public void LearnHelper(TimeSeries ts, float p /*pearson*/, string f1, string f2, Point[] ps)
{
if (p > threshold)
{
int len = ts.getNumOfTimesteps();
correlatedFeatures c = new correlatedFeatures(f1, f2, p, AnomalyDetectionUtil.LinearReg(ps, len), findThreshold(ps, len, AnomalyDetectionUtil.LinearReg(ps, len)) * (float)1.1);
cf.Add(c);
}
}
public void UpdateAnomaly(List <AnomalyReport> ar, Point p, correlatedFeatures cor, int timeStepIndex, string f1, string f2)
{
if (AnomalyDetectionUtil.Dev(p, cor.LineReg) > cor.Threshhold)
{
string desc = f1 + "-" + f2;
long timeS = timeStepIndex + 1;
AnomalyReport rep = new AnomalyReport(desc, timeS);
ar.Add(rep);
}
}
public List <correlatedFeatures> LearnNormal(TimeSeries ts)
{
List <string> atts = ts.getFeaturesNames();
int len = ts.getNumOfTimesteps();
float[,] vals = new float[atts.Count, len];
for (int i = 0; i < atts.Count; i++)
{
List <float> x = ts.getAllFeatureValues(atts[i]);
for (int j = 0; j < len; j++)
{
vals[i, j] = x[j];
}
}
for (int i = 0; i < atts.Count; i++)
{
string f1 = atts[i];
float max = 0;
int jmax = 0;
for (int j = i + 1; j < atts.Count; j++)
{
float[] valFeatureI = Enumerable.Range(0, vals.GetLength(1))
.Select(x => vals[i, x])
.ToArray();
float[] valFeatureJ = Enumerable.Range(0, vals.GetLength(1))
.Select(x => vals[j, x])
.ToArray();
float retVal = AnomalyDetectionUtil.Pearson(valFeatureI, valFeatureJ, len);
float p = Math.Abs(AnomalyDetectionUtil.Pearson(valFeatureI, valFeatureJ, len));
if (p > max)
{
max = p;
jmax = j;
}
}
string f2 = atts[jmax];
Point[] ps = toPoints(ts.getAllFeatureValues(f1), ts.getAllFeatureValues(f2));
LearnHelper(ts, max, f1, f2, ps);
}
return(cf);
}