Combining complementary neural network and error-correcting output codes for multiclass classification problems

Abstract

This paper presented an innovative method, combining Complementary Neural Networks (CMTNN) and Error-Correcting Output Codes (ECOC), to solve multiclass classification problem. CMTNN consist of truth neural network and falsity neural network created based on truth and falsity information, respectively. In the experiment, we deal with feed-forward backpropagation neural networks, trained using 10 fold cross-validation method and classified based on minimum distance. The proposed approach has been tested with three benchmark problems: balance, vehicle and nursery from the UCI machine learning repository. We found that our approach provides better performance compared to the existing techniques considering on either CMTNN or ECOC.