/// <summary>
/// Detect Cypher Key based on Correlation Coefficient
/// </summary>
/// <param name="encryptedText">encryptedText</param>
/// <returns>expected key</returns>
public int DetectCypherKey(string encryptedText)
{
var characterCounts = new List <CharacterWithValue <int> >();
var characterFrequencies = new List <CharacterWithValue <double> >();
foreach (var item in StaticData.EnglishLanguageFrequencies)
{
// calculate character existing count
int count = encryptedText.Select(i => i).Count(i => i == item.Character);
characterCounts.Add(new CharacterWithValue <int>()
{
Character = item.Character,
Value = count
});
// calculate frequency of each character on the phrase (percentage)
float frequency = (float)count / (float)encryptedText.Length;
characterFrequencies.Add(new CharacterWithValue <double>()
{
Character = item.Character,
Value = frequency
});
}
double bestCorrelation = -1;
int bestShiftIndexKey = -1;
int shiftIndexKey = 0;
// get basic language characters frequencies
double[] basicLangauageFrequencies = StaticData.EnglishLanguageFrequencies.Select(i => i.Value).ToArray();
// try all possibillities for shiting index
while (shiftIndexKey < basicLangauageFrequencies.Count() - 1)
{
// shift frequencies
double[] shiftedFrequencies = ArrayManupilation.ShiftByIndex <double>(characterFrequencies.Select(i => i.Value).ToArray(), shiftIndexKey);
// calculate new correlation
double correlation = CorrelationCoefficient.Calculate(basicLangauageFrequencies, shiftedFrequencies, basicLangauageFrequencies.Count());
// find best correlation exist (best correlation is the expected key)
if (correlation > bestCorrelation)
{
bestCorrelation = correlation;
bestShiftIndexKey = shiftIndexKey;
}
shiftIndexKey++;
}
return(basicLangauageFrequencies.Count() - bestShiftIndexKey);
}
/**
* Load and parse data from dat_mes.csv
*/
internal static void LoadData(StudyData studyData)
{
if (IsDataLoaded)
{
return;
}
List <string[]> rows = new List <string[]>();
using (StreamReader reader = File.OpenText("Data/dat_mes.csv"))
{
while (reader.Peek() >= 0)
{
string line = reader.ReadLine();
string[] rowArray = CSVRowToStringArray(line);
if (rowArray.Length > 0)
{
rows.Add(rowArray);
}
}
}
string[] headers = rows[0];
rows.Remove(headers);
/**
* Parse each row array into a Study object.
* Assumes CSV column ordering:
* id,PublicationYear,n,r,VariablesControlled,StudyDesign,
* AdherenceMeasure,ConscientiousnessMeasure ,MeanAge,MethodologicalQuality
*/
foreach (string[] row in rows)
{
PublicationYear publicationYear = studyData.PublicationYears.AddUnique(row[1]);
CorrelationCoefficient correlationCoefficient = studyData.CorrelationCoefficients.AddUnique(row[2]);
SampleSize sampleSize = studyData.SampleSizes.AddUnique(row[3]);
VariablesControlled variablesControlled = studyData.VariablesControlled.AddUnique(row[4]);
StudyDesign studyDesign = studyData.StudyDesigns.AddUnique(row[5]);
AdherenceMeasure adherenceMeasure = studyData.AdherenceMeasures.AddUnique(row[6]);
ConscientiousnessMeasure conscientiousnessMeasure = studyData.ConscientiousnessMeasures.AddUnique(row[7]);
MeanAge meanAge = studyData.MeanAges.AddUnique(row[8]);
MethodologicalQuality methodologicalQuality = studyData.MethodologicalQualities.AddUnique(row[9]);
Study newStudy = new Study
{
Id = row[0],
PublicationYear = publicationYear,
CorrelationCoefficient = correlationCoefficient,
SampleSize = sampleSize,
VariablesControlled = variablesControlled,
StudyDesign = studyDesign,
AdherenceMeasure = adherenceMeasure,
ConscientiousnessMeasure = conscientiousnessMeasure,
MeanAge = meanAge,
MethodologicalQuality = methodologicalQuality
};
studyData.Studies.Add(newStudy);
}
IsDataLoaded = true;
}
