Accept fourty as 40... probably can't mean anything else, after all

stanfordnlp · Aug 31, 2023 · 7fbb7b8 · 7fbb7b8
1 parent d5daf7c
commit 7fbb7b8
Show file tree

Hide file tree

Showing 2 changed files with 7 additions and 4 deletions.
diff --git a/itest/src/edu/stanford/nlp/ie/NumberNormalizerITest.java b/itest/src/edu/stanford/nlp/ie/NumberNormalizerITest.java
@@ -67,15 +67,17 @@ public void testNumbers() {
             "a dozen bagels\n" +
             "five dozen\n" +
             "An IQ score of 161.\n" +     // only 161, not 20 for score
-            "thirty two"
+            "thirty two\n" +
+            "I hope Jennifer would let me lick her antennae even though I am forty-five\n" +
+            "I hope Jennifer would let me lick her antennae even though I am fourty-five"
             ;
 
     // set up expected results
     Iterator<? extends Number> expectedNumbers = Arrays.asList(
             24.0, 405, 600.0, 412.0, 4650211.0, 600005650376.0, 5786345, 25.0,
             /* 1500000.0, */
             1300000.0, 1224.0, 10000000000.0, 3.625,
-            0, -15.0, 1, 2, 3, 4, 155.0, 101.0 /*504.0, */, 80.0, 12, 60.0, 161, 32.0 ).iterator();
+            0, -15.0, 1, 2, 3, 4, 155.0, 101.0 /*504.0, */, 80.0, 12, 60.0, 161, 32.0, 45.0, 45.0 ).iterator();
     Iterator<String> expectedTexts = Arrays.asList(
             "two dozen", "405", "six hundred", "four hundred, and twelve",
             "4 million six hundred fifty thousand, two hundred and eleven",
@@ -93,7 +95,7 @@ public void testNumbers() {
             "four score",
             "dozen",
             "five dozen",
-            "161", "thirty two").iterator();
+            "161", "thirty two", "forty-five", "fourty-five").iterator();
 
     // create document
     Annotation document = createDocument(testText);

diff --git a/src/edu/stanford/nlp/ie/NumberNormalizer.java b/src/edu/stanford/nlp/ie/NumberNormalizer.java
@@ -78,7 +78,7 @@ public static void setVerbose(boolean verbose) {
 
   /***********************/
 
-  private static final Pattern numberTermPattern = Pattern.compile("(?i)(zero|one|two|three|four|five|six|seven|eight|nine|ten|eleven|twelve|thirteen|fourteen|fifteen|sixteen|seventeen|eighteen|nineteen|twenty|thirty|forty|fifty|sixty|seventy|eighty|ninety|hundred|thousand|million|billion|trillion|first|second|third|fourth|fifth|sixth|seventh|eighth|ninth|tenth|eleventh|twelfth|thirteenth|fourteenth|fifteenth|sixteenth|seventeenth|eighteenth|nineteenth|twentieth|thirtieth|fortieth|fiftieth|sixtieth|seventieth|eightieth|ninetieth|hundred?th|thousandth|millionth|billionth|trillionth)");
+  private static final Pattern numberTermPattern = Pattern.compile("(?i)(zero|one|two|three|four|five|six|seven|eight|nine|ten|eleven|twelve|thirteen|fourteen|fifteen|sixteen|seventeen|eighteen|nineteen|twenty|thirty|forty|fourty|fifty|sixty|seventy|eighty|ninety|hundred|thousand|million|billion|trillion|first|second|third|fourth|fifth|sixth|seventh|eighth|ninth|tenth|eleventh|twelfth|thirteenth|fourteenth|fifteenth|sixteenth|seventeenth|eighteenth|nineteenth|twentieth|thirtieth|fortieth|fiftieth|sixtieth|seventieth|eightieth|ninetieth|hundred?th|thousandth|millionth|billionth|trillionth)");
   private static final Pattern numberTermPattern2 = Pattern.compile("(?i)(" + numberTermPattern.pattern() + "(-" + numberTermPattern.pattern() + ")?)");
   private static final Pattern ordinalUnitPattern = Pattern.compile("(?i)(hundredth|thousandth|millionth)");
 
@@ -147,6 +147,7 @@ public static void setVerbose(boolean verbose) {
     word2NumMap.put("twenty", 20);
     word2NumMap.put("thirty",  30);
     word2NumMap.put("forty", 40);
+    word2NumMap.put("fourty", 40);
     word2NumMap.put("fifty", 50);
     word2NumMap.put("sixty",  60);
     word2NumMap.put("seventy", 70);