2021/8/3-第五天作业-CSDN博客

本文链接：https://blog.csdn.net/qq_42758052/article/details/119351459

1.根据决策表定义，写出下图中的 $\mathbf{U}, \mathbf{C}, \mathbf{D} , \mathbf{V}$ 和 $I$ 注: 最后两个属性为决策属性.

Patient	Headache	Temperature	Lymphocyte	Leukocyte	Eosinophil	Heartbeat	Flu
$x_1$	Yes	High	High	High	High	Normal	Yes
$x_2$	Yes	High	Normal	High	High	Abnormal	Yes
$x_3$	Yes	High	High	High	Normal	Abnorma	Yes
$x_4$	No	High	Normal	Normal	Normal	Normal	No
$x_5$	Yes	Normal	Normal	Low	High	Abnorma	No
$x_6$	Yes	Normal	Low	High	Normal	Abnorma	No
$x_7$	Yes	Low	Low	High	Normal	Normal	Yes

答：实例集合 $\mathbf{U} = \{x_1, x_2, x_3, x_4, x_5, x_6, x_7\}$ ，条件属性集合 $\mathbf{C} = \{\textrm{Headache, Temperature, Lymphocyte , Leukocyte, Eosinophil}\}$ ，决策属性集合 $\mathbf{D} = \{\rm Heartbeat, Flu\}$ ，值域 $\mathbf{V} = \{\rm No,Yes, High, Normal, Low, Abnormal\}$ ，函数 $I$ 可以由上图的二维表进行表示。

2. 定义一个标签分布系统, 即各标签的值不是 0/1, 而是 [0,1] 区间的实数, 且同一对象的标签和为 1.

答： A label distribution system is a tuple $(\mathbf{X, Y})$ where $\mathbf{X} = [x_{ij}]_{n \times m} \in \mathbb{R}^{n \times m}$ is a data matrix, $\mathbf{Y} = [y_{ik}]_{n \times l} \in [0, 1]^{n \times l}$ is the label matrix satisfying $\forall y_{ij} \in \mathbf{Y}$ , $\sum_{k=1}^l y_{ik} = 1$ , where $n$ is the number of instances, $m$ is the number of features, and $l$ is the number of labels.