Combining classifiers based on confidence values
- 1 January 1999
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
Abstract
The paper describes our investigation into the neural gas (NG) network algorithm and the hierarchical overlapped architecture (HONG) which we have built by retaining the essence of the original NG algorithm. By defining an implicit ranking scheme, the NG algorithm was made to run faster in its sequential implementation. Each HONG network generated multiple classifications for every sample data presented as confidence values. These confidence values were combined to obtain the final classification of the HONG architecture. Three HONG networks based on three different feature sets with global and structural features were also trained to obtain better classification on conflicting handwritten data. An excellent recognition rate for the NIST SD3 database was consequently obtained.Keywords
This publication has 8 references indexed in Scilit:
- On the importance of sorting in "neural gas" training of vector quantizersPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Hierarchical overlapped SOM's for pattern classificationIEEE Transactions on Neural Networks, 1999
- Fast sequential implementation of "neural-gas" network for vector quantizationIEEE Transactions on Communications, 1998
- Off-line, handwritten numeral recognition by perturbation methodPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1997
- Combining the results of several neural network classifiersNeural Networks, 1994
- 'Neural-gas' network for vector quantization and its application to time-series predictionIEEE Transactions on Neural Networks, 1993
- Computer recognition of unconstrained handwritten numeralsProceedings of the IEEE, 1992
- The self-organizing mapProceedings of the IEEE, 1990