internal Dictionary <string, int> GetPsychProfile(string[] documents)
{
string[] d = new string[] { documents.Aggregate((aggr, cur) => aggr + " " + cur) };
double[][] inputs = TFIDF.Normalize(TFIDF.Transform(d));
int[] output = new int[9];
Dictionary <string, int> result = new Dictionary <string, int>(9);
output[0] = modelDenial.Decide(inputs[0]);
output[1] = modelRepression.Decide(inputs[0]);
output[2] = modelRegression.Decide(inputs[0]);
output[3] = modelCompensation.Decide(inputs[0]);
output[4] = modelProjection.Decide(inputs[0]);
output[5] = modelDisplacement.Decide(inputs[0]);
output[6] = modelRationalization.Decide(inputs[0]);
output[7] = modelReactionFormation.Decide(inputs[0]);
output[8] = output[0] + output[1] + output[2] + output[3] + output[4] + output[5] + output[6] + output[7];
output[8] = Convert.ToInt32(Math.Round((double)output[8] / 8));
result.Add("Отрицание", output[0]);
result.Add("Вытеснение", output[1]);
result.Add("Регрессия", output[2]);
result.Add("Компенсация", output[3]);
result.Add("Проекция", output[4]);
result.Add("Замещение", output[5]);
result.Add("Рационализация", output[6]);
result.Add("Гиперкомпенсация", output[7]);
result.Add("Общий уровень", output[8]);
return(result);
}
public string TFIDF_score(string[] webcontent)
{
TFIDF idf = new TFIDF();
double[][] inputs = idf.Transform(webcontent, 0);
inputs = TFIDF.Normalize(inputs);
string final = "";
for (int index = 0; index < inputs.Length; index++)
{
final = final + webcontent[index] + "\n";
foreach (double value in inputs[index])
{
final = value + ",";
}
final = final + "\n\n";
}
return(final);
}
/// <summary>
/// Generate KMean Clustering from text file and save to file
/// </summary>
/// <param name="filePath"> path of CVS file that contains data</param>
/// <param name="columnname">Name of Column that has text to be considerd</param>
/// <param name="ct">Column type</param>
/// <param name="k">Number of clusters</param>
/// <param name="maxiter">Maximun number of iterations</param>
/// <param name="outputfile">Path where to save clustered data.</param>
public void doIt(string filePath, string columnname, List <Type> ct, int k, int maxiter, string outputfile)
{
Common c = new Common();
string[,] dt = c.getStringMatrix(filePath, ',', ct);
string[] inp = c.getStringVector(dt, columnname);
List <string> vocab = new List <string>();
double[][] inputs = TFIDF.Transform(inp, ref vocab);
inputs = TFIDF.Normalize(inputs);
double[][] tf_labels = Common.getArrayRange(inputs, 0, 1, dt.Length);
double[][] tf_centroids = smart_centroid_initalization(inputs, k, 0);
List <double> heter = new List <double>();
Tuple <double[][], int[]> result = kmeans(tf_labels, k, tf_centroids, maxiter, ref heter);
List <List <double[]> > lld = new List <List <double[]> >();
List <List <string> > lll = new List <List <string> >();
for (int i = 0; i < k; i++)
{
lld.Add(new List <double[]>());
lll.Add(new List <string>());
}
for (int i = 0; i < result.Item2.Length; i++)
{
lld[result.Item2[i]].Add(tf_labels[i]);
}
string[] vocabul = new string[k];
for (int i = 0; i < k; i++)
{
List <double> ccc = result.Item1[i].ToList();
ccc.Sort();
ccc.Reverse();
for (int j = 0; j < k; j++)
{
vocabul[i] += vocab[Array.IndexOf(result.Item1[i], ccc[j])] + ": " + result.Item1[i][Array.IndexOf(result.Item1[i], ccc[j])] + " ";
}
}
using (TextWriter writer = File.CreateText(outputfile))
{
for (int i = 0; i < result.Item2.Length; i++)
{
writer.WriteLine(inp[i] + "," + vocabul[result.Item2[i]]);
lll[result.Item2[i]].Add(inp[i]);
}
}
Console.WriteLine("Finished!!!");
Console.ReadKey();
}
