static void Main(string[] args)
{
var TekstA = new List <string>();
var TekstB = new List <string>();
string[] stringSeparators = { ", ", ".", "!", "?", ";", ":", " ", "-", "\"", "(", ")" };
var test1 = new LoadEachWordToList(@"C:\Users\Patri\OneDrive\Dokumenter\Visual Studio 2015\Projects\CompareTexts\Nyheder_Database\All_articles\False\03_11_2016_Higgs_News_Network_BBC_And_CNN_News_Caught_Staging_FAKE_News.txt");
var test2 = new LoadEachWordToList(@"C:\Users\Patri\OneDrive\Dokumenter\Visual Studio 2015\Projects\CompareTexts\Nyheder_Database\All_articles\False\05_03_2017_Koran_Bible_Same1.txt");
var test3 = new LoadEachWordToList(@"C:\Users\Patri\OneDrive\Dokumenter\Visual Studio 2015\Projects\CompareTexts\Nyheder_Database\All_articles\True\20_04_2017_Syria_news_BBC.txt");
string[] tags = { "Brexit" };
var testJaccard = new CompareTextUsingJaccardSimilarity(test1.Words, null);
Console.WriteLine("False: " + testJaccard.FalseArticlesSimilarity);
Console.WriteLine("True: " + testJaccard.TrueArticlesSimilarity);
var testCosine = new CompareTextUsingCosineSimilarity(test2.Words, tags);
Console.WriteLine("\n");
Console.WriteLine("False: " + (decimal)testCosine.FalseArticlesSimilarity);
Console.WriteLine("True: " + (decimal)testCosine.TrueArticlesSimilarity);
Console.ReadLine();
}
public void FullImplementationFalseArticleSimilarity(string fileName, string tag, string trueOrFalse, int expectedResult,
params string[] tags)
{
string path = GetFilePath(fileName, tag, trueOrFalse);
var text = new LoadEachWordToList(path);
var test = new CompareTextUsingCosineTestClass(text.Words, tags);
Assert.AreEqual(expectedResult, test.FalseArticlesSimilarity);
}
static void Main(string[] args)
{
Stopwatch stopwatch = new Stopwatch();
Console.WriteLine(Stopwatch.IsHighResolution);
stopwatch.Start();
#region LevenshteinDistance
//Indlæser hver linje af teksten til en liste af strings
LoadStringToList tekstA = new LoadStringToList(@"C:\Users\Patri\Dropbox\Projekt\P2\Program\Nyheder_Database\Koran_Bible_Same1.txt");
LoadStringToList tekstB = new LoadStringToList(@"C:\Users\Patri\Dropbox\Projekt\P2\Program\Nyheder_Database\Koran_Bible_Same2.txt");
// Beregner LevenshteinDistance
LevenshteinDistance levDis = new LevenshteinDistance(tekstA.Lines, tekstB.Lines, tekstA.GetAmountOfChars(),
tekstB.GetAmountOfChars(), tekstA.LinesInText, tekstB.LinesInText);
levDis.Print(); // Printer LevenshteinDistance mellem de to tekster
#endregion
Console.WriteLine(stopwatch.ElapsedMilliseconds);
stopwatch.Restart();
#region CosineDistance
//Indlæser hvert ord fra teksten til en liste af strings
LoadEachWordToList TextA = new LoadEachWordToList(@"C:\Users\Patri\Dropbox\Projekt\P2\Program\Nyheder_Database\Koran_Bible_Same1.txt");
LoadEachWordToList TextB = new LoadEachWordToList(@"C:\Users\Patri\Dropbox\Projekt\P2\Program\Nyheder_Database\Koran_Bible_Same2.txt");
CalculateCosine CalcCos = new CalculateCosine(TextA.Words, TextB.Words); // Beregner CosineDistance
CalcCos.Print(); // Printer CosineSimilarity mellem de to tekster
#endregion
Console.WriteLine(stopwatch.ElapsedMilliseconds);
stopwatch.Restart();
#region JaccardDistance
JaccardSimilarity nytekstA = new JaccardSimilarity(TextA.Words, TextB.Words); // TextA og TextB er indlæst i #region CosineDistance - (LoadEachWordToList)
nytekstA.Print(); // Printer JaccardSimilarity mellem de to tekster
#endregion
Console.WriteLine(stopwatch.ElapsedMilliseconds);
stopwatch.Stop();
#region TryDiffrentLoadMethod (Resources)
//var TekstA = new List<string>();
//var TekstB = new List<string>();
//string[] stringSeparators = { ",", ".", "!", "?", ";", ":", " ", "-", "\"", "(", ")" };
//TekstA = Resources.Pizzagate1.Split(stringSeparators, StringSplitOptions.RemoveEmptyEntries).ToList();
//TekstB = Resources.Pizzagate2.Split(stringSeparators, StringSplitOptions.RemoveEmptyEntries).ToList();
//CalculateCosine CalcCos2 = new CalculateCosine(TekstA, TekstB); // Beregner CosineDistance
//CalcCos2.Print(); // Printer CosineDistance mellem de to tekster
#endregion
Console.ReadLine();
}
public void FullImplementationTrueArticleSimilarity(string fileName, string tag, string trueOrFalse, decimal expectedResult,
params string[] tags)
{
string path = GetFilePath(fileName, tag, trueOrFalse);
var text = new LoadEachWordToList(path);
var test = new CompareTextUsingJaccardTestClass(text.Words, tags);
decimal result = Math.Round(test.TrueArticlesSimilarity, 5);
Assert.AreEqual(expectedResult, result);
}
private void calculateValue_Click(object sender, EventArgs e)
{
LoadEachWordToList tekstA = new LoadEachWordToList(@"C:\Users\Aryan\Dropbox\P1 Projekt\P2\Program\Nyheder_Database\" + firstText);
LoadEachWordToList tekstB = new LoadEachWordToList(@"C:\Users\Aryan\Dropbox\P1 Projekt\P2\Program\Nyheder_Database\" + secondText);
JaccardSimilarity nytekstA = new JaccardSimilarity();
JaccardSimilarity nytekstB = new JaccardSimilarity();
nytekstA.GetShingles(tekstA.Words);
nytekstB.GetShingles(tekstB.Words);
jaccardSimilarity = nytekstA.CalculateSimilarity(nytekstA.shingles, nytekstB.shingles);
this.label1.Text = "The Jaccard similarity of the two texts is: " + jaccardSimilarity;
}
// Returns the greatest JaccardSimilarity optained by comparing the text to texts in the directory
public override decimal CompareWithTexts(List <string> paths)
{
decimal greatestSimilarity = 0;
foreach (string path in paths) // Gets JaccardSimilarity for all false articles
{
var databaseText = new LoadEachWordToList(path);
var compareTexts = new JaccardSimilarity(TextToBeCompared, databaseText.Words);
// Happens if the jaccardSimilarity between the two current texts are the greatest so far
if (compareTexts.Similarity > greatestSimilarity)
{
greatestSimilarity = compareTexts.Similarity;
}
}
return(greatestSimilarity);
}
// Returns the greatest CosineSimilarity optained by comparing the text to all texts in the directory
public override decimal CompareWithTexts(List <string> paths)
{
int greatestSimilarity = 0;
foreach (string path in paths) // Gets CosineSimilarity for all false articles
{
var databaseText = new LoadEachWordToList(path);
var compareTexts = new CalculateCosine(TextToBeCompared, databaseText.Words);
// Happens if the CosineSimilarity between the two current texts are the greatest so far
if (compareTexts.Procent > greatestSimilarity)
{
greatestSimilarity = compareTexts.Procent;
}
}
return((decimal)greatestSimilarity);
}
