1.距离函数总结
- 闵可夫斯基距离：给定样本
x
→
i
=
(
x
i
,
1
,
x
i
,
2
,
⋯
 
,
x
i
,
n
)
T
\overrightarrow{\mathbf{x}}_{i}=\left(x_{i, 1}, x_{i, 2}, \cdots, x_{i, n}\right)^{T}
xi=(xi,1,xi,2,⋯,xi,n)T,
x
→
j
=
(
x
j
,
1
,
x
j
,
2
,
⋯
 
,
x
j
,
n
)
T
\overrightarrow{\mathbf{x}}_{j}=\left(x_{j, 1}, x_{j, 2}, \cdots, x_{j, n}\right)^{T}
xj=(xj,1,xj,2,⋯,xj,n)T，则闵可夫斯基距离定义为：
distance( ( x → i , x → j ) = ( ∑ d = 1 n ∣ x i , d − x j , d ∣ p ) 1 / p \text { distance( }\left(\overrightarrow{\mathbf{x}}_{i}, \overrightarrow{\mathbf{x}}_{j}\right)=\left(\sum_{d=1}^{n}\left|x_{i, d}-x_{j, d}\right|^{p}\right)^{1 / p} distance( (xi,xj)=(d=1∑n∣xi,d−xj,d∣p)1/p - 当p=2时，闵可夫斯基距离就是欧式距离：
distance ( x → i , x → j ) = ∥ x → i − x → j ∥ 2 = ∑ d = 1 n ∣ x i , d − x j , d ∣ 2 \operatorname{distance}\left(\overrightarrow{\mathbf{x}}_{i}, \overrightarrow{\mathbf{x}}_{j}\right)=\left\|\overrightarrow{\mathbf{x}}_{i}-\overrightarrow{\mathbf{x}}_{j}\right\|_{2}=\sqrt{\sum_{d=1}^{n}\left|x_{i, d}-x_{j, d}\right|^{2}} distance(xi,xj)=∥∥∥xi−xj∥∥∥2=d=1∑n∣xi,d−xj,d∣2 - 当p=1时，闵可夫斯基距离就是曼哈顿距离：
distance ( x → i , x → j ) = ∥ x → i − x → j ∥ 1 = ∑ d = 1 n ∣ x i , d − x j , d ∣ \text { distance }\left(\overrightarrow{\mathbf{x}}_{i}, \overrightarrow{\mathbf{x}}_{j}\right)=\left\|\overrightarrow{\mathbf{x}}_{i}-\overrightarrow{\mathbf{x}}_{j}\right\|_{1}=\sum_{d=1}^{n}\left|x_{i, d}-x_{j, d}\right| distance (xi,xj)=∥∥∥xi−xj∥∥∥1=d=1∑n∣xi,d−xj,d∣
1万+
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
