调优线性 SVM

R 中的支持向量机

Kailash Awati

Instructor

线性 SVM，默认 cost

library(e1071)
svm_model <- svm(y ~ ., 
                data = trainset, 
                type = "C-classification", 
                kernel = "linear", 
                scale = FALSE)
# Print model summary
svm_model

Call:
svm(formula = y ~ .,
    data = trainset,
    type = "C-classification", 
    kernel = "linear",
    scale = FALSE)

Parameters:
SVM-Type:  C-classification 
 SVM-Kernel:  linear 
       cost:  1 
      gamma:  0.5 
Number of Support Vectors:  55

第 2.3 章 - 线性可分数据集，默认 cost 线性核，含支持向量、决策边界与间隔

线性 SVM（cost = 100）

library(e1071)
svm_model <- svm(y ~ ., 
                data = trainset, 
                type = "C-classification", 
                kernel = "linear", 
                cost = 100,
                scale = FALSE)
# Print model summary
svm_model

Call:
svm(formula = y ~ .,
    data = trainset,
    type = "C-classification", 
    kernel = "linear",
    cost = 100,
    scale = FALSE)

Parameters:
SVM-Type:  C-classification 
 SVM-Kernel:  linear 
       cost:  100 
      gamma:  0.5 
Number of Support Vectors:  6

第 2.3 章 - 线性可分数据，cost = 100 线性核，含支持向量、决策边界与间隔

含义

若已知决策边界为线性，可适当减小间隔
但现实中很少如此

第 2.3 章 - 非线性可分数据集

非线性数据集，线性 SVM（cost = 100）

使用 80% 训练集，构建 cost=100 的模型

# Build model
library(e1071)
svm_model<- svm(y ~ ., 
                data = trainset, 
                type = "C-classification", 
                kernel = "linear", 
                cost = 100,
                scale = FALSE)

计算准确率

# Train and test accuracy
pred_train <- predict(svm_model, trainset)
mean(pred_train == trainset$y)

0.8208333

pred_test <- predict(svm_model, testset)
mean(pred_test == testset$y)

0.85

在 50 次随机划分上，测试准确率均值：82.9%

第 2.3 章 - 非线性可分数据集，cost=100 线性核，含支持向量、决策边界与间隔

非线性数据集，线性 SVM（cost = 1）

重新构建模型，设置 cost = 1

# Trainset contains 80% of data
# Same train/test split as before.
# Build model
svm_model <- svm(y ~ ., 
                data = trainset, 
                type = "C-classification", 
                kernel = "linear", 
                cost = 1,
                scale = FALSE)

计算测试集准确率

# Test accuracy
pred_test <- predict(svm_model, testset)
mean(pred_test == testset$y)

0.8666667

在 50 次随机划分上，测试准确率均值：83.7%

第 2.3 章 - 非线性可分数据集，cost=1 线性核，含支持向量、决策边界与间隔

Passons à la pratique !

R 中的支持向量机