Optimising ECOC matrices in multi-class classification problems

Official URL: http://risc01.sabanciuniv.edu/record=b1534414 (Table of Contents)

Error Correcting Output Coding (ECOC) is a multi-class classiffication technique in which multiple binary classiffiers are trained according to a preset code matrix, such that each one learns a separate dichotomy of the classes. While ECOC is one of the best solutions to multi-class problems, it is suboptimal since the code matrix and the base classiffiers are not learned simultaneously. In this thesis, we present three different algorithms that iteratively updates the ECOC code matrix to improve the performance of the ensemble by reducing the decoupling. Firstly, we applied the previously developed FlipECOC+ update algorithm. Second method is applying simulated annealing method on updating ECOC matrix by flipping proposed entries according to ascending order. Last method is applying beam search to find updated ECOC matrix which has highest validation accuracy. We applied all three algorithms on UCI (University of California Irvine) data sets. Beam search algorithm gives the best result on UCI data sets. All of the proposed update algorithms does not involve further training of the classiffiers and can be applied to any ECOC ensemble.