Subsequently, we need to keep the d

Subsequently, we need to keep the dimension of the population <br>and halve the concatenated matrix to obtain. To do so, we <br>first introduce the accuracy and the fitness function of the <br>classification process as equation (25, and owing to the feature selection is considered as a multi-objective problem, it is necessary to minimize the number of features and maximize the classification accuracy of a given classifier at the same time. Therefore, equation (24) is used to evaluate the joint feature matrix H, expressed as follows:

Thenly, we need to keep the dimension of the population<br>and halve si'r concatenated matrix to obtain. To do so, we<br>first first first the terness and the fitness function of the the<br>the classification process as equation (25, and the owing to the feature selection is considered as a multi-objective problem, it is is is the dos sita sings of the number of the and maximize the the classification accuracy of a given classifier at the same time. Theso, equation (24) is used to evaluate the joint feature matrix H, expressed as follows: ...

随后，我们需要保持人口的规模<br>将连接矩阵减半得到。为此，我们<br>首先介绍<br>将分类过程作为方程（25），由于特征选择被认为是一个多目标问题，因此需要在最小化特征个数的同时，最大化给定分类器的分类精度。因此，方程（24）用于计算关节特征矩阵H，表示如下：<br>