static void Main(string[] args)
{
SampleColection sampleColection = new SampleColection();
sampleColection.addSamples();
sampleColection.normalizeSamples();
foreach (double att in sampleColection.samples[0].attributes)
{
Console.WriteLine(att);
}
Console.WriteLine(sampleColection.samples[0].decision);
System.Console.ReadLine();
}
Dictionary <int, double> countSumOfKminDistances(Sample sampleToCheck, SampleColection data, int k)
{
Dictionary <int, List <double> > distances = countDistances(sampleToCheck, data);
double sumOfKminDistances = 0;
var minDistances = new Dictionary <int, double>();
foreach (var item in distances)
{
item.Value.Sort();
for (int i = 0; i < k; i++)
{
sumOfKminDistances += item.Value[i];
}
minDistances.Add(item.Key, sumOfKminDistances);
}
return(minDistances);
}
public int?chooseDecision(Sample sampleToCheck, SampleColection data, int k)
{
Dictionary <int, double> minDistances = countSumOfKminDistances(sampleToCheck, data, k);
double min = minDistances.First().Value;
int decision = minDistances.First().Key;
for (int i = 1; i < minDistances.Count; i++)
{
decision = minDistances.Keys.ElementAt(i);
if (minDistances[decision] < min)
{
min = minDistances[decision];
}
}
if (isOnlyOneMin(minDistances, min))
{
return(decision);
}
return(null);
}
Dictionary <int, List <double> > countDistances(Sample sampleToCheck, SampleColection data)
{
var distances = new Dictionary <int, List <double> >();
double distance;
foreach (Sample sample in data.samples)
{
distance = Metrics.euklideanMetric(sampleToCheck.attributes, sample.attributes);
if (!distances.ContainsKey(sample.decision))
{
distances.Add(sample.decision, new List <double> {
distance
});
}
else
{
distances[sample.decision].Add(distance);
}
}
return(distances);
}