A Hybrid Machine Learning Model for Position Falsification Attacks for Intrusion Detection in VANET

The majority of self-driving vehicles are susceptible to various types of attacks because of their dynamic network along with communication design topology. These kinds of vehicles are dependent on outside VANET communication bases. Industry and academia have shown a lot of interest in it, but road safety, traffic congestion and security have not been correctly addressed in fresh centuries. Building a safe framework for the VANET communication scheme along with being able to sense various kinds of attacks are the greatest significant requirements of network security, have been adequately studied by numerous students, to address these issues. This work suggests a Hybrid DSVM scheme that is based on Support Vector Machine (SVM) and Decision Tree (DT) Machine Learning algorithms to build a secure framework to detect attack, to improve performance as well as adapt to the VANET scenario. The exploratory outcomes show that this approach gives the improved outcomes when contrasted with various AI based Calculations to identify assault.