Imbalance 01

How to handle Imbalanced Data?

ROSE

Data

##install.packages("ROSE")
library(ROSE)
data(hacide)
str(hacide.train)

## 'data.frame':    1000 obs. of  3 variables:
##  $ cls: Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
##  $ x1 : num  0.2008 0.0166 0.2287 0.1264 0.6008 ...
##  $ x2 : num  0.678 1.5766 -0.5595 -0.0938 -0.2984 ...

table(hacide.train$cls)

## 
##   0   1 
## 980  20

library(rpart)
treeimb <- rpart(cls ~ ., data = hacide.train)
pred.treeimb <- predict(treeimb, newdata = hacide.test)
accuracy.meas(hacide.test$cls, pred.treeimb[,2])

## 
## Call: 
## accuracy.meas(response = hacide.test$cls, predicted = pred.treeimb[, 
##     2])
## 
## Examples are labelled as positive when predicted is greater than 0.5 
## 
## precision: 1.000
## recall: 0.200
## F: 0.167

roc.curve(hacide.test$cls, pred.treeimb[,2], plotit = F)

## Area under the curve (AUC): 0.600

data_balanced_over <- ovun.sample(cls ~ ., data = hacide.train, method = "over",N = 1960)$data
table(data_balanced_over$cls)

## 
##   0   1 
## 980 980

data_balanced_under <- ovun.sample(cls ~ ., data = hacide.train, method = "under", N = 40, seed = 1)$data
table(data_balanced_under$cls)

## 
##  0  1 
## 20 20

data_balanced_both <- ovun.sample(cls ~ ., data = hacide.train, method = "both", p=0.5, N=1000, seed = 1)$data
table(data_balanced_both$cls)

## 
##   0   1 
## 520 480

data.rose <- ROSE(cls ~ ., data = hacide.train, seed = 1)$data
table(data.rose$cls)

## 
##   0   1 
## 520 480

#build decision tree models
tree.rose <- rpart(cls ~ ., data = data.rose)
tree.over <- rpart(cls ~ ., data = data_balanced_over)
tree.under <- rpart(cls ~ ., data = data_balanced_under)
tree.both <- rpart(cls ~ ., data = data_balanced_both)
#make predictions on unseen data
pred.tree.rose <- predict(tree.rose, newdata = hacide.test)
pred.tree.over <- predict(tree.over, newdata = hacide.test)
pred.tree.under <- predict(tree.under, newdata = hacide.test)
pred.tree.both <- predict(tree.both, newdata = hacide.test)
ROSE.holdout <- ROSE.eval(cls ~ ., data = hacide.train, learner = rpart, 
                          method.assess = "holdout", extr.pred = function(obj)obj[,2], seed = 1)
ROSE.holdout

## 
## Call: 
## ROSE.eval(formula = cls ~ ., data = hacide.train, learner = rpart, 
##     extr.pred = function(obj) obj[, 2], method.assess = "holdout", 
##     seed = 1)
## 
## Holdout estimate of auc: 0.980

Figure

roc.curve(hacide.test$cls, pred.tree.rose[,2])

## Area under the curve (AUC): 0.993

roc.curve(hacide.test$cls, pred.tree.over[,2])

## Area under the curve (AUC): 0.798

roc.curve(hacide.test$cls, pred.tree.under[,2])

## Area under the curve (AUC): 0.924

roc.curve(hacide.test$cls, pred.tree.both[,2])

## Area under the curve (AUC): 0.798

imbalance

Data

library("imbalance")
data(newthyroid1)
head(newthyroid1)

##   T3resin Thyroxin Triiodothyronine Thyroidstimulating TSH_value    Class
## 1     105      7.3              1.5                1.5      -0.1 negative
## 2      67     23.3              7.4                1.8      -0.6 positive
## 3     111      8.4              1.5                0.8       1.2 negative
## 4      89     14.3              4.1                0.5       0.2 positive
## 5     105      9.5              1.8                1.6       3.6 negative
## 6     110     20.3              3.7                0.6       0.2 positive

summary(newthyroid1)

##     T3resin         Thyroxin      Triiodothyronine Thyroidstimulating
##  Min.   : 65.0   Min.   : 0.500   Min.   : 0.20    Min.   : 0.10     
##  1st Qu.:103.0   1st Qu.: 7.100   1st Qu.: 1.35    1st Qu.: 1.00     
##  Median :110.0   Median : 9.200   Median : 1.70    Median : 1.30     
##  Mean   :109.6   Mean   : 9.805   Mean   : 2.05    Mean   : 2.88     
##  3rd Qu.:117.5   3rd Qu.:11.300   3rd Qu.: 2.20    3rd Qu.: 1.70     
##  Max.   :144.0   Max.   :25.300   Max.   :10.00    Max.   :56.40     
##    TSH_value           Class    
##  Min.   :-0.700   negative:180  
##  1st Qu.: 0.550   positive: 35  
##  Median : 2.000                 
##  Mean   : 4.199                 
##  3rd Qu.: 4.100                 
##  Max.   :56.300

numPositive <- length(which(newthyroid1$Class == "positive"))
numNegative <- length(which(newthyroid1$Class == "negative"))
nInstances <- numNegative - numPositive
newSamples <- pdfos(dataset = newthyroid1, numInstances = 80,
                    classAttr = "Class")
# Bind a balanced dataset
newDataset <- rbind(newthyroid1, newSamples)
head(newthyroid1)

##   T3resin Thyroxin Triiodothyronine Thyroidstimulating TSH_value    Class
## 1     105      7.3              1.5                1.5      -0.1 negative
## 2      67     23.3              7.4                1.8      -0.6 positive
## 3     111      8.4              1.5                0.8       1.2 negative
## 4      89     14.3              4.1                0.5       0.2 positive
## 5     105      9.5              1.8                1.6       3.6 negative
## 6     110     20.3              3.7                0.6       0.2 positive

summary(newthyroid1)

##     T3resin         Thyroxin      Triiodothyronine Thyroidstimulating
##  Min.   : 65.0   Min.   : 0.500   Min.   : 0.20    Min.   : 0.10     
##  1st Qu.:103.0   1st Qu.: 7.100   1st Qu.: 1.35    1st Qu.: 1.00     
##  Median :110.0   Median : 9.200   Median : 1.70    Median : 1.30     
##  Mean   :109.6   Mean   : 9.805   Mean   : 2.05    Mean   : 2.88     
##  3rd Qu.:117.5   3rd Qu.:11.300   3rd Qu.: 2.20    3rd Qu.: 1.70     
##  Max.   :144.0   Max.   :25.300   Max.   :10.00    Max.   :56.40     
##    TSH_value           Class    
##  Min.   :-0.700   negative:180  
##  1st Qu.: 0.550   positive: 35  
##  Median : 2.000                 
##  Mean   : 4.199                 
##  3rd Qu.: 4.100                 
##  Max.   :56.300

Figure

# Plot a visual comparison between new and old dataset
plotComparison(newthyroid1, newDataset,
               attrs = names(newthyroid1)[1:3], classAttr = "Class")