/**
* Extracts termClaimsDescriptionAbstractTitles of the documents; Adds them to vector in the same
* order
*
* @param hits
* @param i
* @param j
* @return relevantDocsTerms docs must be in order
* @throws java.io.IOException
*/
public Map<TermFreqVector, String> getDocsTerms(TopDocs hits, int i, int j) throws IOException {
Map<TermFreqVector, String> docsTerms = new HashMap<>();
// Process each of the documents
while (i < j && i < hits.totalHits && i >= 0) {
ScoreDoc scoreDoc = hits.scoreDocs[i];
if (sourceField.equals(PatentQuery.all)) {
Terms termTitle = ir.getTermVector(scoreDoc.doc, PatentQuery.getFields()[1]);
TermFreqVector docTermsTitle = new TermFreqVector(termTitle);
docsTerms.put(docTermsTitle, PatentQuery.getFields()[1]);
Terms termAbstract = ir.getTermVector(scoreDoc.doc, PatentQuery.getFields()[2]);
TermFreqVector docTermsAbstract = new TermFreqVector(termAbstract);
docsTerms.put(docTermsAbstract, PatentQuery.getFields()[2]);
Terms termDescription = ir.getTermVector(scoreDoc.doc, PatentQuery.getFields()[3]);
TermFreqVector docTermsDescription = new TermFreqVector(termDescription);
docsTerms.put(docTermsDescription, PatentQuery.getFields()[3]);
Terms termClaims = ir.getTermVector(scoreDoc.doc, PatentQuery.getFields()[5]);
TermFreqVector docTermsClaims = new TermFreqVector(termClaims);
docsTerms.put(docTermsClaims, PatentQuery.getFields()[5]);
} else {
Terms term = ir.getTermVector(scoreDoc.doc, sourceField); // get termvector for document
// Create termVector and add it to vector
TermFreqVector docTerms = new TermFreqVector(term);
docsTerms.put(docTerms, sourceField);
}
i++;
}
return docsTerms;
}
/*
* listTermVectorField displays the term vector for a field in
* a document in an index (specified by reader).
*/
static void listTermVectorField(IndexReader reader, String docidString, String field)
throws IOException {
System.out.println("\nTermVector: docid " + docidString + ", field " + field);
int docid = Integer.parseInt(docidString);
if ((docid < 0) || (docid >= reader.numDocs())) {
System.out.println("ERROR: " + docidString + " is a bad document id.");
return;
}
;
Terms terms = reader.getTermVector(docid, field);
termVectorDisplay(terms);
}