Machine Learning-Based Aging Gene Feature Selection and Classification

算法1 层次特征选择HFS
输入: 数据集D、特征集F、测试样本t、层次结构G
输出: 特征子集S
1. P ← Ø, S ← Ø;
2. F′ ← 返回F中在样本t中取1的特征子集;
3. for eachleaf in G中所有叶结点 do
4.　 p ←根据G构造从叶节点leaf到根的路径;
5. 　将路径p添加到集合P;
6. for eachp in Pdo
7. 　q ← F′ ∩ {f \|f∈p};
8. 　m ←q中特征的互信息平均值;
9. 　for eachf_i in ρdo
10. 　　ifM（C, f_i） > mthen
11. 　　　将f_i添加到集合S; 12. returnS;

Table 1. [in Chinese]
View table
View in Article
Table 1. [in Chinese]
数据集样本数特征数依赖数正例率/%
小鼠 130 2885 1773 65.4
果蝇 198 1596 692 63.1
蠕虫 861 2509 1459 32.9
酵母 400 2759 1772 12.8

Table 2. [in Chinese]
View table
View in Article
Table 2. [in Chinese]
数据集 KNN NB SVM LR DT RF
小鼠 63.6 65.9 64.1 64.6 61.2 64.7
果蝇 60.1 60.8 59.5 60.3 55.3 59.9
蠕虫 52.3 56.4 60.3 57.8 58.3 60.2
酵母 43.9 50.8 43.3 41.1 47.7 48.1

Table 3. [in Chinese]
View table
View in Article
Table 3. [in Chinese]
类别 NB SP SM CFS MR RPV HFS
小鼠 65.9 74.1 71.6 71.5 74.8 66.7 75.5
果蝇 60.8 64.3 59.8 64.0 62.1 61.7 64.5
蠕虫 56.4 61.1 58.4 59.2 58.2 61.8 62.9
酵母 50.8 65.0 65.0 59.7 56.2 59.2 67.7

Tools

Get Citation

Copy Citation Text

Jie ZENG, Quanwang WU, Dehui LI, Junmin GAO. Machine Learning-Based Aging Gene Feature Selection and Classification[J]. Experiment Science and Technology, 2025, 23(2): 84

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites