and selection
    // (the filters are applied in this order)
    AnswerSelection.clearFilters();
    // - answer extraction filters
    AnswerSelection.addFilter(new AnswerTypeFilter());
    AnswerSelection.addFilter(new AnswerPatternFilter());
    AnswerSelection.addFilter(new WebDocumentFetcherFilter());
    AnswerSelection.addFilter(new PredicateExtractionFilter());
    AnswerSelection.addFilter(new FactoidsFromPredicatesFilter());
    AnswerSelection.addFilter(new TruncationFilter());
    // - answer selection filters
    AnswerSelection.addFilter(new StopwordFilter());
    AnswerSelection.addFilter(new QuestionKeywordsFilter());
    AnswerSelection.addFilter(new AnswerProjectionFilter(resultsCorp));
    AnswerSelection.addFilter(new ScoreNormalizationFilter(NORMALIZER));
    AnswerSelection.addFilter(new ScoreCombinationFilter());
    AnswerSelection.addFilter(new FactoidSubsetFilter());
    AnswerSelection.addFilter(new DuplicateFilter());
    AnswerSelection.addFilter(new ScoreSorterFilter());
    AnswerSelection.addFilter(new ResultLengthFilter());
}
Also used : ScoreCombinationFilter(info.ephyra.answerselection.filters.ScoreCombinationFilter) ScoreSorterFilter(info.ephyra.answerselection.filters.ScoreSorterFilter) Ontology(info.ephyra.nlp.semantics.ontologies.Ontology) AnswerPatternFilter(info.ephyra.answerselection.filters.AnswerPatternFilter) PredicateExtractionFilter(info.ephyra.answerselection.filters.PredicateExtractionFilter) ScoreNormalizationFilter(info.ephyra.answerselection.filters.ScoreNormalizationFilter) WebDocumentFetcherFilter(info.ephyra.answerselection.filters.WebDocumentFetcherFilter) StopwordFilter(info.ephyra.answerselection.filters.StopwordFilter) TruncationFilter(info.ephyra.answerselection.filters.TruncationFilter) WordNet(info.ephyra.nlp.semantics.ontologies.WordNet) BagOfWordsG(info.ephyra.querygeneration.generators.BagOfWordsG) PredicateG(info.ephyra.querygeneration.generators.PredicateG) AnswerTypeFilter(info.ephyra.answerselection.filters.AnswerTypeFilter) ResultLengthFilter(info.ephyra.answerselection.filters.ResultLengthFilter) QuestionReformulationG(info.ephyra.querygeneration.generators.QuestionReformulationG) QuestionKeywordsFilter(info.ephyra.answerselection.filters.QuestionKeywordsFilter) DuplicateFilter(info.ephyra.answerselection.filters.DuplicateFilter) FactoidSubsetFilter(info.ephyra.answerselection.filters.FactoidSubsetFilter) BagOfTermsG(info.ephyra.querygeneration.generators.BagOfTermsG) BingKM(info.ephyra.search.searchers.BingKM) QuestionInterpretationG(info.ephyra.querygeneration.generators.QuestionInterpretationG) FactoidsFromPredicatesFilter(info.ephyra.answerselection.filters.FactoidsFromPredicatesFilter) AnswerProjectionFilter(info.ephyra.answerselection.filters.AnswerProjectionFilter)

Aggregations

AnswerPatternFilter (info.ephyra.answerselection.filters.AnswerPatternFilter)2 AnswerTypeFilter (info.ephyra.answerselection.filters.AnswerTypeFilter)2 FactoidsFromPredicatesFilter (info.ephyra.answerselection.filters.FactoidsFromPredicatesFilter)2 PredicateExtractionFilter (info.ephyra.answerselection.filters.PredicateExtractionFilter)2 TruncationFilter (info.ephyra.answerselection.filters.TruncationFilter)2 WebDocumentFetcherFilter (info.ephyra.answerselection.filters.WebDocumentFetcherFilter)2 Ontology (info.ephyra.nlp.semantics.ontologies.Ontology)2 WordNet (info.ephyra.nlp.semantics.ontologies.WordNet)2 BagOfTermsG (info.ephyra.querygeneration.generators.BagOfTermsG)2 BagOfWordsG (info.ephyra.querygeneration.generators.BagOfWordsG)2 PredicateG (info.ephyra.querygeneration.generators.PredicateG)2 QuestionInterpretationG (info.ephyra.querygeneration.generators.QuestionInterpretationG)2 QuestionReformulationG (info.ephyra.querygeneration.generators.QuestionReformulationG)2 AnswerProjectionFilter (info.ephyra.answerselection.filters.AnswerProjectionFilter)1 DuplicateFilter (info.ephyra.answerselection.filters.DuplicateFilter)1 FactoidSubsetFilter (info.ephyra.answerselection.filters.FactoidSubsetFilter)1 QuestionKeywordsFilter (info.ephyra.answerselection.filters.QuestionKeywordsFilter)1 ResultLengthFilter (info.ephyra.answerselection.filters.ResultLengthFilter)1 ScoreCombinationFilter (info.ephyra.answerselection.filters.ScoreCombinationFilter)1 ScoreNormalizationFilter (info.ephyra.answerselection.filters.ScoreNormalizationFilter)1