Improving Committee Diagnosis with Resampling Techniques
Bambang Parmanto, Paul Munro, Howard R. Doyle · 1995
Central to the performance improvement of a committee relative to individual networks is the error correlation between networks in the committee. We investigated methods of achieving error indepen-dence between the networks by training the networks with different resampling sets from the original training set. The methods were tested on the sinwave artificial task and the real-world problems of hepatoma (liver cancer) and breast cancer diagnoses. 1