欧几里得距离(欧氏距离):
d
(
x
,
y
)
=
[
∑
i
=
1
d
(
x
i
−
y
i
)
2
]
1
/
2
d(x, y)=\left[\sum_{i=1}^{d}\left(x_{i}-y_{i}\right)^{2}\right]^{1 / 2}
d(x,y)=[i=1∑d(xi−yi)2]1/2也可表示为:
d
(
x
,
y
)
=
∥
x
−
y
∥
2
=
(
x
−
y
)
T
(
x
−
y
)
d(x, y)=\|x-y\|_{2}=\sqrt{(x-y)^{T}(x-y)}
d(x,y)=∥x−y∥2=(x−y)T(x−y)
曼哈顿距离(街市距离):
d
(
x
,
y
)
=
∑
i
=
1
d
∣
x
i
−
y
i
∣
d(x, y)=\sum_{i=1}^{d}\left|x_{i}-y_{i}\right|
d(x,y)=i=1∑d∣xi−yi∣
也可表示为:
d
(
x
,
y
)
=
∥
x
−
y
∥
2
d(x, y)=\|x-y\|_{2}
d(x,y)=∥x−y∥2
各个方向上距离和。
切比雪夫距离
d
(
x
,
y
)
=
max
(
x
i
−
y
i
)
d(x, y)=\max \left(x_{i}-y_{i}\right)
d(x,y)=max(xi−yi)
也可表示为:
d
(
x
,
y
)
=
∥
x
−
y
∥
∝
d(x, y)=\|x-y\|_{\propto}
d(x,y)=∥x−y∥∝
明可夫斯基距离
dist
(
X
,
Y
)
=
(
∑
i
=
1
n
∣
x
i
−
y
i
∣
p
)
1
/
p
\operatorname{dist}(X, Y)=\left(\sum_{i=1}^{n}\left|x_{i}-y_{i}\right|^{p}\right)^{1 / p}
dist(X,Y)=(i=1∑n∣xi−yi∣p)1/p
余弦相似度:
cos
(
θ
)
=
∑
i
=
1
n
(
x
i
×
y
i
)
∑
i
=
1
n
(
x
i
)
2
×
∑
i
=
1
n
(
y
i
)
2
\cos (\theta)=\frac{\sum_{i=1}^{n}\left(x_{i} \times y_{i}\right)}{\sqrt{\sum_{i=1}^{n}\left(x_{i}\right)^{2}} \times \sqrt{\sum_{i=1}^{n}\left(y_{i}\right)^{2}}}
cos(θ)=∑i=1n(xi)2×∑i=1n(yi)2∑i=1n(xi×yi)
c
o
s
(
θ
)
cos(\theta)
cos(θ)越趋于1,数据越相似。
皮尔森相关系数
r
(
X
,
Y
)
=
n
∑
x
y
−
∑
x
∑
y
n
∑
x
2
−
(
∑
x
)
2
⋅
n
∑
y
2
−
(
∑
y
)
2
r(X, Y)=\frac{n \sum x y-\sum x \sum y}{\sqrt{n \sum x^{2}-\left(\sum x\right)^{2}} \cdot \sqrt{n \sum y^{2}-\left(\sum y\right)^{2}}}
r(X,Y)=n∑x2−(∑x)2⋅n∑y2−(∑y)2n∑xy−∑x∑y