public String printVector() { if (string == null) return null; if (string.length() == 0) return null; StringBuilder res = new StringBuilder(); // token string (1) res.append(string); // lowercase string (1) res.append(" ").append(string.toLowerCase()); // prefix (4) res.append(" " + TextUtilities.prefix(string, 1)); res.append(" " + TextUtilities.prefix(string, 2)); res.append(" " + TextUtilities.prefix(string, 3)); res.append(" " + TextUtilities.prefix(string, 4)); // suffix (4) res.append(" " + TextUtilities.suffix(string, 1)); res.append(" " + TextUtilities.suffix(string, 2)); res.append(" " + TextUtilities.suffix(string, 3)); res.append(" " + TextUtilities.suffix(string, 4)); // line information (1) res.append(" ").append(lineStatus); // capitalisation (1) if (digit.equals("ALLDIGIT")) res.append(" NOCAPS"); else res.append(" ").append(capitalisation); // digit information (1) res.append(" ").append(digit); // character information (1) if (singleChar) res.append(" 1"); else res.append(" 0"); // lexical information (8) if (properName) res.append(" 1"); else res.append(" 0"); if (commonName) res.append(" 1"); else res.append(" 0"); if (firstName) res.append(" 1"); else res.append(" 0"); if (locationName) res.append(" 1"); else res.append(" 0"); if (year) res.append(" 1"); else res.append(" 0"); if (month) res.append(" 1"); else res.append(" 0"); if (email) res.append(" 1"); else res.append(" 0"); if (http) res.append(" 1"); else res.append(" 0"); // bibliographical information(4) if (isKnownJournalTitle || isKnownAbbrevJournalTitle) res.append(" 1"); else res.append(" 0"); if (isKnownConferenceTitle) res.append(" 1"); else res.append(" 0"); if (isKnownPublisher) res.append(" 1"); else res.append(" 0"); // punctuation information (1) res.append(" ").append(punctType); // in case the token is a punctuation (NO otherwise) // relative position in the sequence (1) res.append(" ").append(relativePosition); // label - for training data (1) if (label != null) res.append(" ").append(label).append("\n"); else res.append(" 0\n"); return res.toString(); }