Skip to content
Related Articles

Related Articles

Improve Article

ML | Handling Imbalanced Data with SMOTE and Near Miss Algorithm in Python

  • Difficulty Level : Expert
  • Last Updated : 08 Sep, 2021

In Machine Learning and Data Science we often come across a term called Imbalanced Data Distribution, generally happens when observations in one of the class are much higher or lower than the other classes. As Machine Learning algorithms tend to increase accuracy by reducing the error, they do not consider the class distribution. This problem is prevalent in examples such as Fraud Detection, Anomaly Detection, Facial recognition, etc.
Standard ML techniques such as Decision Tree and Logistic Regression have a bias towards the majority class, and they tend to ignore the minority class. They tend only to predict the majority class, hence, having major misclassification of the minority class in comparison with the majority class. In more technical words, if we have imbalanced data distribution in our dataset then our model becomes more prone to the case when the minority class has a negligible or very lesser recall.
Imbalanced Data Handling Techniques: There are mainly 2 mainly algorithms that are widely used for handling imbalanced class distribution. 

  1. SMOTE
  2. Near Miss Algorithm

SMOTE (Synthetic Minority Oversampling Technique) – Oversampling

SMOTE (synthetic minority oversampling technique) is one of the most commonly used oversampling methods to solve the imbalance problem. 
It aims to balance class distribution by randomly increasing minority class examples by replicating them. 
SMOTE synthesizes new minority instances between existing minority instances. It generates the virtual training records by linear interpolation for the minority class. These synthetic training records are generated by randomly selecting one or more of the k-nearest neighbors for each example in the minority class. After the oversampling process, the data is reconstructed and several classification models can be applied for the processed data. 
More Deep Insights into how SMOTE Algorithm works! 
 

Attention reader! Don’t stop learning now. Get hold of all the important Machine Learning Concepts with the Machine Learning Foundation Course at a student-friendly price and become industry ready.



  • Step 1: Setting the minority class set A, for each 
    *** QuickLaTeX cannot compile formula:
     
    
    *** Error message:
    Error: Nothing to show, formula is empty
    

    , the k-nearest neighbors of x are obtained by calculating the Euclidean distance between x and every other sample in set A.

  • Step 2: The sampling rate N is set according to the imbalanced proportion. For each 
    *** QuickLaTeX cannot compile formula:
     
    
    *** Error message:
    Error: Nothing to show, formula is empty
    

    , N examples (i.e x1, x2, …xn) are randomly selected from its k-nearest neighbors, and they construct the set $A_1$  .

  • Step 3: For each example 
    *** QuickLaTeX cannot compile formula:
     
    
    *** Error message:
    Error: Nothing to show, formula is empty
    

    (k=1, 2, 3…N), the following formula is used to generate a new example: 
    $x' = x + rand(0, 1) * \mid x - x_k \mid$
    in which rand(0, 1) represents the random number between 0 and 1.

NearMiss Algorithm – Undersampling

NearMiss is an under-sampling technique. It aims to balance class distribution by randomly eliminating majority class examples. When instances of two different classes are very close to each other, we remove the instances of the majority class to increase the spaces between the two classes. This helps in the classification process. 
To prevent the problem of information loss in most under-sampling techniques, near-neighbor methods are widely used. 
The basic intuition about the working of near-neighbor methods is as follows: 
 



  • Step 1: The method first finds the distances between all instances of the majority class and the instances of the minority class. Here, the majority class is to be under-sampled.
  • Step 2: Then, n instances of the majority class that have the smallest distances to those in the minority class are selected. 
  • Step 3: If there are k instances in the minority class, the nearest method will result in k*n instances of the majority class.

For finding n closest instances in the majority class, there are several variations of applying the NearMiss Algorithm : 
 

  1. NearMiss – Version 1: It selects samples of the majority class for which average distances to the k closest instances of the minority class is smallest.
  2. NearMiss – Version 2: It selects samples of the majority class for which average distances to the k farthest instances of the minority class is smallest.
  3. NearMiss – Version 3: It works in 2 steps. Firstly, for each minority class instance, their M nearest-neighbors will be stored. Then finally, the majority class instances are selected for which the average distance to the N nearest-neighbors is the largest.

This article helps in better understanding and hands-on practice on how to choose best between different imbalanced data handling techniques.
 

Load libraries and data file

The dataset consists of transactions made by credit cards. This dataset has 492 fraud transactions out of 284, 807 transactions. That makes it highly unbalanced, the positive class (frauds) account for 0.172% of all transactions. 
The dataset can be downloaded from here
 

Python3




# import necessary modules
import pandas  as pd
import matplotlib.pyplot as plt
import numpy as np
from sklearn.linear_model import LogisticRegression
from sklearn.preprocessing import StandardScaler
from sklearn.metrics import confusion_matrix, classification_report
 
# load the data set
data = pd.read_csv('creditcard.csv')
 
# print info about columns in the dataframe
print(data.info())

Output: 
 

RangeIndex: 284807 entries, 0 to 284806
Data columns (total 31 columns):
Time      284807 non-null float64
V1        284807 non-null float64
V2        284807 non-null float64
V3        284807 non-null float64
V4        284807 non-null float64
V5        284807 non-null float64
V6        284807 non-null float64
V7        284807 non-null float64
V8        284807 non-null float64
V9        284807 non-null float64
V10       284807 non-null float64
V11       284807 non-null float64
V12       284807 non-null float64
V13       284807 non-null float64
V14       284807 non-null float64
V15       284807 non-null float64
V16       284807 non-null float64
V17       284807 non-null float64
V18       284807 non-null float64
V19       284807 non-null float64
V20       284807 non-null float64
V21       284807 non-null float64
V22       284807 non-null float64
V23       284807 non-null float64
V24       284807 non-null float64
V25       284807 non-null float64
V26       284807 non-null float64
V27       284807 non-null float64
V28       284807 non-null float64
Amount    284807 non-null float64
Class     284807 non-null int64

 

Python3




# normalise the amount column
data['normAmount'] = StandardScaler().fit_transform(np.array(data['Amount']).reshape(-1, 1))
 
# drop Time and Amount columns as they are not relevant for prediction purpose
data = data.drop(['Time', 'Amount'], axis = 1)
 
# as you can see there are 492 fraud transactions.
data['Class'].value_counts()

Output: 
 

       0    284315
       1       492

 

Split the data into test and train sets

 



Python3




from sklearn.model_selection import train_test_split
 
# split into 70:30 ration
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.3, random_state = 0)
 
# describes info about train and test set
print("Number transactions X_train dataset: ", X_train.shape)
print("Number transactions y_train dataset: ", y_train.shape)
print("Number transactions X_test dataset: ", X_test.shape)
print("Number transactions y_test dataset: ", y_test.shape)

Output: 
 

      Number transactions X_train dataset:  (199364, 29)
      Number transactions y_train dataset:  (199364, 1)
      Number transactions X_test dataset:  (85443, 29)
      Number transactions y_test dataset:  (85443, 1)

 

Now train the model without handling the imbalanced class distribution

 

Python3




# logistic regression object
lr = LogisticRegression()
 
# train the model on train set
lr.fit(X_train, y_train.ravel())
 
predictions = lr.predict(X_test)
 
# print classification report
print(classification_report(y_test, predictions))

Output: 
 

                precision   recall   f1-score  support

           0       1.00      1.00      1.00     85296
           1       0.88      0.62      0.73       147

    accuracy                           1.00     85443
   macro avg       0.94      0.81      0.86     85443
weighted avg       1.00      1.00      1.00     85443

The accuracy comes out to be 100% but did you notice something strange? 
The recall of the minority class is very less. It proves that the model is more biased towards the majority class. So, it proves that this is not the best model. 
Now, we will apply different imbalanced data handling techniques and see their accuracy and recall results.
 

Using SMOTE Algorithm

You can check all the parameters from here
 

Python3




print("Before OverSampling, counts of label '1': {}".format(sum(y_train == 1)))
print("Before OverSampling, counts of label '0': {} \n".format(sum(y_train == 0)))
 
# import SMOTE module from imblearn library
# pip install imblearn (if you don't have imblearn in your system)
from imblearn.over_sampling import SMOTE
sm = SMOTE(random_state = 2)
X_train_res, y_train_res = sm.fit_sample(X_train, y_train.ravel())
 
print('After OverSampling, the shape of train_X: {}'.format(X_train_res.shape))
print('After OverSampling, the shape of train_y: {} \n'.format(y_train_res.shape))
 
print("After OverSampling, counts of label '1': {}".format(sum(y_train_res == 1)))
print("After OverSampling, counts of label '0': {}".format(sum(y_train_res == 0)))

Output: 
 

Before OverSampling, counts of label '1': [345]
Before OverSampling, counts of label '0': [199019] 

After OverSampling, the shape of train_X: (398038, 29)
After OverSampling, the shape of train_y: (398038, ) 

After OverSampling, counts of label '1': 199019
After OverSampling, counts of label '0': 199019

Look! that SMOTE Algorithm has oversampled the minority instances and made them equal to the majority class. Both categories have an equal amount of records. More specifically, the minority class has been increased to the total number of the majority class. 
Now see the accuracy and recall results after applying SMOTE algorithm (Oversampling). 
 

Prediction and Recall
 

Python3




lr1 = LogisticRegression()
lr1.fit(X_train_res, y_train_res.ravel())
predictions = lr1.predict(X_test)
 
# print classification report
print(classification_report(y_test, predictions))

Output: 
 

                precision   recall   f1-score  support

           0       1.00      0.98      0.99     85296
           1       0.06      0.92      0.11       147

    accuracy                           0.98     85443
   macro avg       0.53      0.95      0.55     85443
weighted avg       1.00      0.98      0.99     85443

Wow, We have reduced the accuracy to 98% as compared to the previous model but the recall value of the minority class has also improved to 92 %. This is a good model compared to the previous one. The recall is great. 
Now, we will apply the NearMiss technique to Under-sample the majority class and see its accuracy and recall results. 
 

NearMiss Algorithm:

You can check all the parameters from here
 

Python3




print("Before Undersampling, counts of label '1': {}".format(sum(y_train == 1)))
print("Before Undersampling, counts of label '0': {} \n".format(sum(y_train == 0)))
 
# apply near miss
from imblearn.under_sampling import NearMiss
nr = NearMiss()
 
X_train_miss, y_train_miss = nr.fit_sample(X_train, y_train.ravel())
 
print('After Undersampling, the shape of train_X: {}'.format(X_train_miss.shape))
print('After Undersampling, the shape of train_y: {} \n'.format(y_train_miss.shape))
 
print("After Undersampling, counts of label '1': {}".format(sum(y_train_miss == 1)))
print("After Undersampling, counts of label '0': {}".format(sum(y_train_miss == 0)))

Output: 
 

Before Undersampling, counts of label '1': [345]
Before Undersampling, counts of label '0': [199019] 

After Undersampling, the shape of train_X: (690, 29)
After Undersampling, the shape of train_y: (690, ) 

After Undersampling, counts of label '1': 345
After Undersampling, counts of label '0': 345

The NearMiss Algorithm has undersampled the majority instances and made it equal to majority class. Here, the majority class has been reduced to the total number of the minority classes, so that both classes will have an equal number of records. 
 

Prediction and Recall
 

Python3




# train the model on train set
lr2 = LogisticRegression()
lr2.fit(X_train_miss, y_train_miss.ravel())
predictions = lr2.predict(X_test)
 
# print classification report
print(classification_report(y_test, predictions))

Output: 
 

               precision    recall   f1-score   support

           0       1.00      0.56      0.72     85296
           1       0.00      0.95      0.01       147

    accuracy                           0.56     85443
   macro avg       0.50      0.75      0.36     85443
weighted avg       1.00      0.56      0.72     85443

This model is better than the first model because it classifies better and also the recall value of minority class is 95 %. But due to the undersampling of the majority class, its recall has decreased to 56 %. So in this case, SMOTE is giving me great accuracy and recall, I’ll go ahead and use that model! 🙂
 




My Personal Notes arrow_drop_up
Recommended Articles
Page :