This paper presents an innovative approach for detecting Distributed Denial-of-Service (DDoS) attacks using evolutionary algorithms (EAs) optimization combined with machine learning (ML) techniques. The study proposes three methods: XGB-GA Optimization, RF-GA Optimization, and SVM-GA Optimization, which integrate genetic algorithms (GA) with tree-based pipelines optimization tool (TPOT)-genetic programming. These methods were tested on DDoS attack datasets, achieving high accuracy scores of 99.99%, 99.50%, and 99.99%, respectively. The study also employed TPOT to identify the optimal algorithm for constructing a machine learning model, with XGB-GA identified as the most effective choice. The proposed methodology significantly enhances the accuracy and robustness of DDoS attack detection, contributing to improved cybersecurity and digital infrastructure protection. The research utilized publicly available datasets, including KDD Cup 99 and CIC-IDS 2017, and employed 10-fold cross-validation for model evaluation. The study also discussed the use of TPOT for automated machine learning pipeline optimization, highlighting its effectiveness in identifying the best ML pipeline for DDoS detection. The results demonstrated that the proposed methods outperformed other existing techniques in terms of accuracy and efficiency. The study also compared the performance of different ML algorithms, including XGBoost, Random Forest, and SVM, and found that the XGB-GA method achieved the highest accuracy. The research concludes that integrating EAs with ML techniques provides a powerful solution for detecting DDoS attacks, offering a robust and accurate methodology for enhancing cybersecurity.This paper presents an innovative approach for detecting Distributed Denial-of-Service (DDoS) attacks using evolutionary algorithms (EAs) optimization combined with machine learning (ML) techniques. The study proposes three methods: XGB-GA Optimization, RF-GA Optimization, and SVM-GA Optimization, which integrate genetic algorithms (GA) with tree-based pipelines optimization tool (TPOT)-genetic programming. These methods were tested on DDoS attack datasets, achieving high accuracy scores of 99.99%, 99.50%, and 99.99%, respectively. The study also employed TPOT to identify the optimal algorithm for constructing a machine learning model, with XGB-GA identified as the most effective choice. The proposed methodology significantly enhances the accuracy and robustness of DDoS attack detection, contributing to improved cybersecurity and digital infrastructure protection. The research utilized publicly available datasets, including KDD Cup 99 and CIC-IDS 2017, and employed 10-fold cross-validation for model evaluation. The study also discussed the use of TPOT for automated machine learning pipeline optimization, highlighting its effectiveness in identifying the best ML pipeline for DDoS detection. The results demonstrated that the proposed methods outperformed other existing techniques in terms of accuracy and efficiency. The study also compared the performance of different ML algorithms, including XGBoost, Random Forest, and SVM, and found that the XGB-GA method achieved the highest accuracy. The research concludes that integrating EAs with ML techniques provides a powerful solution for detecting DDoS attacks, offering a robust and accurate methodology for enhancing cybersecurity.