Multi-objective Particle Swarm Optimization for Botnet Detection in Internet of Things
Author(s)
Habib, Maria
Aljarah, Ibrahim
Faris, Hossam
Mirjalili, Seyedali
Griffith University Author(s)
Year published
2020
Metadata
Show full item recordAbstract
Nowadays, the world witnesses an immense growth in Internet of things devices. Such devices are found in smart homes, wearable devices, retail, health care, industry, and transportation. As we are entering Internet of things (IoT) digital era, IoT devices not only hack our world, but also start to hack our personal life. The widespread IoT has created a rich platform for potential IoT cyberattacks. Data mining and machine learning techniques have significant roles in the field of IoT botnet detection. The aim of this chapter is to develop detection model based on multi-objective particle swarm optimization (MOPSO) for ...
View more >Nowadays, the world witnesses an immense growth in Internet of things devices. Such devices are found in smart homes, wearable devices, retail, health care, industry, and transportation. As we are entering Internet of things (IoT) digital era, IoT devices not only hack our world, but also start to hack our personal life. The widespread IoT has created a rich platform for potential IoT cyberattacks. Data mining and machine learning techniques have significant roles in the field of IoT botnet detection. The aim of this chapter is to develop detection model based on multi-objective particle swarm optimization (MOPSO) for identifying the malicious behaviors in IoT network traffic. The performance of MOPSO is verified against multi-objective non-dominating sorting genetic algorithm (NSGA-II), common traditional machine learning algorithms, and some conventional filter-based feature selection methods. As per the obtained results, MOPSO is competitive and outperforms NSGA-II, traditional machine learning methods, and filter-based methods in most of the studied datasets.
View less >
View more >Nowadays, the world witnesses an immense growth in Internet of things devices. Such devices are found in smart homes, wearable devices, retail, health care, industry, and transportation. As we are entering Internet of things (IoT) digital era, IoT devices not only hack our world, but also start to hack our personal life. The widespread IoT has created a rich platform for potential IoT cyberattacks. Data mining and machine learning techniques have significant roles in the field of IoT botnet detection. The aim of this chapter is to develop detection model based on multi-objective particle swarm optimization (MOPSO) for identifying the malicious behaviors in IoT network traffic. The performance of MOPSO is verified against multi-objective non-dominating sorting genetic algorithm (NSGA-II), common traditional machine learning algorithms, and some conventional filter-based feature selection methods. As per the obtained results, MOPSO is competitive and outperforms NSGA-II, traditional machine learning methods, and filter-based methods in most of the studied datasets.
View less >
Book Title
Evolutionary Machine Learning Techniques
Subject
Artificial intelligence