On exploiting hierarchical label structure with pairwise classifiers

Johannes Fürnkranz, Jan Frederik Sima
2011 SIGKDD Explorations  
The goal of this work was to test whether the performance of a regular pairwise classifier can be improved when additional information about the hierarchical class structure is added to the training sets. Somewhat surprisingly, the additional information seems to hurt the performance. We explain this with the fact that the structure of the class hierarchy is not reflected in the distribution of the instances.
doi:10.1145/1964897.1964903 fatcat:vxw53xx6drgwdkdagdev5cdjoe