In automated pattern recognition systems, the two main challenges are feature selection and extraction. The features selected directly affects the number of measurements required; and extracting low-dimensional features from the selected ones reduces the computational complexity of the classifier In traditional approaches, human expertise is obligatory for feature selection and statistical techniques are employed for feature projection. In this paper, a constrained genetic algorithm for performing these two tasks simultaneously, in conjunction with the k-nearest neighbor classifier is proposed. This algorithm requires minimal human intervention as it realizes good tradeoff solutions between classification. accuracy, feature measurement requirements, and computational complexity.
|Title of host publication||International Conference on Computational Intelligence and Security CIS’2007|
|Publication status||Published - 2007|
|Event||International Conference on Computational Intelligence and Security CIS’2007 - |
Duration: 1 Jan 2007 → …
|Conference||International Conference on Computational Intelligence and Security CIS’2007|
|Period||01/01/2007 → …|