The emergence of the Internet of Things, cloud computing, and quantum computing has led to a rapid increase in the amount of data produced, commonly referred to as ‘big data’. This deluge of data is a tremendous asset for machine learning (ML) model training, with revolutionary potential to strengthen security protocols. The growing intricacy of cyber threats within network security necessitates a shift toward more adaptable and resilient solutions, especially when it comes to intrusion detection systems (IDS). Firewalls and other traditional security measures are unable to keep up with the constantly changing threats, which leads to a high false positive and negative rate. Recognizing the urgency for robust network protocols, this paper engages in an in-depth study, surveying diverse papers that explore the integration of supervised learning and unsupervised learning ML techniques into IDS. ‘Neural networks’, ‘support vector machines’, ‘decision trees’, and ‘clustering algorithms’ are some of the machine learning methods that are being examined. In the digital age, big data and machine learning integration appears to be a potential path toward improving cybersecurity. The goal of the paper is to improve the accuracy of recognizing attacks on the given dataset by outlining important machine learning techniques and to identify the most effective one in improving intrusion detection performance.