Self-organization and associative memory: 3rd edition
Self-organization and associative memory: 3rd edition
Meta Analysis of Classification Algorithms for Pattern Recognition
IEEE Transactions on Pattern Analysis and Machine Intelligence
The Handbook of Brain Theory and Neural Networks
The Handbook of Brain Theory and Neural Networks
Measuring Classification Complexity of Image Databases: A Novel Approach
ICIAP '99 Proceedings of the 10th International Conference on Image Analysis and Processing
Using boundary methods for estimating class separability
Using boundary methods for estimating class separability
Multiresolution Estimates of Classification Complexity
IEEE Transactions on Pattern Analysis and Machine Intelligence
| Hi-index | 0.00 |
We present in this paper an auto-adaptive neural network tree structure that is used to solve classification problems. Two main ideas are behind this neural network tree structure. The first one is that it is easier to solve simple problems than difficult ones, so we divide an initial hard classification problem into several simplerleasier ones in a recursive way: "Divide To Simplify". In order to perform the classification problem decomposition, we need to answer two following questions: is the problem difficult? And then, how to simplify it? The second idea is to use some classification complexity estimation methods to evaluate the problem difficulty. We propose an algorithm which estimates the problem complexity, divides it into several easier ones and builds a self organized neural tree structure that solves this problem efficiently.