Get access

THEORETICAL FOUNDATIONS AND EXPERIMENTAL RESULTS FOR A HIERARCHICAL CLASSIFIER WITH OVERLAPPING CLUSTERS

Authors


Igor T. Podolak, Łojasiewicza 6, 30-348 Cracow, Poland; e-mail: igor.podolak@uj.edu.pl. This research was partially funded by the Polish National Science Center grant no. 6548/B/T02/2011/40.

Abstract

This paper proposes a classification framework based on simple classifiers organized in a tree-like structure. It is observed that simple classifiers, even though they have high error rate, find similarities among classes in the problem domain. The authors propose to trade on this property by recognizing classes that are mistaken and constructing overlapping subproblems. The subproblems are then solved by other classifiers, which can be very simple, giving as a result a hierarchical classifier (HC). It is shown that HC, together with the proposed training algorithm and evaluation methods, performs well as a classification framework. It is also proven that such constructs give better accuracy than the root classifier it is built upon.

Ancillary