public void Createhashtag()
{
instance = WordOccurrence.Create(helper.Handler.Object, helper.RawTextExractor.Object, helper.InquirerManager.Object, "#word", null, POSTags.Instance.NN);
Assert.AreEqual(NamedEntities.Hashtag, instance.Entity);
instance.Entity = NamedEntities.Date;
Assert.AreEqual(NamedEntities.Hashtag, instance.Entity);
}
public TwitterProfile Train(TwitterProfile profile, TextContent content)
{
if (profile == null || content?.Text == null)
{
throw new ArgumentNullException();
}
var regex = new Regex(@"(\.|,| |!|\?)");
var excluded = new Regex(@"(^@|^#|^;|^:|^http|^HTTP|\/|\\|…$|&\S+|^\""$|^\"".*[^\""]$|^[^\""].*\""$|^RT$)");
//TODO move remove @ and # to tweet generator instead of trainer
var words = regex.Split(content.Text)
.Where(word => !string.IsNullOrWhiteSpace(word))
.Where(word => !excluded.IsMatch(word))
.Select(word => new Word(word)).ToList();
WordOccurrence lastWordOccurrence = null;
Word temporaryWord;
foreach (var word in words)
{
if (profile.Vocabulary.Any(w => w.Equals(word)))
{
temporaryWord = profile.Vocabulary.SingleOrDefault(w => w.Equals(word));
}
else if (_wordRepository.Get(word) != null)
{
temporaryWord = _wordRepository.Get(word);
}
else
{
temporaryWord = _wordRepository.Add(word);
}
if (temporaryWord == null)
{
continue;
}
var currentWordOccurrence = profile.Words.SingleOrDefault(wo => wo.Word == temporaryWord);
if (currentWordOccurrence == null)
{
currentWordOccurrence = profile.AddWord(temporaryWord);
}
else
{
currentWordOccurrence.Occurrence++;
}
lastWordOccurrence?.AddOccurrence(currentWordOccurrence.Word);
lastWordOccurrence = currentWordOccurrence;
}
return(profile);
}
public void CreateArguments()
{
Assert.Throws <ArgumentNullException>(() => WordOccurrence.Create(null, helper.RawTextExractor.Object, helper.InquirerManager.Object, "Test", null, POSTags.Instance.NN));
Assert.Throws <ArgumentNullException>(() => WordOccurrence.Create(helper.Handler.Object, null, helper.InquirerManager.Object, "Test", null, POSTags.Instance.NN));
Assert.Throws <ArgumentNullException>(() => WordOccurrence.Create(helper.Handler.Object, helper.RawTextExractor.Object, null, "Test", null, POSTags.Instance.NN));
Assert.Throws <ArgumentException>(() => WordOccurrence.Create(helper.Handler.Object, helper.RawTextExractor.Object, helper.InquirerManager.Object, null, null, POSTags.Instance.NN));
Assert.Throws <ArgumentNullException>(() => WordOccurrence.Create(helper.Handler.Object, helper.RawTextExractor.Object, helper.InquirerManager.Object, "Test", null, null));
Assert.Throws <ArgumentException>(() => WordOccurrence.Create(helper.Handler.Object, helper.RawTextExractor.Object, helper.InquirerManager.Object, "Test", null, POSTags.Instance.SBAR));
}
public void Setup()
{
helper = new WordsHandlerHelper();
helper.RawTextExractor.Setup(item => item.GetWord("test")).Returns("T");
helper.Handler.Setup(item => item.CheckSentiment(It.IsAny <WordOccurrence>())).Returns(new SentimentValue(new TestWordItem(), 2));
helper.Handler.Setup(item => item.IsStop(It.IsAny <WordOccurrence>())).Returns(true);
helper.Handler.Setup(item => item.MeasureQuantifier(It.IsAny <WordOccurrence>())).Returns(2);
instance = WordOccurrence.Create(helper.Handler.Object, helper.RawTextExractor.Object, helper.InquirerManager.Object, "Test", null, POSTags.Instance.NN);
}
public void Add()
{
Assert.Throws <ArgumentNullException>(() => instance.Add(null));
instance.Add(WordOccurrence.Create(helper.Handler.Object, helper.RawTextExractor.Object, helper.InquirerManager.Object, "Test", null, POSTags.Instance.NN));
instance.Add(WordOccurrence.Create(helper.Handler.Object, helper.RawTextExractor.Object, helper.InquirerManager.Object, "Test", null, POSTags.Instance.NN));
Assert.AreEqual("test test", instance.Text);
Assert.AreEqual("t t", instance.Stemmed);
Assert.AreEqual("NN", instance.POS.Tag);
Assert.IsNotNull(instance.Relationship);
Assert.IsFalse(instance.IsSentiment);
Assert.IsFalse(instance.IsTopAttribute);
Assert.AreEqual(2, instance.QuantValue);
Assert.IsFalse(instance.IsQuestion);
Assert.IsFalse(instance.IsFeature);
Assert.IsFalse(instance.IsFixed);
Assert.IsFalse(instance.IsStopWord);
Assert.IsFalse(instance.IsSimple);
Assert.AreEqual(2, instance.AllWords.Count());
}
public void CreateFixed()
{
instance = WordOccurrence.Create(helper.Handler.Object, helper.RawTextExractor.Object, helper.InquirerManager.Object, "xxxbad", null, POSTags.Instance.NN);
Assert.IsTrue(instance.IsFixed);
}
private static void BuildFrequencyDictionary()
{
var result = new FrequencyResults();
Console.WriteLine("How many sentences do you want to parse?");
var nbOfSentencesToParse = int.Parse(Console.ReadLine());
var nbOfAlreadyParsedSentences = 0;
var frequencyDirectory = Utilities.PathToDownloadDirectory + "frequencies";
if (!Directory.Exists(frequencyDirectory))
{
Directory.CreateDirectory(frequencyDirectory);
}
var frequencyFilePath = frequencyDirectory + "/frequencies.txt";
var excludedFrequencyFilePath = frequencyDirectory + "/excluded-frequencies.txt";
var nbOfSentencesParsedFilePath = frequencyDirectory + "/nbOfSentencesParsed.txt";
var parsingResumed = false;
if (File.Exists(nbOfSentencesParsedFilePath))
{
int nbOfSentencesParsed;
if (int.TryParse(File.ReadAllText(nbOfSentencesParsedFilePath), out nbOfSentencesParsed))
{
Console.WriteLine("{0} sentences have already been parsed. Resume parsing? (y/n)", nbOfSentencesParsed);
var resumeParsing = string.Equals(Console.ReadLine(), "Y", StringComparison.InvariantCultureIgnoreCase);
if (resumeParsing)
{
nbOfAlreadyParsedSentences = nbOfSentencesParsed;
parsingResumed = true;
}
}
}
var sw = Stopwatch.StartNew();
Console.WriteLine("Building of frequency dictionary started");
// Tokenize the sentences and compute the frequencies
Func <string[], int, bool> extractTokens = (tokens, sentenceCounter) =>
{
for (var i = 0; i < tokens.Length; i++)
{
var wordOccurence = new WordOccurrence()
{
IsFirstTokenInSentence = i == 0,
Word = tokens[i]
};
result.AddOccurence(wordOccurence);
}
return(true);
};
Utilities.ExtractTokensFromTxtFiles(extractTokens, nbOfSentencesToParse, nbOfAlreadyParsedSentences);
// Load previous frequency dictionaries that were already computed
if (parsingResumed)
{
result.LoadFrequencyDictionary(frequencyFilePath);
result.LoadFrequencyDictionary(excludedFrequencyFilePath);
}
// Save frequency files on disk
result.SaveFrequencyDictionary(frequencyFilePath);
result.SaveExcludedFrequencyDictionary(excludedFrequencyFilePath);
// Save the nb of sentences parsed (for information and being able to relaunch the parsing at this point)
File.WriteAllText(nbOfSentencesParsedFilePath, nbOfSentencesToParse.ToString());
Console.WriteLine("Building of frequency dictionary done");
Console.WriteLine("=====================================");
sw.Stop();
Console.WriteLine("Ellapsed time: {0}", sw.Elapsed.ToString("g"));
}
public IDictionaryEnumerator GetWordsByOccurrenceEnumerator()
{
SortedList sl = new SortedList();
IDictionaryEnumerator enumer = GetWordsAlphabeticallyEnumerator();
while (enumer.MoveNext())
{
WordOccurrence wo = new WordOccurrence((int)enumer.Value, (string)enumer.Key);
sl.Add(wo, null);
}
return sl.GetEnumerator();
}
protected override void CalculateRatingLogic()
{
TextVectorCell[] cells = Review.Vector.GetCells().ToArray();
(double Probability, double Normalization, VectorData Vector)result = Model.GetVector(cells);
VectorData vector = result.Vector;
if (vector == null ||
vector.Length == 0)
{
Rating = Review.CalculateRawRating();
return;
}
var bias = vector.RHO;
var fallbackWeight = 0.1;
VectorCell lexicon = default;
foreach (VectorCell item in vector.Cells)
{
var cell = (TextVectorCell)item.Data;
if (cell.Name == Constants.RATING_STARS)
{
lexicon = item;
}
if (cell.Item != null)
{
var word = (IWordItem)cell.Item;
Add(new SentimentValue(word, word.Text, new SentimentValueData(item.Calculated, SentimentSource.AdjustedSVM)));
}
else
{
bias += item.Calculated;
}
}
var notAddedSentiments = new List <SentimentValue>();
foreach (SentimentValue sentimentValue in Review.GetAllSentiments())
{
if (!ContainsSentiment(sentimentValue.Owner))
{
notAddedSentiments.Add(sentimentValue);
}
}
if (lexicon != null)
{
var totalWords = Review.GetAllSentiments().Length;
fallbackWeight = Math.Abs(lexicon.Theta) / totalWords;
}
if (notAddedSentiments.Count > 0)
{
foreach (SentimentValue sentiment in notAddedSentiments)
{
Add(new SentimentValue(sentiment.Owner, sentiment.Span, new SentimentValueData(sentiment.DataValue.Value * fallbackWeight, SentimentSource.AdjustedCalculated)));
}
}
if (TotalSentiments > 0)
{
Add(new SentimentValue(
WordOccurrence.CreateBasic(Constants.BIAS, POSTags.Instance.JJ),
"BIAS",
new SentimentValueData(bias, SentimentSource.AdjustedSVM)));
}
if (Rating.HasValue)
{
if (Rating.IsPositive.Value &&
result.Probability < 0.5)
{
log.LogDebug("Mistmatch in sentiment with machine prediction: {0} - {1}", Rating.IsPositive, result.Probability);
}
}
}
