Table 3.

Security solutions based on ML.

Citation	Year	Focused Area	Solution Technique	ML Type	Algorithm	Attack Type	Object
[135]	2019	Security	Machine Learning	SL	Decision-tree classifier	Vehicle misbehavior	Detect vehicle misbehavior
[136]	2019			SL	KNN and SVM	Malicious node attacks	Detect malicious node
[137]	2019			SL	CatBoost	Jamming attacks	Detect jamming attacks
[138]	2020			SL	Plausibility checks and traditional SL	A data-centric misbehavior	Misbehavior detection system for IoVs
[139]	2022			SL	RF, NB, and KNN	Backdoor, DDoS, and MITM attacks	To detect and mitigate various IoV attacks using ML algorithms
[140]	2022			SL	Eight SL models	Malicious messages	Classification of normal and malicious messages in vehicle network
[141]	2023			SL	RF	Falsification attacks	To protect IoV data, identify and prevent falsification attacks.
[142]	2019			UL	DCAEs	DoS attacks	Defend against DoS attacks
[143]	2020			UL	UL	Four types of attacks	Detect DoS attacks and three other types of attacks
[144]	2022			UL	K-Means, Gaussian Mixture, and Dbscan Clustering	DoS attack	To identify and mitigate DoS attacks that compromise connected vehicle function and safety
[145]	2023			UL	Median Absolute Deviation	Anomalies in V2V communication	To detect malicious nodes with low false-positive rates
[146]	2018			RL	Q-learning	Spoofing attack	Find spoofing data
[149]	2019			RL	DRL	Malicious node attacks	Signal authentication
[151]	2019			RL	Q-learning	DDoS attacks	Detect DDoS attacks
[152]	2019			RL	Q-learning	Jamming attack	Prevent jamming attack
[153]	2022			RL	Q-learning	Malicious data transmission in V2X communication	Classifying incoming data as legitimate or malicious improves security
[154]	2023			RL	DRL and ILP	Edge attacks	To improve network stability and enhancing security mechanisms