Robust Blockchain and Artificial Intelligence-Based Efficient Techniques for Network Intrusion Detection System to Improve Cybersecurity

IoT expansion and distributed system developments together create new cybersecurity obstacles while demanding solutions that scale effectively and deliver precise results. The current generation of intrusion detection systems (IDS) demonstrates inadequate capability to deliver both data security assurance and immediate performance. This paper proposes a robust Blockchain and Artificial Intelligence (AI)-based IDS to enhance cybersecurity. The system uses the UNSW-NB15 dataset together with advanced preprocessing techniques along with feature selection and SMOTE-based class balancing to achieve optimal data quality and deal with imbalanced datasets. Random Forest (RF) demonstrates the best performance at 98.6% accuracy, while other models, KNN and SVM, detect malicious activities with the solution. Blockchain technology maintains data authentication by ensuring immutability combined with smart contracts for secure distributed device communication. Performance evaluation criteria demonstrate the system's success by measuring F1-score along with recall, accuracy, and precision. This technique provides scalable protection that operates in real time with encryption capabilities to handle current cybersecurity requirements.