載入套件

library(e1071)
## Warning: package 'e1071' was built under R version 3.3.3
library(ggplot2)
## Warning: package 'ggplot2' was built under R version 3.3.3
library(caret)
## Warning: package 'caret' was built under R version 3.3.3
## Loading required package: lattice

整理資料

dfev <- read.csv("fev.csv")
dfev <- dfev[,c(2:4,6)]

想要看新生兒肺功能指數是否會受到年齡、周遭抽菸因子的影響 畫出年齡與肺功能的散布圖，並加上受到的不同抽菸狀態的影響

g<-ggplot(data=dfev,aes(x =FEV , y = Age)) 
g1 <- g + geom_point(aes(color = Smoker))
g1

利用線性回歸模型來看看關係

H0:肺功能不會受到年齡、抽菸與否的影響

m1 <- lm(FEV~Age+Smoker,data=dfev)
summary(m1)
## 
## Call:
## lm(formula = FEV ~ Age + Smoker, data = dfev)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -1.6653 -0.3564 -0.0508  0.3494  2.0894 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)    
## (Intercept) 0.158378   0.131003   1.209  0.22712    
## Age         0.230605   0.008184  28.176  < 2e-16 ***
## SmokerNon   0.208995   0.080745   2.588  0.00986 ** 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.5651 on 651 degrees of freedom
## Multiple R-squared:  0.5766, Adjusted R-squared:  0.5753 
## F-statistic: 443.3 on 2 and 651 DF,  p-value: < 2.2e-16

發現年紀與是否抽菸的p-value均<0.05，固可推測二者確實會對新生兒的肺功能有影響

進一步看看抽菸與年齡是否會產生交互作用，對新生兒的肺功能有影響

m2 <- lm(FEV~Age*Smoker,data=dfev)
summary(m2)
## 
## Call:
## lm(formula = FEV ~ Age * Smoker, data = dfev)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -1.76645 -0.34947 -0.03364  0.33679  2.05990 
## 
## Coefficients:
##               Estimate Std. Error t value Pr(>|t|)    
## (Intercept)    2.19697    0.40596   5.412 8.78e-08 ***
## Age            0.07986    0.02959   2.699  0.00713 ** 
## SmokerNon     -1.94357    0.41428  -4.691 3.31e-06 ***
## Age:SmokerNon  0.16270    0.03074   5.293 1.65e-07 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.5537 on 650 degrees of freedom
## Multiple R-squared:  0.5941, Adjusted R-squared:  0.5922 
## F-statistic: 317.1 on 3 and 650 DF,  p-value: < 2.2e-16

發現年齡與抽菸的交互作用項，其p-value<0.05，固可推測二者會產生交互作用並且對新生兒的肺功能有影響

SVM來預測新生兒周邊環境抽菸情形

#建模型
x <- subset(dfev,select = -Smoker)
y <- dfev$Smoker
model <- svm(Smoker~.,data = dfev,cost= 10,kernal = "radial", scale = FALSE)
print(model)
## 
## Call:
## svm(formula = Smoker ~ ., data = dfev, cost = 10, kernal = "radial", 
##     scale = FALSE)
## 
## 
## Parameters:
##    SVM-Type:  C-classification 
##  SVM-Kernel:  radial 
##        cost:  10 
##       gamma:  0.3333333 
## 
## Number of Support Vectors:  197
#預測:x與y是在對正確答案
presult <- predict(model,x)

confusion matrix

confusionMatrix(presult,y)
## Confusion Matrix and Statistics
## 
##           Reference
## Prediction Current Non
##    Current      31   5
##    Non          34 584
##                                           
##                Accuracy : 0.9404          
##                  95% CI : (0.9194, 0.9573)
##     No Information Rate : 0.9006          
##     P-Value [Acc > NIR] : 0.0001926       
##                                           
##                   Kappa : 0.5844          
##  Mcnemar's Test P-Value : 7.34e-06        
##                                           
##             Sensitivity : 0.47692         
##             Specificity : 0.99151         
##          Pos Pred Value : 0.86111         
##          Neg Pred Value : 0.94498         
##              Prevalence : 0.09939         
##          Detection Rate : 0.04740         
##    Detection Prevalence : 0.05505         
##       Balanced Accuracy : 0.73422         
##                                           
##        'Positive' Class : Current         
##

結論:由於此筆資料，雖然未抽菸者在整筆資料裡面明顯占有多數，但此模型預測的準確率有高達94%，精準的預測出抽菸與未抽菸的人