Sign in Subscribe
Algorithms

Optimizing Imbalanced Learning with Genetic Algorithm

Strategic Insights

— 1 min read

Research investigators in the Department of Computer Science at Namal University in Pakistan have proposed a novel approach to generate synthetic data using Genetic Algorithms (GAs) to address the challenge of class imbalance in machine learning models. The team aimed to outperform traditional methods such as SMOTE, ADASYN, GAN, and VAE in terms of model performance. Experimental results on three datasets, including Credit Card Fraud Detection, PIMA Indian Diabetes, and PHONEME, demonstrated that the proposed method significantly outperformed the previous techniques based on commonly used performance metrics.

Key Takeaways:

  • The researchers proposed a novel approach to generate synthetic data using Genetic Algorithms (GAs) to address the challenge of class imbalance in machine learning models.
  • The team aimed to outperform traditional methods such as SMOTE, ADASYN, GAN, and VAE in terms of model performance.
  • Experimental results on three datasets (Credit Card Fraud Detection, PIMA Indian Diabetes, and PHONEME) demonstrated that the proposed method significantly outperformed the previous techniques.
  • The proposed method was able to analyze the Simple as well as the Elitist Genetic Algorithms, along with Logistic Regression and Support Vector Machines to evaluate the population initialization and fitness function.
  • The researchers concluded that this highlights the potential of GAs in the development of accurate and reliable AI models for imbalanced datasets.

Statistics:

  • The proposed method was evaluated on three datasets: Credit Card Fraud Detection, PIMA Indian Diabetes, and PHONEME.
  • The method achieved significant improvement in performance metrics, including accuracy, precision, recall, F1-score, ROC-AUC, and AP (Accuracy-Precision) curve.
  • The experimental results demonstrated that the proposed method outperformed the previous techniques by up to 15% in terms of accuracy.

Sources:

  • Optimizing imbalanced learning with genetic algorithm. Scientific Reports, 2025;15(1):34857.
  • Nature Portfolio, Heidelberger Platz 3, Berlin, 14197, Germany.

Read more