参考资料：《数据探掘 R语言实战》 p65-P68

install.packages("rattle") # 获取实验数据集
install.packages("ellipse") # 获取构建相关图的函数plotcorr

rm(list = ls())
library("ellipse") # 加载包
library("rattle") 

data(weather) # 加载数据集
head(weather) # 查看数据集

##         Date Location MinTemp MaxTemp Rainfall Evaporation Sunshine
## 1 2007-11-01 Canberra     8.0    24.3      0.0         3.4      6.3
## 2 2007-11-02 Canberra    14.0    26.9      3.6         4.4      9.7
## 3 2007-11-03 Canberra    13.7    23.4      3.6         5.8      3.3
## 4 2007-11-04 Canberra    13.3    15.5     39.8         7.2      9.1
## 5 2007-11-05 Canberra     7.6    16.1      2.8         5.6     10.6
## 6 2007-11-06 Canberra     6.2    16.9      0.0         5.8      8.2
##   WindGustDir WindGustSpeed WindDir9am WindDir3pm WindSpeed9am
## 1          NW            30         SW         NW            6
## 2         ENE            39          E          W            4
## 3          NW            85          N        NNE            6
## 4          NW            54        WNW          W           30
## 5         SSE            50        SSE        ESE           20
## 6          SE            44         SE          E           20
##   WindSpeed3pm Humidity9am Humidity3pm Pressure9am Pressure3pm Cloud9am
## 1           20          68          29      1019.7      1015.0        7
## 2           17          80          36      1012.4      1008.4        5
## 3            6          82          69      1009.5      1007.2        8
## 4           24          62          56      1005.5      1007.0        2
## 5           28          68          49      1018.3      1018.5        7
## 6           24          70          57      1023.8      1021.7        7
##   Cloud3pm Temp9am Temp3pm RainToday RISK_MM RainTomorrow
## 1        7    14.4    23.6        No     3.6          Yes
## 2        3    17.5    25.7       Yes     3.6          Yes
## 3        7    15.4    20.2       Yes    39.8          Yes
## 4        7    13.5    14.1       Yes     2.8          Yes
## 5        7    11.1    15.4       Yes     0.0           No
## 6        5    10.9    14.8        No     0.2           No

test_data <- weather[, 12:21] # 第12到21列为数值型
cor_matrix <- cor(test_data, use = "pairwise") # 两两变量求相关系数
cor_matrix # 显示结果

##              WindSpeed9am WindSpeed3pm Humidity9am Humidity3pm Pressure9am
## WindSpeed9am   1.00000000   0.47296617  -0.2706229  0.14665712 -0.35633183
## WindSpeed3pm   0.47296617   1.00000000  -0.2660925 -0.02636775 -0.35980011
## Humidity9am   -0.27062286  -0.26609247   1.0000000  0.54671844  0.13572697
## Humidity3pm    0.14665712  -0.02636775   0.5467184  1.00000000 -0.08794614
## Pressure9am   -0.35633183  -0.35980011   0.1357270 -0.08794614  1.00000000
## Pressure3pm   -0.24795238  -0.33732535   0.1344205 -0.01005189  0.96789496
## Cloud9am       0.10184246  -0.02642642   0.3928416  0.55163264 -0.15755279
## Cloud3pm      -0.02247149   0.00720724   0.2719381  0.51010790 -0.14100043
## Temp9am        0.06407405  -0.01776636  -0.4365506 -0.25568147 -0.46041819
## Temp3pm       -0.23518635  -0.18756965  -0.3551186 -0.58167615 -0.25367375
##              Pressure3pm    Cloud9am    Cloud3pm     Temp9am    Temp3pm
## WindSpeed9am -0.24795238  0.10184246 -0.02247149  0.06407405 -0.2351864
## WindSpeed3pm -0.33732535 -0.02642642  0.00720724 -0.01776636 -0.1875697
## Humidity9am   0.13442050  0.39284158  0.27193809 -0.43655057 -0.3551186
## Humidity3pm  -0.01005189  0.55163264  0.51010790 -0.25568147 -0.5816761
## Pressure9am   0.96789496 -0.15755279 -0.14100043 -0.46041819 -0.2536738
## Pressure3pm   1.00000000 -0.12894408 -0.14383718 -0.49263629 -0.3454853
## Cloud9am     -0.12894408  1.00000000  0.52521793  0.02104135 -0.2023440
## Cloud3pm     -0.14383718  0.52521793  1.00000000  0.04094519 -0.1728142
## Temp9am      -0.49263629  0.02104135  0.04094519  1.00000000  0.8444058
## Temp3pm      -0.34548531 -0.20234405 -0.17281423  0.84440581  1.0000000

col <- 1:10 # 填充颜色
plotcorr(cor_matrix, col = col, type = "lower", diag = F)

相关性越强圆形越窄，左倾斜（\）表示负相关，右倾斜（/）表示正相关，例如Temp3pm和Temp9am为正相关

# numbers = T, diag = T
plotcorr(cor_matrix, numbers = T, type = "lower", diag = T)

• col 设置椭圆填充颜色

• type 设置显示上三角、下三角、全部显示（upper、lower、full）

• diag 逻辑值，是否显示主对角线

• numbers 逻辑值，是否用相关系数值取代椭圆，数值会增大10倍四舍五入