/**
* Returns the k instances of the given data set that are the closest to the instance that is
* given as a parameter.
*
* @param dm the distance measure used to calculate the distance between instances
* @param inst the instance for which we need to find the closest
* @return the instances from the supplied data set that are closest to the supplied instance
*/
@Override
public Set<Instance> kNearest(int k, Instance inst, DistanceMeasure dm) {
Map<Instance, Double> closest = new HashMap<Instance, Double>();
double max = dm.getMaxValue();
for (Instance tmp : this) {
double d = dm.measure(inst, tmp);
if (dm.compare(d, max) && !inst.equals(tmp)) {
closest.put(tmp, d);
if (closest.size() > k) max = removeFarthest(closest, dm);
}
}
return closest.keySet();
}
/*
* Removes the element from the vector that is farthest from the supplied
* element.
*/
private double removeFarthest(Map<Instance, Double> vector, DistanceMeasure dm) {
Instance tmp = null; // ; = vector.get(0);
double max = dm.getMinValue();
// System.out.println("minvalue:"+max);
for (Instance inst : vector.keySet()) {
double d = vector.get(inst);
if (dm.compare(max, d)) {
max = d;
tmp = inst;
}
// System.out.println("d="+d+"\t"+max);
}
vector.remove(tmp);
return max;
}
/** XXX DOC */
public double score(Dataset[] datas) {
Instance[] centroids = new Instance[datas.length];
for (int i = 0; i < datas.length; i++) {
centroids[i] = DatasetTools.average(datas[i]);
}
double sum = 0;
for (int i = 0; i < datas.length; i++) {
for (int j = 0; j < datas[i].size(); j++) {
double error = dm.measure(datas[i].instance(j), centroids[i]);
sum += error;
}
}
return sum;
}