Fusing neural networks through space partitioning and fuzzy integration

Neural Processing Letters

Volumn 16, Issue 1, 2002, Pages 53-65

  Verikas, Antanas ^a   Lipnickas, Arunas ^b  

^a Halmstad University   (Lithuania)

^b KAUNAS UNIVERSITY OF TECHNOLOGY   (Lithuania)

Author keywords

Decision fusion; Fuzzy integral; Half Half bagging; Neural network

Indexed keywords

DECISION THEORY; ERROR ANALYSIS; FUZZY SETS; INTEGRATION; PROBLEM SOLVING;

FUZZY INTEGRATION;

NEURAL NETWORKS;

EID: 0036671543     PISSN: 13704621     EISSN: None     Source Type: Journal    
DOI: 10.1023/A:1019703911322     Document Type: Article

References (36)

1. Democracy in neural nets: Voting schemes for classification
2. An asymptotic analysis of some expert fusion methods
3. Improved dynamic programming-based handwritten word recognition using optimal order statistics
4. Experiments with a new boosting algorithm
5. A decision-theoretic generalization of on-line learning and an application to boosting
6. Fusion of handwritten word classifiers
7. Classification by fuzzy integral: Performance and tests
8. Optimal linear combinations of neural networks
9. Decision combination in multiple classifier systems
10. A method of combining multiple classifiers-A neural network approach
11. Convergence results of the EM approach to mixtures of experts architectures
12. Advances in fuzzy integration for pattern-recognition
13. On combining classifiers
14. Decision templates for multiple classifter fusion
15. Bayesian interpolation
16. Generating accurate and diverse members of a neural-network ensemble
17. When networks disagree: Ensamble method for neural networks
18. Learning with ensembles: How over-fitting can be useful
19. Fuzzy measures and fuzzy integrals: A survey
20. Averaging regularized estimators
21. Combining estimators using non-constant weighting functions
22. Fuzzy kohonen clustering networks
23. Linear and order statistics combiners for pattern classification
24. Adaptive character recognition system
25. Soft combination of neural classifiers: A comparative study
26. Training neural networks by stochastic optimisation
27. Fusing neural networks through fuzzy integration
28. Ensemble methods for phoneme classification
29. Combination of multiple classifiers using local accuracy estimates
30. Methods for combining multiple classifiers and their appli-cations to handwriting recognition