Approximating optimally discrete probability distribution with kth-order dependency for combining multiple decisions
✍ Scribed by Hee-Joong Kang; Kawon Kim; Jin H. Kim
- Publisher
- Elsevier Science
- Year
- 1997
- Tongue
- English
- Weight
- 704 KB
- Volume
- 62
- Category
- Article
- ISSN
- 0020-0190
No coin nor oath required. For personal study only.
✦ Synopsis
A probabilistic combination of K classifiers' decisions obtained from samples needs a (K + &t-order probability distribution. as well as proposed an approximation scheme of such a high-order distribution with a product of only first-order tree dependencies. However, if a classifier follows mom than two classifiers, such first-order dependency does not estimate adequately a high-order distribution. Therefore, a new method is proposed to approximate optimally the (K + l)st-order distribution with a product set of M-order dependencies where 1 Q k Q K, which are identified by a systematic dependency-directed approach. And also, a new method is presented to combine probabilistitally multiple decisions with the product set of the M-order dependencies, using a Bayesian formalism.