public IndexProcessor(
IWordsExtractor wordsExtractor,
IDocumentWithExtractedWordsStore documentWithExtractedWordsStore)
{
this.wordsExtractor = wordsExtractor;
this.documentWithExtractedWordsStore = documentWithExtractedWordsStore;
}
public ProbabilityProvider(ITrainingSample trainingSample, IWordsExtractor wordsExtractor, IWordsPreprocessor wordsPreprocessor)
{
spamCountWithWord = CalculateWordsCount(trainingSample.SpamFiles, wordsExtractor, wordsPreprocessor);
totalSpamCount = trainingSample.SpamFiles.Count();
hamCountWithWord = CalculateWordsCount(trainingSample.HamFiles, wordsExtractor, wordsPreprocessor);
totalHamCount = trainingSample.HamFiles.Count();
}
public BayesianSpamFilter(IProbabilityProvider probabilityProvider, IWordsExtractor wordsExtractor, IWordsPreprocessor wordsPreprocessor)
{
this.probabilityProvider = probabilityProvider;
this.wordsExtractor = wordsExtractor;
this.wordsPreprocessor = wordsPreprocessor;
}
private static Dictionary <string, int> CalculateWordsCount(IEnumerable <FileInfo> msgFiles, IWordsExtractor wordsExtractor, IWordsPreprocessor wordsPreprocessor)
{
var wordToCountMessagesWithThisWord = new Dictionary <string, int>();
foreach (var messageFile in msgFiles)
{
var rawWords = wordsExtractor.ExtractWords(messageFile.OpenRead());
var preprocessedWords = wordsPreprocessor.PreprocessWords(rawWords);
foreach (var word in preprocessedWords.Distinct())
{
if (!wordToCountMessagesWithThisWord.ContainsKey(word))
{
wordToCountMessagesWithThisWord[word] = 0;
}
wordToCountMessagesWithThisWord[word]++;
}
}
return(wordToCountMessagesWithThisWord);
}
