Classifier Combination Using Random Walks on the Space of Concepts

JORGE SÁNCHEZ; JAVIER REDOLFI

Buenos Aires

Congreso; 7th Iberoamerican Congress on Pattern Recognition, CIARP 2012,; 2012

Universidad de Buenos Aires

We propose a novel approach for the combination of classifiers based on two commonly adopted strategies in multiclass classification: one-vs-all and one-vs-one. The method relies on establishing the relevance of nodes in a graph defined in the space of concepts. Following a similar approach as in the ranking of websites, the relative strength of the nodes is given by the stationary distribution of a Markov chain defined on that graph. The proposed approach do not requires the base classifiers to provide calibrated probabilities. Experiments on the challenging problem of multiclass image classification show the potentiality of our approach.