Federated Learning and its Application in Malware Detection
Sakshi Bhagwat, Govind P. Gupta · 2023
In last few years, we have seen a rapid increase in Android malware. Cyber-crimes have increased in which the main weapon for attack is malware, and the medium used to execute the malware attack is the Internet. On mobile systems, malware can enter through mail, Web sites, or applications downloaded from the Internet. There are different types of malware, including adware and spyware as well as the worm, Trojan, virus, rootkit, backdoor, ransomware, and browser hijacker. Static and dynamic techniques are used to detect malware and benign samples in malware datasets. Due to the rise of malicious activities, malware detection is necessary such that the use of machine learning and deep learning is increasing. To increase the accuracy of detection, artificial intelligence methods are used for efficient monitoring and detection systems. Federated leaning (FL) is a distributed learning framework. It is a new machine learning technique for training algorithms across local data available at decentralized edge devices or the server. FL provides confidentiality for the data; the dataset shared between participant and server is limited to the model. This chapter aims to provide a comparative study about malware detection using FL.