Genetic Algorithm-Based Feature Selection for Network Intrusion Detection Using Machine Learning

@article{1707438,
      author = {Umukoro Gift , Fasanmi Olufemi Ajiroghene Ezekiel},
      title = {Genetic Algorithm-Based Feature Selection for Network Intrusion Detection Using Machine Learning},
      journal = {Iconic Research And Engineering Journals},
      year = {2025},
      volume = {8},
      number = {9},
      pages = {288-296},
      issn = {2456-8880},
      url = {https://www.irejournals.com/formatedpaper/1707438.pdf},
      abstract = {Network Intrusion Detection NID) plays a critical role in identifying and mitigating security threats in modern networks. In this study, we use an MLP classifier combined with a genetic algorithm (GA) for feature selection to enhance the model?s performance in NID tasks. We investigate the model across different generations?N=5, N=10, N=15, N=20, and N=25?to assess its performance with selected features. The results are compared with a non-optimized model to highlight the improvements gained through feature selection. Key metrics such as Accuracy, Precision, Recall, and F1 Score demonstrate significant gains as the number of generations increases. The model achieves peak performance at N=20, with accuracy reaching 99.23%, after which further generations show minimal improvement, indicating the presence of Overlapping Behavior (OBE). These findings suggest that the genetic algorithm converges to an optimal feature set by the 20th generation, showcasing the importance of feature selection in improving NID model performance while optimizing computational efficiency.},
      keywords = {Genetic Algorithm, Multi-Layer Perceptron (MLP), Network Intrusion Detection, Machine Learning.},
      month = {March},
  }