/**
* Run the randomization test
*
* @param baseline
* @param target
* @return
*/
public double test(HashMap<String, Double> target, HashMap<String, Double> baseline) {
double[] b = new double[baseline.keySet().size()]; // baseline
double[] t = new double[target.keySet().size()]; // target
int c = 0;
for (String key : baseline.keySet()) {
b[c] = baseline.get(key).doubleValue();
t[c] = target.get(key).doubleValue();
c++;
}
double trueDiff = Math.abs(BasicStats.mean(b) - BasicStats.mean(t));
double pvalue = 0.0;
double[] pb = new double[baseline.keySet().size()]; // permutation of baseline
double[] pt = new double[target.keySet().size()]; // permutation of target
for (int i = 0; i < nPermutation; i++) {
char[] bits = randomBitVector(b.length).toCharArray();
for (int j = 0; j < b.length; j++) {
if (bits[j] == '0') {
pb[j] = b[j];
pt[j] = t[j];
} else {
pb[j] = t[j];
pt[j] = b[j];
}
}
double pDiff = Math.abs(BasicStats.mean(pb) - BasicStats.mean(pt));
if (pDiff >= trueDiff) pvalue += 1.0;
}
return pvalue / nPermutation;
}
@Override
public final Explanation explain(BasicStats stats) {
return Explanation.match(
lambda(stats),
getClass().getSimpleName() + ", computed from: ",
Explanation.match(stats.getDocFreq(), "docFreq"),
Explanation.match(stats.getNumberOfDocuments(), "numberOfDocuments"));
}
/**
* Returns an explanation for the normalized term frequency.
*
* <p>The default normalization methods use the field length of the document and the average field
* length to compute the normalized term frequency. This method provides a generic explanation for
* such methods. Subclasses that use other statistics must override this method.
*/
public Explanation explain(BasicStats stats, float tf, float len) {
Explanation result = new Explanation();
result.setDescription(getClass().getSimpleName() + ", computed from: ");
result.setValue(tfn(stats, tf, len));
result.addDetail(new Explanation(tf, "tf"));
result.addDetail(new Explanation(stats.getAvgFieldLength(), "avgFieldLength"));
result.addDetail(new Explanation(len, "len"));
return result;
}
@Override
public final float lambda(BasicStats stats) {
return (stats.getDocFreq() + 1F) / (stats.getNumberOfDocuments() + 1F);
}