This is a simple utility function that make word-level ngrams from a set of words @param words @param ngrams @param size
(String[] words, Set<String> ngrams, int size)
| 46 | * @param size |
| 47 | */ |
| 48 | protected static void makeNGram(String[] words, Set<String> ngrams, int size) { |
| 49 | int stop = words.length - size + 1; |
| 50 | for (int i = 0; i < stop; i++) { |
| 51 | StringBuilder sb = new StringBuilder(); |
| 52 | for (int j = 0; j < size; j++) { |
| 53 | sb.append(words[i + j]).append(" "); |
| 54 | } |
| 55 | sb.deleteCharAt(sb.length() - 1); |
| 56 | ngrams.add(sb.toString()); |
| 57 | } |
| 58 | if (size > 1) { |
| 59 | makeNGram(words, ngrams, size - 1); |
| 60 | } |
| 61 | } |
| 62 | |
| 63 | } |