private static float GetDocumentDistance(DocumentVectorTest doc1, DocumentVectorTest doc2)
{
var dist = 0.0f;
for (var i = 0; i < doc1.VectorSpace.Length; i++)
{
dist += (float)Math.Pow((double)(doc1.VectorSpace[i] - doc2.VectorSpace[i]), 2.0);
}
dist = (float)Math.Pow((double)dist, 0.5);
return(dist);
}
private static float Move(DocumentVectorTest documentVector1, DocumentVectorTest documentVector2, float G)
{
int length = documentVector1.VectorSpace.Count();
float[] d = new float[length];
var distance = GetDocumentDistance(documentVector1, documentVector2);
for (int i = 0; i < length; i++)
{
d[i] = documentVector2.VectorSpace[i] - documentVector1.VectorSpace[i];
}
for (var i = 0; i < length; i++)
{
documentVector1.VectorSpace[i] = documentVector1.VectorSpace[i] + distance * (G / (float)Math.Pow(distance, 3.0)); //--last function
//documentVector1.VectorSpace[i] = documentVector1.VectorSpace[i] + d[i] * (G / (float)Math.Pow(distance, 3.0));
}
return(distance);
}
public static List <DocumentVectorTest> CreatingTheDocVectorCollection(string fileName)
{
List <DocumentVectorTest> TestDocVectorList = new List <DocumentVectorTest>();
int number_of_lines = 0;
char[] separators = { ' ', ',', '/', '.', '-', '\t' };
const Int32 BufferSize = 128;
using (var fileStream = File.OpenRead(fileName))
using (var streamReader = new StreamReader(fileStream, Encoding.UTF8, true, BufferSize))
{
String line;
while ((line = streamReader.ReadLine()) != null)
{
number_of_lines++;
DocumentVectorTest testDoc = new DocumentVectorTest();
testDoc.VectorSpace = new float[2];
var lineitem = line.TrimStart(' ');
var lineitem2 = lineitem.TrimEnd(' ');
var items = lineitem2.Split(separators);
testDoc.VectorSpace[0] = float.Parse(items[0]);
testDoc.VectorSpace[1] = float.Parse(items[1]);
/*
* for (int i = 0; i < items.Count(); i++)
* {
* if (!items[i].Contains(" "))
* testDoc.VectorSpace[i] = float.Parse(items[i]);
* else
* continue;
* }
*/
//testDoc.Content = "testDataPoint" + number_of_lines;
testDoc.Content = number_of_lines.ToString();
TestDocVectorList.Add(testDoc);
}
}
return(TestDocVectorList);
}
public static List <TestCentroid> UpdateMeans(List <TestCentroid> fillCentroidCollection, List <DocumentVectorTest> vectorSpace)
{
List <TestCentroid> result = new List <TestCentroid>();
List <DocumentVectorTest> newVectorSpace = new List <DocumentVectorTest>(vectorSpace);
int length = vectorSpace[0].VectorSpace.Length;
float[] newVectorSpaceArray = new float[length];
float[] minDistancesToCluster = new float[0];
for (int i = 0; i < length; i++)
{
newVectorSpaceArray[i] = 0.0F;
}
for (int c = 0; c < fillCentroidCollection.Count; c++)
{
for (int gd = 0; gd < fillCentroidCollection[c].GroupedDocument.Count; gd++)
{
for (int k = 0; k < fillCentroidCollection[c].GroupedDocument[gd].VectorSpace.Length; k++)
{
newVectorSpaceArray[k] += fillCentroidCollection[c].GroupedDocument[gd].VectorSpace[k];
}
}
}
for (int c1 = 0; c1 < fillCentroidCollection.Count; c1++)
{
for (int gd1 = 0; gd1 < fillCentroidCollection[c1].GroupedDocument.Count; gd1++)
{
for (int k1 = 0; k1 < fillCentroidCollection[c1].GroupedDocument[gd1].VectorSpace.Length; k1++)
{
newVectorSpaceArray[k1] = newVectorSpaceArray[k1] / fillCentroidCollection[c1].GroupedDocument.Count;
}
}
}
float minDist = 0.1F;
float currentValue = 0.1F;
int index = 0;
for (int i = 0; i < fillCentroidCollection.Count; i++)
{
minDistancesToCluster = new float[fillCentroidCollection[i].GroupedDocument.Count];
for (int j = 0; j < fillCentroidCollection[i].GroupedDocument.Count; j++)
{
//minDistancesToCluster = new float[fillCentroidCollection[i].GroupedDocument.Count];
minDistancesToCluster[j] = SimilarityMatrixCalculations.FindEuclideanDistance(fillCentroidCollection[i].GroupedDocument.First().VectorSpace, fillCentroidCollection[i].GroupedDocument[j].VectorSpace);
//}
for (int z = 0; z < minDistancesToCluster.Length; z++)
{
currentValue = minDistancesToCluster[z];
if (currentValue <= minDist && currentValue != 0)
{
minDist = currentValue;
index = z;
}
//here we must to find the closest document to new vectorSpace;
//for all docs in cluster create the vectorSpace
}
/*
* DocumentVector newClusterCenter = fillCentroidCollection[i].GroupedDocument[index];
* fillCentroidCollection[i].GroupedDocument.Clear();
* fillCentroidCollection[i].GroupedDocument.Add(newClusterCenter);
*/
}
DocumentVectorTest newClusterCenter = fillCentroidCollection[i].GroupedDocument[index];
index = 0;
fillCentroidCollection[i].GroupedDocument.Clear();
fillCentroidCollection[i].GroupedDocument.Add(newClusterCenter);
}
minDistancesToCluster = new float[0];
result = new List <TestCentroid>(fillCentroidCollection);
return(result);
}
/*
* private static bool UpdateMeans(List<DocumentVectorTest> data, List<TestCentroid> clustering, List<TestCentroid> means)
* {
* int numClusters = means.Count;
* bool changed = false;
*
* List<TestCentroid> newClustering = new List<TestCentroid>(numClusters);
* float[] distances = new float[numClusters];
* for(int i = 0; i < data.Count; i++)
* {
* for (int k = 0; k < numClusters; k++)
* distances[k] = Logic.ClusteringAlgorithms.SimilarityMatrixCalculations.FindEuclideanDistance(data[i].VectorSpace, means[k].GroupedDocument[0].VectorSpace);
*
* int newClusterID = MinIndex(distances);
*
* if(newClustering[newClusterID].GroupedDocument[0].Content != newClustering[i].GroupedDocument[0].Content)
* {
* changed = true;
* newClustering[i] = newClustering[newClusterID];
* }
* }
* if (changed == false)
* return false;
*
*
* List<TestCentroid> clusterCounts = new List<TestCentroid>(numClusters);
* for(int i = 0; i < data.Count; i++)
* {
* var cluster = newClustering[i];
++clusterCounts.IndexOf(cluster);
* }
*
* return true;
* }
*/
#endregion
#region OldUpdateMeans
/*
* public static bool UpdateMeans(List<DocumentVectorTest> data, List<TestCentroid> clustering, List<TestCentroid> means)
* {
* List<DocumentVectorTest> newVectorSpace = new List<DocumentVectorTest>(data);
* int vectorSpaceLength = data[0].VectorSpace.Length;
* int numClusters = means.Count;
* int[] clusterCounts = new int[numClusters];
*
#region AccordingToTheAuthorsWeCanOmitThis
*
* for(int c = 0; c < data.Count; c++)
* {
* var cluster = data[c];
++clusterCounts[data.IndexOf(cluster)];
* }
*
#endregion
*
* for (int k = 0; k < numClusters; k++)
* if (clusterCounts[k] == 0)
* return false;
*
* for (int cl = 0; cl < means.Count; cl++)
* clustering[cl].GroupedDocument.Clear();
*
* for(int i = 0; i < data.Count; i++)
* {
* var cluster = clustering.ElementAt(i);
* for (int j = 0; j < cluster.GroupedDocument.Count; j++)
* for (int v = 0; v < cluster.GroupedDocument[j].VectorSpace.Length; v++)
* clustering[clustering.IndexOf(cluster)].GroupedDocument[j].VectorSpace[v] += data[i].VectorSpace[v];
* }
*
* for (int k = 0; k < clustering.Count; ++k)
* for (int j = 0; j < clustering[k].GroupedDocument.Count; j++)
* for (int z = 0; z < clustering[k].GroupedDocument[j].VectorSpace.Length; z++)
* clustering[k].GroupedDocument[j].VectorSpace[z] /= clusterCounts[k];
* return true;
* }
*/
#endregion
public static Tuple <bool, List <TestCentroid> > UpdateMeans(List <DocumentVectorTest> data, List <TestCentroid> clustering, List <TestCentroid> means)
{
Tuple <bool, List <TestCentroid> > result;
int numClusters = means.Count;
bool changed = false;
float minVal = float.MaxValue;
int minIndex = 0;
List <TestCentroid> centers = new List <TestCentroid>(numClusters);
centers = TestCentroidInitializer(numClusters);
List <TestCentroid> newMeans = new List <TestCentroid>(numClusters);
newMeans = TestCentroidInitializer(numClusters);
for (int i = 0; i < means.Count; i++)
{
DocumentVectorTest newCenter = new DocumentVectorTest();
newCenter.VectorSpace = new float[means[i].GroupedDocument[0].VectorSpace.Length];
for (int j = 0; j < means[i].GroupedDocument.Count; j++)
{
for (int v = 0; v < means[i].GroupedDocument[j].VectorSpace.Length; v++)
{
newCenter.VectorSpace[v] += means[i].GroupedDocument[j].VectorSpace[v];
}
for (int z = 0; z < means[i].GroupedDocument[j].VectorSpace.Length; z++)
{
newCenter.VectorSpace[z] = newCenter.VectorSpace[z] / means[i].GroupedDocument.Count;
}
}
newCenter.Content = "ideal center of " + i.ToString() + " cluster";
centers[i].GroupedDocument.Add(newCenter);
}
for (int i = 0; i < clustering.Count; i++)
{
float[] distances = new float[clustering[i].GroupedDocument.Count];
for (int j = 0; j < clustering[i].GroupedDocument.Count; j++)
{
distances[j] = Distance(clustering[i].GroupedDocument[j].VectorSpace, centers[i].GroupedDocument[0].VectorSpace);
}
for (int j = 0; j < distances.Length; j++)
{
if (distances[j] < minVal)
{
minVal = distances[j];
minIndex = j;
}
}
newMeans[i].GroupedDocument.Add(clustering[i].GroupedDocument[minIndex]);
minVal = float.MaxValue;
minIndex = 0;
}
for (int i = 0; i < clustering.Count; i++)
{
if (means[i].GroupedDocument[0].Content == newMeans[i].GroupedDocument[0].Content)
{
changed = true;
}
else
{
changed = false;
}
}
means = newMeans;
result = new Tuple <bool, List <TestCentroid> >(changed, means);
return(result);
}
