A Maximum Likelihood Framework for Integrating Taxonomies

Abstract:

Many approaches have been proposed for the problem of mapping classes from a source taxonomy to classes in a master taxonomy. Most of these techniques however ignore the hierarchical structure of the taxonomies. Our approach exploits this structure to obtain a more natural mapping between the classes. Furthermore, unlike previous work, our technique also inserts source classes into appropriate places of the master hierarchy creating new categories if needed. In this paper, we propose a Maximum Likelihood based framework to integrate classes from a source taxonomy into a master hierarchy. We evaluate our approach on text and hyperspectral datasets.