处理分类特征

R 异常检测入门

Alastair Rushworth

Data Scientist

检查列的类别

单列的类别

class(sat$V1)

"numeric"

所有列的类别

sapply(X = sat, FUN = class)

      label          V1          V2          V3         V4          V5          V6    high_low
  "numeric"   "numeric"   "numeric"   "numeric"  "numeric"   "numeric"   "numeric" "character"

Isolation Forest

将分类特征编码为 factor

sat$high_low <- as.factor(sat$high_low)

class(sat$high_low)

"factor"

训练 Isolation Forest

sat_for <- iForest(sat[, -1], nt = 100)

含 factor 的 LOF

Gower 距离 度量含分类和数值特征的点间距离

library(cluster)
sat_dist <- daisy(sat[, -1], metric = "gower")

将 sat_dist 传给 lof

sat_lof <- lof(sat_dist, k = 10)

探索 Gower 距离矩阵

将对象转换为矩阵

sat_distmat <- as.matrix(sat_dist)

查找点间距离的最大最小值

range(sat_distmat)

0.0000000 0.8680774

让我们来练习！

R 异常检测入门