Algebraic-logic synthesis of correct recognition procedures based on elementary algorithms
Computational Mathematics and Mathematical Physics
A decision-theoretic generalization of on-line learning and an application to boosting
Journal of Computer and System Sciences - Special issue: 26th annual ACM symposium on the theory of computing & STOC'94, May 23–25, 1994, and second annual Europe an conference on computational learning theory (EuroCOLT'95), March 13–15, 1995
A simple, fast, and effective rule learner
AAAI '99/IAAI '99 Proceedings of the sixteenth national conference on Artificial intelligence and the eleventh Innovative applications of artificial intelligence conference innovative applications of artificial intelligence
Improved Boosting Algorithms Using Confidence-rated Predictions
Machine Learning - The Eleventh Annual Conference on computational Learning Theory
Construction of an ensemble of logical correctors on the basis of elementary classifiers
Pattern Recognition and Image Analysis
| Hi-index | 0.00 |
The construction of correct recognition procedures based on incorrect elementary classifiers (fragments of learning object descriptions) is studied for problems with integer data. A number of special Boolean functions are considered as the correction functions. A family of correct sets of elementary classifiers (a family of correctors) is constructed at the learning stage. An original technique is proposed to construct this family. The technique helps find the most informative correctors without significant time expenditures. The results are given for the constructed recognition procedures tested on real problems. Earlier studies dealt with the simplest case when the elementary classifier was a separate feature value in the description of a learning object.