Study of Machine and Deep Learning Classifications in Cyber Physical System

Cyber physical system is the most widely used infrastructure for solving different challenges in our day to day lives. In big data environment, taking a correct and rapid decision remains as a big task. Internet of Things or cyber physical systems will be used for transforming the manufacturing industry and other applications to the next level. But, along with these benefits of Cyber physical system, there are numerous problems that one face in their life due to the lack of smart analytical tools that affect the industries, which are not handling the large amount of data generated from the Internet connected devices. Even the skilled people are unable to handle, track or analyze these infrastructures. This creates a necessity for the integration of ML and cyber security in CPS, where it makes the skilled people to track threats on web within a less time period. The proposed research study is performed on the different frameworks used for Cyber-attack detection using learning approach, this proves the importance of machine learning and deep learning in Cyber physical system for detecting the threats in a better way. Security analytics is used by various researchers and also by using it one can prioritize the signals and alerts. The proposed study on different attacks has also highlighted the researchers to be more aware about uncommon attacks that can become very dangerous. Further, the study of various works done in analyzing different attacks are done using various approaches and dataset that is covered along with pros and cons to help in choosing the best approach according to the requirement.