Merge pull request eugenp#198 from Doha2012/master

update classifier
JeffersonAlmeida · Apr 21, 2015 · 91ea275 · 91ea275
2 parents f880c5f + 42369da
commit 91ea275
Showing 1 changed file with 13 additions and 10 deletions.
diff --git a/spring-security-oauth/src/main/java/org/baeldung/reddit/classifier/RedditClassifier.java b/spring-security-oauth/src/main/java/org/baeldung/reddit/classifier/RedditClassifier.java
@@ -25,7 +25,7 @@ public class RedditClassifier {
 
     public static int GOOD = 0;
     public static int BAD = 1;
-    public static int MIN_SCORE = 7;
+    public static int MIN_SCORE = 10;
     public static int NUM_OF_FEATURES = 1000;
 
     private final AdaptiveLogisticRegression classifier;
@@ -42,9 +42,9 @@ public class RedditClassifier {
 
     public RedditClassifier() {
         classifier = new AdaptiveLogisticRegression(2, NUM_OF_FEATURES, new L2());
-        classifier.setPoolSize(50);
+        classifier.setPoolSize(150);
         titleEncoder = new AdaptiveWordValueEncoder("title");
-        titleEncoder.setProbes(1);
+        titleEncoder.setProbes(2);
         domainEncoder = new StaticWordValueEncoder("domain");
         domainEncoder.setProbes(1);
     }
@@ -65,13 +65,15 @@ public void trainClassifier(String fileName) throws IOException {
     }
 
     public Vector convertPost(String title, String domain, int hour) {
-        final Vector features = new RandomAccessSparseVector(4);
-        final int noOfWords = Splitter.onPattern("\\W").omitEmptyStrings().splitToList(title).size();
-        titleEncoder.addToVector(title, features);
-        domainEncoder.addToVector(domain, features);
-        features.set(2, hour);
-        features.set(3, noOfWords);
-        return features;
+        final Vector vector = new RandomAccessSparseVector(NUM_OF_FEATURES);
+        final List<String> words = Splitter.onPattern("\\W").omitEmptyStrings().splitToList(title);
+        vector.set(0, hour);
+        vector.set(1, words.size());
+        domainEncoder.addToVector(domain, vector);
+        for (final String word : words) {
+            titleEncoder.addToVector(word, vector);
+        }
+        return vector;
     }
 
     public int classify(Vector features) {
@@ -106,6 +108,7 @@ private void evaluateClassifier(List<NamedVector> vectors) throws IOException {
         System.out.println("Eval count ========= Good = " + evalCount[0] + " ___ Bad = " + evalCount[1]);
         System.out.println("Test result ======== Correct prediction = " + correct + " -----  Wrong prediction = " + wrong);
         System.out.println("Test result ======== Correct Good = " + correctCount[0] + " -----  Correct Bad = " + correctCount[1]);
+        System.out.println("Test result ======== Good accuracy = " + (correctCount[0] / (evalCount[0] + 0.0)) + " ----- Bad accuracy = " + (correctCount[1] / (evalCount[1] + 0.0)));
         this.accuracy = correct / (wrong + correct + 0.0);
     }