“SynthDefender”: Orchestrating Cyber Resilience with SVM-ANN Union against Reconnaissance Intrusion

In order to find vulnerabilities in a system or network, an attacker will do reconnaissance to gather relevant data. Other types of attacks, such as denial-of-service (DoS) and malware insertion, may be planned and carried out using this data. It is critical to find security holes and ensure sensitive data is protected. Innovative methods, including machine learning, may be able to identify and foil these cybercriminals, making the world's IoT networks more trustworthy and secure. In order to detect possible reconnaissance attacks, this research employs a mixed method that combines Support Vector Machine (SVM) and Artificial Neural Network (ANN). By integrating ANN for pattern recognition and Support Vector Machines (SVM) for point classification, an effective learning approach is generated. They want to enhance the accuracy and efficiency of Reconnaissance Intruder detection in order to reduce the effect of network attacks. Finding surveillance incursions using data from the CICIoT23 database is the main objective of the research. Algorithms that detect assassination attacks in IoT networks might benefit greatly from this dataset, because of its large and diverse sample size.