Intrusion Detection Method Based on Genetic Algorithm of Optimizing LightGBM
Zhanbo Li, Xiaoyang Li · Proceedings of the 2021 5th International Conference on Electronic Information Technology and Computer Engineering · 2021
In response to the high-precision requirements of intrusion detection systems, an intrusion detection method based on genetic algorithm of optimizing LightGBM is proposed. The recursive feature elimination algorithm is used to select the optimal feature subset, and a weighted loss function is designed to solve the problem of imbalanced network traffic data. Aiming at the problem that the performance of LightGBM is greatly affected by parameters and the cumbersome parameter adjustment, the powerful global search capability of genetic algorithm is used to optimize LightGBM and automatically determine the optimal parameter combination. Using the CIC-IDS2017 data set experiment, the experimental results show that the accuracy of this method is as high as 99.88%, which has higher detection accuracy than other methods.