PDF version of this entire document

Weighted Similarity Measure

There is an experiment underway where weighting is applied based on the lessons taken from an analysis with the new tool. So far there are no classification mistakes, but it ought to run further and compare more pairs before meaningful conclusions can be drawn and the weighing tested to a point where ideal values are found.

Figure: Overview of the debugging process with examples from two true pairs (same person) and one false pair (different people), with the eye component discrepancies shown at the top and the nose at the bottom

Figure: The ROC curve obtained by using a weighted form of the similarity measure