Hybridization of Evolutionary Mechanisms for Feature Subset Selection in Unsupervised Learning
MICAI '09 Proceedings of the 8th Mexican International Conference on Artificial Intelligence
BR: A New Method for Computing All Typical Testors
CIARP '09 Proceedings of the 14th Iberoamerican Conference on Pattern Recognition: Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications
A fast implementation of the CT_EXT algorithm for the testor property identification
MICAI'10 Proceedings of the 9th Mexican international conference on Artificial intelligence conference on Advances in soft computing: Part II
Identification of risk factors for TRALI using a hybrid algorithm
MICAI'12 Proceedings of the 11th Mexican international conference on Advances in Artificial Intelligence - Volume Part I
Evolutionary computation in the identification of risk factors. Case of TRALI
Expert Systems with Applications: An International Journal
| Hi-index | 0.00 |
Typical testors are a useful tool for feature selection and for determining feature relevance in supervised classification problems, especially when quantitative and qualitative features are mixed. Nowadays, computing all typical testors is a highly costly procedure; all described algorithms have exponential complexity. Existing algorithms are not acceptable methods owing to several problems (particularly run time) which are dependent on matrix size. Because of this, different approaches, such as sequential algorithms, parallel processing, genetic algorithms, heuristics and others have been developed. This paper describes a novel external type algorithm that improves the run time of all other reported algorithms. We analyze the behaviour of the algorithm in some experiments, whose results are presented here.