Taxonomy grounded aggregation of classifiers with different label sets
share
We describe the problem of aggregating the label predictions of diverse classifiers using a class taxonomy. Such a taxonomy may not have been available or referenced when the individual classifiers were designed and trained, yet mapping the output labels into the taxonomy is desirable to integrate the effort spent in training the constituent classifiers. A hierarchical taxonomy representing some domain knowledge may be different from, but partially mappable to, the label sets of the individual classifiers. We present a heuristic approach and a principled graphical model to aggregate the label predictions by grounding them into the available taxonomy. Our model aggregates the labels using the taxonomy structure as constraints to find the most likely hierarchically consistent class. We experimentally validate our proposed method on image and text classification tasks.
READ FULL TEXT