Download now Free registration required
Feature selection is a critical step in building a large scale identification system based on Electro-CardioGram (ECG) signal. During this phase, the set of features that deemed to be the most effective attributes are extracted in order to construct suitable identification algorithms. A key problem that many researchers face is how to choose the optimal set of features since not all features are relevant to the identification algorithm, and in some cases, irrelevant and redundant features can introduce noisy data that distracts the learning algorithm and therefore severely degrade the accuracy of the identification system and cause slow training and testing process.
- Format: PDF
- Size: 498.6 KB