Data Archive

Imbalanced Datasets

Handling imbalanced datasets is a common challenge in machine learning, particularly in classification tasks where one class significantly outnumbers the other(s).

In Classification tasks, an imbalanced dataset can lead to a model that performs well on the majority class but poorly on the minority class. This is because the model may learn to predict the majority class more often due to its prevalence.

For Regression tasks, handling outliers or data skewness might be necessary.

Techniques

  1. Resampling:

    • Oversampling: Increase the number of instances in the minority class by duplicating existing samples or generating new ones.
    • Undersampling: Reduce the number of instances in the majority class by randomly removing samples.

  2. Synthetic Data Generation:

    • SMOTE (Synthetic Minority Over-sampling Technique): Generate synthetic samples for the minority class by interpolating between existing samples.

  3. Weighted Loss Functions:

    • Assign higher weights to the minority class during training to penalize misclassification more heavily.

  4. Model Evaluation

    • Use metrics like precision, recall, F1-score, and AUC-ROC instead of accuracy to better evaluate model performance on imbalanced data.

Example: Handling Imbalanced Data with SMOTE

Consider a scenario where you have an imbalanced dataset of resumes, with a majority of male resumes and a minority of female resumes. You want to build a model to predict gender based on resume features.

import pandas as pd
from sklearn.model_selection import train_test_split
from imblearn.over_sampling import SMOTE
from sklearn.ensemble import RandomForestClassifier
from sklearn.metrics import classification_report
 
# Sample dataset
data = {
    'feature1': [1, 2, 3, 4, 5, 6, 7,8 , 9, 10],
    'feature2': [5, 4, 3, 2, 1, 0, 1, 2, 3, 4],
    'gender': ['male', 'male', 'male', 'male', 'male', 'female', 'female', 'female', 'female', 'female']
}
 
df = pd.DataFrame(data)
 
# Split data into features and target
X = df[['feature1', 'feature2']]
y = df['gender']
 
# Split into training and test sets
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
 
# Apply SMOTE to balance the dataset
smote = SMOTE(random_state=42)
X_resampled, y_resampled = smote.fit_resample(X_train, y_train)
 
# Train a classifier
clf = RandomForestClassifier(random_state=42)
clf.fit(X_resampled, y_resampled)
 
# Predict and evaluate
y_pred = clf.predict(X_test)
print(classification_report(y_test, y_pred))

SMOTE: This technique generates synthetic samples for the minority class (female resumes) by creating new instances that are interpolations of existing ones.

Handling imbalanced datasets is crucial for building robust models that perform well across all classes. Techniques like SMOTE, resampling, and using appropriate evaluation metrics can significantly improve model performance on minority classes.

Graph View