In the emerging era of Web 3.0, securing cyberspace has gradually evolved into a
critical organizational and national research agenda inviting interest from a multidisciplinary
scientific workforce. There are many avenues into this area, and, in recent
research, machine-learning and data-mining techniques have been applied to design,
develop, and improve algorithms and frameworks for cybersecurity system design.
Intellectual products in this domain have appeared under various topics, including
machine learning, data mining, cybersecurity, data management and modeling,
and privacy preservation. Several conferences, workshops, and journals focus on the
fragmented research topics in this area. However, transcendent and interdisciplinary
assessment of past and current works in the field and possible paths for future research
in the area are essential for consistent research and development.
This interdisciplinary assessment is especially useful for students, who typically
learn cybersecurity, machine learning, and data mining in independent courses.
Machine learning and data mining play significant roles in cybersecurity, especially
as more challenges appear with the rapid development of information discovery
techniques, such as those originating from the sheer dimensionality and heterogeneous
nature of the network data, the dynamic change of threats, and the severe
imbalanced classes of normal and anomalous behaviors. In this book, we attempt
to combine all the above knowledge for a single advanced course.
This book surveys cybersecurity problems and state-of-the-art machine-learning
and data-mining solutions that address the overarching research problems, and it is
designed for students and researchers studying or working on machine learning and
data mining in cybersecurity applications. The inclusion of cybersecurity in machinelearning
research is important for academic research. Such an inclusion inspires fundamental
research in machine learning and data mining, such as research in the
subfields of imbalanced learning, feature extraction for data with evolving characteristics,
and privacy-preserving data mining.