diff options
author | Peter Wu <peter@lekensteyn.nl> | 2015-03-31 18:57:57 +0200 |
---|---|---|
committer | Peter Wu <peter@lekensteyn.nl> | 2015-03-31 18:57:57 +0200 |
commit | bec3ac2780fce9f90830f9f21a03b6c5487cd21c (patch) | |
tree | 6facce81c2a27c5f1aa284fe24587e539d6e6bab | |
parent | ecce7ed7b203ac6937df1ff705cd4e14217cb3ca (diff) | |
download | assignment4-bec3ac2780fce9f90830f9f21a03b6c5487cd21c.tar.gz |
Add simple getSmoothedCount implementation
-rw-r--r-- | spellchecker/src/CorpusReader.java | 10 |
1 files changed, 7 insertions, 3 deletions
diff --git a/spellchecker/src/CorpusReader.java b/spellchecker/src/CorpusReader.java index 1392654..536a41b 100644 --- a/spellchecker/src/CorpusReader.java +++ b/spellchecker/src/CorpusReader.java @@ -99,6 +99,10 @@ public class CorpusReader { return vocabulary.contains(word); } + /** + * Returns a smoothed value based on the number of occurrences of the n-gram + * in the corpus. + */ public double getSmoothedCount(String NGram) { if (NGram == null || NGram.length() == 0) { throw new IllegalArgumentException("NGram must be non-empty."); @@ -106,9 +110,9 @@ public class CorpusReader { double smoothedCount = 0.0; - /** - * ADD CODE HERE * - */ + // simplest smoothing implementation: plus 1. + smoothedCount = getNGramCount(NGram) + 1; + return smoothedCount; } } |