Rule Extraction and Reduction for Hyper Surface Classification
ISNN 2009 Proceedings of the 6th International Symposium on Neural Networks: Advances in Neural Networks - Part II
Combination methodologies of multi-agent hyper surface classifiers: design and implementation issues
AIS-ADM'07 Proceedings of the 2nd international conference on Autonomous intelligent systems: agents and data mining
Detecting unusual pattern with labeled data in two-stage
FSKD'09 Proceedings of the 6th international conference on Fuzzy systems and knowledge discovery - Volume 1
Local bayesian based rejection method for HSC ensemble
ISNN'10 Proceedings of the 7th international conference on Advances in Neural Networks - Volume Part I
Hi-index | 0.00 |
Based on Jordan Curve Theorem, a universal classification method called hyper surface classification (HSC) has recently been proposed. Experimental results are exciting, which show that in three-dimensional space, this method works fairly well in both accuracy and efficiency even for large size data up to 107. However, designing a number of new classifiers is needed with the growing of feature dimension. To solve the problem, a kind of efficient dimension transposition method that is suitable for HSC and without losing any essential information is put forward in this paper. The dimension transposition method rearrange all of the numerals in the higher dimensional data to lower dimensional data without changing each numeral, but only change their position according to some orders. The experiment shows that the method can classify high dimension data with high accuracy.