Neither the MDAS total scores nor the classification into high vs. low fear were related to age or gender.
In total, we evaluated the classification of tissue samples based on different combinations of N genes and investigated the classifiers up to 10-feature model classifier.
For efficiency measuring, we consider the total classification time t as the sum of training set definition time t 1 and time consuming t 2 of the SVM classifier.
The total classification rate of the proposed method can be defined as follows [3]: (9).
The total classification accuracy of the five methods is 84%, 81 %, 79 and 81%and81% respectively.
While with the feature selection algorithm mRMR and IG, the total classification accuracy of WSVM achieves the best classification accuracy.
The total classification accuracy is in the range (93% to 95%) for 100 EEG segments.
The third term ∑ i = 1 m ∑ j = 1 c ξ i j denotes the total classification error for all the samples, which is minimized to achieve high classification accuracy.
In order to test whether the selected discriminated features could serve as a diagnostic biomarker panel, a logistic regression model was used to find a linear combination of the biomarkers that minimizes the total classification error.
(13) vector - minimiz e (x, y ) ∑ i = 1 n x i, ∑ i = 1 l 1 + l 2 y i, subject to (11 ) (12 ) with x i ∈ { 0, 1 }, i ∈ { 1, 2, ⋯, n }, y i ≥ 0, i ∈ { 1, 2, ⋯, l 1 + l 2 } The first term of objective function in Equation (13) is to minimize the number of chosen features, and the second one is to minimize the total classification error.
