From 571897436b042a35f468c80d462ce0de2c55de0b Mon Sep 17 00:00:00 2001 From: Luca Foppiano Date: Tue, 18 Jun 2019 15:41:24 +0900 Subject: [PATCH] use minus character to trigger the HYPEN feature --- .../java/org/grobid/core/features/FeaturesVectorQuantities.java | 2 +- src/main/java/org/grobid/core/features/FeaturesVectorUnits.java | 2 +- .../java/org/grobid/core/features/FeaturesVectorValues.java | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) diff --git a/src/main/java/org/grobid/core/features/FeaturesVectorQuantities.java b/src/main/java/org/grobid/core/features/FeaturesVectorQuantities.java index 01482921..5f7dd6fa 100644 --- a/src/main/java/org/grobid/core/features/FeaturesVectorQuantities.java +++ b/src/main/java/org/grobid/core/features/FeaturesVectorQuantities.java @@ -166,7 +166,7 @@ else if (featureFactory.test_digit(word)) featuresVector.punctType = "DOT"; } else if (word.equals(",")) { featuresVector.punctType = "COMMA"; - } else if (word.equals("-")) { + } else if (word.equals("-") || word.equals("−")) { // we cover hypen and minus sign featuresVector.punctType = "HYPHEN"; } else if (word.equals("\"") || word.equals("\'") || word.equals("`")) { featuresVector.punctType = "QUOTE"; diff --git a/src/main/java/org/grobid/core/features/FeaturesVectorUnits.java b/src/main/java/org/grobid/core/features/FeaturesVectorUnits.java index 4c641681..59803414 100644 --- a/src/main/java/org/grobid/core/features/FeaturesVectorUnits.java +++ b/src/main/java/org/grobid/core/features/FeaturesVectorUnits.java @@ -98,7 +98,7 @@ public static FeaturesVectorUnits addFeaturesUnit(String character, featuresVector.punctType = "DOT"; } else if (character.equals(",")) { featuresVector.punctType = "COMMA"; - } else if (character.equals("-")) { + } else if (character.equals("-") || character.equals("−")) { featuresVector.punctType = "HYPHEN"; } else if (character.equals("\"") || character.equals("\'") || character.equals("`")) { featuresVector.punctType = "QUOTE"; diff --git a/src/main/java/org/grobid/core/features/FeaturesVectorValues.java b/src/main/java/org/grobid/core/features/FeaturesVectorValues.java index b0a18599..453ec42a 100644 --- a/src/main/java/org/grobid/core/features/FeaturesVectorValues.java +++ b/src/main/java/org/grobid/core/features/FeaturesVectorValues.java @@ -75,7 +75,7 @@ public static FeaturesVectorValues addFeatures(String character, String label) { featuresVector.punctType = "DOT"; } else if (character.equals(",")) { featuresVector.punctType = "COMMA"; - } else if (character.equals("-")) { + } else if (character.equals("-") || character.equals("−")) { featuresVector.punctType = "HYPHEN"; } else if (character.equals("\"") || character.equals("\'") || character.equals("`")) { featuresVector.punctType = "QUOTE";