/** * Returns a document in the VSM. */ public DoubleVector mapDocument(BOW bow) { return mapDocument(bow, true); } // end map
public double compare(BOW bow1, BOW bow2) { DoubleVector d1 = mapDocument(bow1); DoubleVector d2 = mapDocument(bow2); DoubleVector pd1 = mapPseudoDocument(d1); DoubleVector pd2 = mapPseudoDocument(d2); return pd1.dot(pd2) / Math.sqrt(pd1.dot(pd1) * pd2.dot(pd2)); }
BOW bow2 = new BOW(s[1].toLowerCase().replaceAll("category:", "_").split("[_ ]")); DoubleVector d1 = mapDocument(bow1); DoubleVector d2 = mapDocument(bow2);