private Set<String> allowedFollowingTags(
Set<String> tags, String previousPreviousTag, String previousTag) {
Set<String> allowedTags = new HashSet<String>();
for (String tag : tags) {
String trigramString = makeTrigramString(previousPreviousTag, previousTag, tag);
if (seenTagTrigrams.contains((trigramString))) {
allowedTags.add(tag);
}
}
return allowedTags;
}
private static void evaluateTagger(
POSTagger posTagger,
List<TaggedSentence> taggedSentences,
Set<String> trainingVocabulary,
boolean verbose) {
double numTags = 0.0;
double numTagsCorrect = 0.0;
double numUnknownWords = 0.0;
double numUnknownWordsCorrect = 0.0;
int numDecodingInversions = 0;
for (TaggedSentence taggedSentence : taggedSentences) {
List<String> words = taggedSentence.getWords();
List<String> goldTags = taggedSentence.getTags();
List<String> guessedTags = posTagger.tag(words);
for (int position = 0; position < words.size() - 1; position++) {
String word = words.get(position);
String goldTag = goldTags.get(position);
String guessedTag = guessedTags.get(position);
if (guessedTag.equals(goldTag)) numTagsCorrect += 1.0;
numTags += 1.0;
if (!trainingVocabulary.contains(word)) {
if (guessedTag.equals(goldTag)) numUnknownWordsCorrect += 1.0;
numUnknownWords += 1.0;
}
}
double scoreOfGoldTagging = posTagger.scoreTagging(taggedSentence);
double scoreOfGuessedTagging = posTagger.scoreTagging(new TaggedSentence(words, guessedTags));
if (scoreOfGoldTagging > scoreOfGuessedTagging) {
numDecodingInversions++;
if (verbose)
System.out.println(
"WARNING: Decoder suboptimality detected. Gold tagging has higher score than guessed tagging.");
}
if (verbose) System.out.println(alignedTaggings(words, goldTags, guessedTags, true) + "\n");
}
System.out.println(
"Tag Accuracy: "
+ (numTagsCorrect / numTags)
+ " (Unknown Accuracy: "
+ (numUnknownWordsCorrect / numUnknownWords)
+ ") Decoder Suboptimalities Detected: "
+ numDecodingInversions);
}
public Counter<String> getLogScoreCounter(LocalTrigramContext localTrigramContext) {
int position = localTrigramContext.getPosition();
String word = localTrigramContext.getWords().get(position);
Counter<String> tagCounter = unknownWordTags;
if (wordsToTags.keySet().contains(word)) {
tagCounter = wordsToTags.getCounter(word);
}
Set<String> allowedFollowingTags =
allowedFollowingTags(
tagCounter.keySet(),
localTrigramContext.getPreviousPreviousTag(),
localTrigramContext.getPreviousTag());
Counter<String> logScoreCounter = new Counter<String>();
for (String tag : tagCounter.keySet()) {
double logScore = Math.log(tagCounter.getCount(tag));
if (!restrictTrigrams
|| allowedFollowingTags.isEmpty()
|| allowedFollowingTags.contains(tag)) logScoreCounter.setCount(tag, logScore);
}
return logScoreCounter;
}