private void copy_from(EnglishStemmer other)
{
B_Y_found = other.B_Y_found;
I_p2 = other.I_p2;
I_p1 = other.I_p1;
copy_from(other);
}
public List<Term> termsInQuery()
{
string documentWithoutPunctuation = "";
foreach (var c in SearchPhrase)
{
if (!char.IsPunctuation(c))
{
documentWithoutPunctuation += c;
}
}
SearchPhrase = documentWithoutPunctuation;
List<string> termStrings = SearchPhrase.ToLower().Split(' ').ToList();
List<string> stemmedTerms = new List<string>();
EnglishStemmer stemmer = new EnglishStemmer();
foreach(var ts in termStrings)
{
stemmedTerms.Add(stemmer.Stem(ts));
}
List<Term> terms = new List<Term>();
foreach(var stemmedterm in stemmedTerms.Distinct())
{
if(db.Terms.Any(i=>i.StemmedText == stemmedterm))
{
terms.Add(db.Terms.First(i => i.StemmedText == stemmedterm));
}
}
return terms;
}
private void btnFolder_Click(object sender, EventArgs e)
{
FolderBrowserDialog dia = new FolderBrowserDialog();
DialogResult res = dia.ShowDialog();
if (res != System.Windows.Forms.DialogResult.OK)
{
return;
}
FSDirectory dir = FSDirectory.GetDirectory(Environment.CurrentDirectory + "\\LuceneIndex");
//Lucene.Net.Store.RAMDirectory dir = new RAMDirectory();
Lucene.Net.Analysis.Standard.StandardAnalyzer an = new Lucene.Net.Analysis.Standard.StandardAnalyzer();
IndexWriter wr = new IndexWriter(dir, an,true);
IStemmer stemmer = new EnglishStemmer();
DirectoryInfo diMain = new DirectoryInfo(dia.SelectedPath);
foreach(FileInfo fi in diMain.GetFiles()){
Document doc = new Document();
doc.Add(new Field("title", fi.Name,Field.Store.YES, Field.Index.NO));
//doc.Add(new Field("text", File.ReadAllText(fi.FullName),Field.Store.YES, Field.Index.TOKENIZED,Field.TermVector.YES));
doc.Add(new Field("text", PerformStemming(stemmer,NLPToolkit.Tokenizer.TokenizeNow(File.ReadAllText(fi.FullName)).ToArray()), Field.Store.YES, Field.Index.TOKENIZED, Field.TermVector.YES));
wr.AddDocument(doc);
}
wr.Optimize();
wr.Flush();
wr.Close();
dir.Close();
IndexReader reader = IndexReader.Open(dir);
for (int i = 0; i < reader.MaxDoc(); i++)
{
if (reader.IsDeleted(i))
continue;
Document doc = reader.Document(i);
String docId = doc.Get("docId");
foreach (TermFreqVector vector in reader.GetTermFreqVectors(i))
{
foreach(string term in vector.GetTerms()){
Console.WriteLine(term);
}
}
// do something with docId here...
}
//IndexSearcher search = new IndexSearcher(wr.GetReader());
//MoreLikeThis mlt = new MoreLikeThis(wr.GetReader());
//FileInfo fitarget = new FileInfo(@"C:\Users\peacemaker\Desktop\TestNoBitcoin\test.txt");
//Query query = mlt.Like(fitarget);
//var hits = search.Search(query, int.MaxValue);
//foreach (ScoreDoc doc in hits.ScoreDocs)
//{
// textBox1.Text += doc.Score + Environment.NewLine;
//}
}
public static void CreateIndex(MongoCollection<TweetItem> collection)
{
DateTime dtmFirst = new DateTime(2014, 05, 17, 0, 0, 0);
DateTime dtmLast = new DateTime(2014, 05, 17, 23, 59, 59);
FSDirectory dir = FSDirectory.GetDirectory(Environment.CurrentDirectory + "\\LuceneIndex");
//Lucene.Net.Store.RAMDirectory dir = new RAMDirectory();
Lucene.Net.Analysis.StopAnalyzer an = new Lucene.Net.Analysis.StopAnalyzer();
IndexWriter wr = new IndexWriter(dir, an, true);
IStemmer stemmer = new EnglishStemmer();
while (dtmFirst.Date <= DateTime.Now.Date)
{
var query = Query<TweetItem>.Where(t => t.CreationDate >= dtmFirst && t.CreationDate <= dtmLast);
List<TweetItem> value = collection.Find(query).ToList();
//DirectoryInfo diMain = new DirectoryInfo(dia.SelectedPath);
using (var client = new HttpClient())
{
client.BaseAddress = new Uri("http://www.datasciencetoolkit.org/text2sentiment");
client.DefaultRequestHeaders.Accept.Clear();
client.DefaultRequestHeaders.Accept.Add(new MediaTypeWithQualityHeaderValue("application/json"));
foreach (TweetItem tweet in value)
{
Document doc = new Document();
//SentimentResult res = await GetSentiment(tweet.Text, client);
string stemmedtext = PerformStemming(stemmer, NLPToolkit.Tokenizer.TokenizeNow(tweet.Text).ToArray());
var scores = classifier.Classify(stemmedtext,DragonHelper.DragonHelper.ExcludeList);
string positiveSentiment = string.Empty;
string negativeSentiment = string.Empty;
positiveSentiment = scores["Positive"].ToString();
negativeSentiment = scores["Negative"].ToString();
doc.Add(new Field("id", tweet._id.ToString(), Field.Store.YES, Field.Index.NO));
doc.Add(new Field("created", tweet.CreationDate.ToString(), Field.Store.YES, Field.Index.NO));
doc.Add(new Field("user", tweet.User, Field.Store.YES, Field.Index.NO));
doc.Add(new Field("text", stemmedtext, Field.Store.YES, Field.Index.TOKENIZED, Field.TermVector.YES));
doc.Add(new Field("possentiment", positiveSentiment , Field.Store.YES, Field.Index.TOKENIZED, Field.TermVector.YES));
doc.Add(new Field("negsentiment", negativeSentiment, Field.Store.YES, Field.Index.TOKENIZED, Field.TermVector.YES));
wr.AddDocument(doc);
}
}
dtmFirst = dtmFirst.AddDays(1);
dtmLast = dtmLast.AddDays(1);
}
wr.Optimize();
wr.Flush();
wr.Close();
dir.Close();
}
private string StemWords(string opText)
{
Dictionary<string, int> currentPostStems = new Dictionary<string, int>();
string[] opArr = opText.Split(null);
string outputText = "";
//currentPostStems = TestStemmer(new EnglishStemmer(), opArr);
//foreach (KeyValuePair<string, int> stem in currentPostStems)
//{
// outputText += "Stem: " + stem.Key + "\t\tOccurs: " + stem.Value + Environment.NewLine;
//}
EnglishStemmer stemmer = new EnglishStemmer();
foreach (string word in opArr) {
outputText += stemmer.Stem(word) + " ";
}
return outputText;
}
public static void InitParams(List<string> t)
{
Eng = new StWdsEng();
Rus = new StWdsRus();
EngStem = new EnglishStemmer();
RusStem = new RussianStemmer();
ts = new List<string>();
ts = t;
Tag = new List<List<Word>>();
}
public static void InitParams(string[] t)
{
string[] temp = t;
Eng = new StWdsEng();
Rus = new StWdsRus();
EngStem = new EnglishStemmer();
RusStem = new RussianStemmer();
ts = new List<string>();
ts.AddRange(temp);
Tag = new List<List<Word>>();
}
private void copy_from(EnglishStemmer other)
{
B_Y_found = other.B_Y_found;
I_p2 = other.I_p2;
I_p1 = other.I_p1;
copy_from(other);
}
public static void InitParams(List<string> t)
{
Eng = new StWdsEng();
Rus = new StWdsRus();
EngStem = new EnglishStemmer();
RusStem = new RussianStemmer();
ts = new List<string>();
ts = t;
fullWdsCollection = new List<List<string>>[2];
fullWdsCollection[0] = new List<List<string>>();
fullWdsCollection[1] = new List<List<string>>();
Tag = new List<List<Word>>[2];
Tag[0] = new List<List<Word>>();
Tag[1] = new List<List<Word>>();
}
public ExternalEnglishStemmer()
{
stemmer = new EnglishStemmer();
}
private void processDocument(Document document, int totalNoOfDocumentsInCollection, List<Term> stopwords, List<Term> allTerms)
{
// remove all punctuation
string documentWithoutPunctuation = "";
foreach (var c in document.Content)
{
if (!char.IsPunctuation(c))
{
documentWithoutPunctuation += c;
}
}
//put the terms in the document onto a string list, splitting whenever there is a space, and whenever there is \r\n in document.
string[] splitStrings = { " ", "\r\n" };
List<string> rawTermsInDocument = documentWithoutPunctuation.ToLower().Split(splitStrings, StringSplitOptions.RemoveEmptyEntries).ToList();
//stems each term in the document, and then adds it to db.terms if it hasn't been encountered before. Also creates a list of all of the terms in the document in stemmed format
EnglishStemmer stemmer = new EnglishStemmer();
List<string> stemmedTermsInDocument = new List<string>();
foreach (var rawTerm in rawTermsInDocument)
{
string stemmedTerm = stemmer.Stem(rawTerm);
stemmedTermsInDocument.Add(stemmedTerm);
Term termObj = new Term { StemmedText = stemmedTerm, Text = rawTerm };
var test = db.Terms.Where(i => i.StemmedText == stemmedTerm).Any();
if (allTerms.Where(i => i.StemmedText == stemmedTerm).Any() == false && !stopwords.Contains(termObj))
{
db.Terms.Add(termObj);
db.SaveChanges();
allTerms.Add(termObj);
}
}
/*goes through all db.terms (ie all terms in the dictionary of sorts) and if the term is in the document, then it creates a term document weight
for the term, and also computes its term frequency*/
foreach (var term in allTerms)
{
if (stemmedTermsInDocument.Contains(term.StemmedText) && !stopwords.Contains(term))
{
TermDocumentWeight termDocumentWeight = new TermDocumentWeight();
termDocumentWeight.DocumentID = document.ID;
termDocumentWeight.TermID = term.ID;
termDocumentWeight.TermFrequency = stemmedTermsInDocument.Count(i => i == term.StemmedText);
db.TermDocumentWeights.Add(termDocumentWeight);
Debug.WriteLine("TDW for " + termDocumentWeight.Term + " added, ID: " + termDocumentWeight.ID);
}
}
db.SaveChanges();
}
