Skip to content

Commit

Permalink
Add suport of bilou and lowercase labels
Browse files Browse the repository at this point in the history
  • Loading branch information
@manning
manning authored and Stanford NLP committed Dec 29, 2014
1 parent 2fac926 commit 697bf66
Showing 1 changed file with 26 additions and 11 deletions.
37 changes: 26 additions & 11 deletions src/edu/stanford/nlp/sequences/IOBUtils.java
View file
Original file line number Diff line number Diff line change
Expand Up @@ -37,6 +37,8 @@ private IOBUtils() {} // static methods
* This code is very specific to the particular CoNLL way of labeling
* classes for IOB-style encoding, but this notation is quite widespread.
* It will work on any of these styles of input.
* This will also recognize BILOU format (B=B, I=I, L=E, O=O, U=S).
* It also works with lowercased names like i-org.
* If the labels are not of the form "C-Y+", where C is a single character,
* then they will be regarded as NOPREFIX labels.
* This method updates the List tokens in place.
Expand Down Expand Up @@ -79,6 +81,9 @@ public static <TOK extends CoreMap> void entitySubclassify(List<TOK> tokens,
case "noprefix":
how = 6;
break;
case "bilou":
how = 7;
break;
default:
throw new IllegalArgumentException("entitySubclassify: unknown style: " + style);
}
Expand All @@ -102,7 +107,7 @@ public static <TOK extends CoreMap> void entitySubclassify(List<TOK> tokens,
char prefix;
if (cAns.length() > 2 && cAns.charAt(1) == '-') {
base = cAns.substring(2, cAns.length());
prefix = cAns.charAt(0);
prefix = Character.toUpperCase(cAns.charAt(0));
} else {
base = cAns;
prefix = ' ';
Expand All @@ -111,7 +116,7 @@ public static <TOK extends CoreMap> void entitySubclassify(List<TOK> tokens,
char pPrefix;
if (pAns.length() > 2 && pAns.charAt(1) == '-') {
pBase = pAns.substring(2, pAns.length());
pPrefix = pAns.charAt(0);
pPrefix = Character.toUpperCase(pAns.charAt(0));
} else {
pBase = pAns;
pPrefix = ' ';
Expand All @@ -120,16 +125,16 @@ public static <TOK extends CoreMap> void entitySubclassify(List<TOK> tokens,
char nPrefix;
if (nAns.length() > 2 && nAns.charAt(1) == '-') {
nBase = nAns.substring(2, nAns.length());
nPrefix = nAns.charAt(0);
nPrefix = Character.toUpperCase(nAns.charAt(0));
} else {
nBase = nAns;
nPrefix = ' ';
}

boolean isStartAdjacentSame = base.equals(pBase) &&
(prefix == 'B' || prefix == 'S' || pPrefix == 'E' || pPrefix == 'S');
(prefix == 'B' || prefix == 'S' || prefix == 'U' || pPrefix == 'E' || pPrefix == 'S' || pPrefix == 'U');
boolean isEndAdjacentSame = base.equals(nBase) &&
(prefix == 'E' || prefix == 'S' || nPrefix == 'B' || pPrefix == 'S');
(prefix == 'E' || prefix == 'L' || prefix == 'S' || prefix == 'U' || nPrefix == 'B' || nPrefix == 'S' || nPrefix == 'U');
boolean isFirst = !base.equals(pBase) || isStartAdjacentSame;
boolean isLast = !base.equals(nBase) || isEndAdjacentSame;
String newAnswer = base;
Expand Down Expand Up @@ -176,7 +181,18 @@ public static <TOK extends CoreMap> void entitySubclassify(List<TOK> tokens,
} else {
newAnswer = "I-" + base;
}
break;
// nothing to do on case 6 as it's just base
case 7:
if (isFirst && isLast) {
newAnswer = "U-" + base;
} else if ( ( ! isFirst) && isLast) {
newAnswer = "L-" + base;
} else if (isFirst && ( ! isLast)) {
newAnswer = "B-" + base;
} else {
newAnswer = "I-" + base;
}
}
}
if (intern) {
Expand Down Expand Up @@ -231,7 +247,7 @@ public static boolean countEntityResults(List<? extends CoreMap> doc,
return false;
} else if (gold.length() > 2 && gold.charAt(1) == '-') {
goldEntity = gold.substring(2, gold.length());
goldPrefix = gold.charAt(0);
goldPrefix = Character.toUpperCase(gold.charAt(0));
} else {
goldEntity = gold;
goldPrefix = ' ';
Expand All @@ -241,7 +257,7 @@ public static boolean countEntityResults(List<? extends CoreMap> doc,
return false;
} else if (guess.length() > 2 && guess.charAt(1) == '-') {
guessEntity = guess.substring(2, guess.length());
guessPrefix = guess.charAt(0);
guessPrefix = Character.toUpperCase(guess.charAt(0));
} else {
guessEntity = guess;
guessPrefix = ' ';
Expand All @@ -251,13 +267,13 @@ public static boolean countEntityResults(List<? extends CoreMap> doc,
// "Guess: " + guess + " (" + guessPrefix + ' ' + guessEntity + ')');

boolean goldIsStartAdjacentSame = goldEntity.equals(previousGoldEntity) &&
(goldPrefix == 'B' || goldPrefix == 'S' || previousGoldPrefix == 'E' || previousGoldPrefix == 'S');
(goldPrefix == 'B' || goldPrefix == 'S' || goldPrefix == 'U' || previousGoldPrefix == 'E' || previousGoldPrefix == 'S' || previousGoldPrefix == 'S');
boolean newGold = ! gold.equals(background) &&
( ! goldEntity.equals(previousGoldEntity) || goldIsStartAdjacentSame);
boolean guessIsStartAdjacentSame = guessEntity.equals(previousGuessEntity) &&
(guessPrefix == 'B' || guessPrefix == 'S' || previousGuessPrefix == 'E' || previousGuessPrefix == 'S');
(guessPrefix == 'B' || guessPrefix == 'S' || guessPrefix == 'U' || previousGuessPrefix == 'E' || previousGuessPrefix == 'L' || previousGuessPrefix == 'S' || previousGuessPrefix == 'U');
boolean newGuess = ! guess.equals(background) &&
( ! guessEntity.equals(previousGuessEntity) || guessIsStartAdjacentSame);;
( ! guessEntity.equals(previousGuessEntity) || guessIsStartAdjacentSame);

boolean goldEnded = ! previousGold.equals(background) &&
( ! goldEntity.equals(previousGoldEntity) || goldIsStartAdjacentSame);
Expand Down Expand Up @@ -334,5 +350,4 @@ public static void main(String[] args) {
}
}


}

0 comments on commit 697bf66

Please sign in to comment.