主成分分析（PCA）

使用 R 的 caret 進行 Machine Learning

Zach Mayer

Data Scientist at DataRobot and co-author of caret

主成分分析

合併低變異且彼此相關的變數
產生一組高變異、互相垂直的預測變數
避免共線性（也就是預測變數彼此相關）

PCA：視覺化示意

第 1 個成分有最高變異
第 2 個成分有次高變異
以此類推…

貼上的影像-919.png

範例：血腦資料

許多預測變數
其中多數變異很低

# Load the blood brain dataset
data(BloodBrain)
names(bbbDescr)[nearZeroVar(bbbDescr)]

[1] "negative"     "peoe_vsa.2.1" "peoe_vsa.3.1"
[4] "a_acid"       "vsa_acid"     "frac.anion7."
[7] "alert"

範例：血腦資料

# Basic model
set.seed(42)
data(BloodBrain)
model <- train(
  bbbDescr, 
  logBBB, 
  method = "glm",
  trControl = trainControl(
    method = "cv", number = 10, verbose = TRUE
  ),
  preProcess = c("zv", "center", "scale")
)
min(model$results$RMSE)

1.107702

範例：血腦資料

# Remove low-variance predictors
set.seed(42)
data(BloodBrain)
model <- train(
  bbbDescr, 
  logBBB, 
  method = "glm",
  trControl = trainControl(
    method = "cv", number = 10, verbose = TRUE
  ),
  preProcess = c("nzv", "center", "scale")
)
min(model$results$RMSE)

0.9796199

範例：血腦資料

# Add PCA
set.seed(42)
data(BloodBrain)
model <- train(
  bbbDescr, 
  logBBB, 
  method = "glm",
  trControl = trainControl(
    method = "cv", number = 10, verbose = TRUE
  ),
  preProcess = c("zv", "center", "scale", "pca")
)
min(model$results$RMSE)

0.9796199

一起來練習吧！

使用 R 的 caret 進行 Machine Learning