public static double MaxEntropy(IRumorNode <TNode, TRumor> node)
{
if (!node.Memory.Any())
{
return(0.0);
}
return(Math.Log(1.0 / node.Memory.First().InformationSpace, 2));
}
public static double DistortionChance <TNode, TRumor>(this IRumorNode <TNode, TRumor> node, float K) where TNode : INetworkNode <TNode> where TRumor : IRumor
{
var Hn = Entropy <TNode, TRumor> .ShannonEntropy(node);
var Hmax = Entropy <TNode, TRumor> .MaxEntropy(node);
return(1.0 / (Math.Exp(K * (Hmax - Hn) / Hmax) + 1.0));
}
public static IDictionary <IRumor, double> InformationFrequency <TNode, TRumor>(this IRumorNode <TNode, TRumor> node) where TNode : INetworkNode <TNode> where TRumor : IRumor
{
int current;
var counts = new Dictionary <IRumor, int>();
foreach (var rumor in node.Memory)
{
if (counts.TryGetValue(rumor, out current))
{
counts[rumor] = current + 1;
}
else
{
counts[rumor] = 0;
}
}
var total = (double)counts.Sum(x => x.Value);
return(counts.ToDictionary(x => x.Key, x => (double)x.Value / total));
}
public static double ShannonEntropy(IRumorNode <TNode, TRumor> node)
{
var f = node.InformationFrequency();
return(-f.Sum(fi => fi.Value * Math.Log(fi.Value, 2)));
}
