Search

To eliminate the signal trend terms, wavelet base functions are introduced for signal decomposition and reconstruction. The selection of wavelet bases may affect the result of eliminating signal trend. The concept of error index of signal trend elimination was proposed and the formula for calculating the index was presented. Using this formula, the error indexes of signal trend elimination of 34 types of wavelet bases were calculated. It was found that the wavelet bases of sym10 and other 5 types could be selected as preferred ones with less error indexes of signal trend elimination. Further, the sym10 and other two non-preferred wavelet bases were used to carry out the signal trend elimination of measured car body acceleration signal. The result shows that wavelet base of sym10 extracts the signal trend more precisely than other two non-preferred wavelet bases and the effectiveness of the presented formula for calculating the error index of signal trend elimination is validated [ABSTRACT FROM AUTHOR]

A new feature selection method based on clustering algorithm is proposed to selected informative features. First, category-unrelated features are kicked out according to Bhattacharyya distance. Then, based on the process of recursive feature elimination, a weighted distance between sample and the cluster center generated by the fuzzy interactive self-organizing data algorithm (ISODATA) is used as the index of feature for separating different classes. Finally, the candidate feature subset with the maximum area under the receiver operating characteristic curve (AUC) value and accuracy rate both in classification and clustering tests is selected as the optimal feature subset. The proposed feature subset selection method is applied to five gene expression profile datasets and experiment results show that the selected features have good performance in terms of both classification and clustering measurements. Results demonstrate that the proposed method is effective for selecting informative features from high-dimensional dataset. [ABSTRACT FROM AUTHOR]