public void TestClassify()
{
var wds = new SimpleWordsDataSource();
var classifier = new BayesianClassifier(wds);
var sentence = new[] { "This", "is", "a", "sentence", "about", "java" };
Assert.AreEqual(IClassifierConstants.NEUTRAL_PROBABILITY, classifier.Classify(ICategorizedClassifierConstants.DEFAULT_CATEGORY, sentence), 0d);
wds.SetWordProbability(new WordProbability("This", .5d));
wds.SetWordProbability(new WordProbability("is", .5d));
wds.SetWordProbability(new WordProbability("a", .5d));
wds.SetWordProbability(new WordProbability("sentence", .2d));
wds.SetWordProbability(new WordProbability("about", .5d));
wds.SetWordProbability(new WordProbability("java", .99d));
Assert.AreEqual(.96d, classifier.Classify(ICategorizedClassifierConstants.DEFAULT_CATEGORY, sentence), .009d);
}
public void TestGetStopWordProvider()
{
var wds = new SimpleWordsDataSource();
ITokenizer tokenizer = new DefaultTokenizer(DefaultTokenizer.BREAK_ON_WORD_BREAKS);
IStopWordProvider stopWordProvider = new DefaultStopWordProvider();
var classifier = new BayesianClassifier(wds, tokenizer, stopWordProvider);
Assert.AreEqual(stopWordProvider, classifier.StopWordProvider);
}
public void TestGetWordsDataSource()
{
var wds = new SimpleWordsDataSource();
var classifier = new BayesianClassifier(wds);
Assert.AreEqual(wds, classifier.WordsDataSource);
}
public void TestGetTokenizer()
{
SimpleWordsDataSource wds = new SimpleWordsDataSource();
ITokenizer tokenizer = new DefaultTokenizer(DefaultTokenizer.BREAK_ON_WORD_BREAKS);
BayesianClassifier classifier = new BayesianClassifier(wds, tokenizer);
Assert.AreEqual(tokenizer, classifier.Tokenizer);
}
protected void Setup()
{
wordsDataSource = new SimpleWordsDataSource();
}
public Classifier(int topicN, string mainDirectory)
{
this.topicN = topicN;
this.mainDirectory = mainDirectory;
this.TopicFileName = "topic" + topicN + ".txt";
this.TopicDir = "" + topicN;
IWordsDataSource wds = new SimpleWordsDataSource();
this.classifier = new BayesianClassifier(wds);
}
