Machine Learning and Big Data Analytics for IoT Security
摘要截稿:
全文截稿: 2019-07-15
影响因子: 6.125
期刊难度:
CCF分类: C类
中科院JCR分区:
• 大类 : 计算机科学 - 2区
• 小类 : 计算机:理论方法 - 1区
Overview
The "Internet of things" heralds the connections of a nearly countless number of devices to the internet thus promising accessibility, boundless scalability, amplified productivity and a surplus of additional paybacks. The hype surrounding the IoT and its applications is already forcing companies to quickly upgrade their current processes, tools, and technology to accommodate massive data volumes and take advantage of insights. Since there is a vast amount of data generated by the IoT, a well-analysed data is extremely valuable. However, the large-scale deployment of IoT will bring new challenges and IoT security is one of them.
The philosophy behind machine learning is to automate the creation of analytical models in order to enable algorithms to learn continuously with the help of available data. Continuously evolving models produce increasingly positive results, reducing the need for human interaction. These evolved models can be used to automatically produce reliable and repeatable decisions. Today's machine learning algorithms comb through data sets that no human could feasibly get through in a year or even a lifetime's worth of work. As the IoT continues to grow, more algorithms will be needed to keep up with the rising sums of data that accompany this growth.
One of the main challenge of the IoT security is the integration with communication, computing, control, and physical environment parameters to analyse, detect and defend cyber-attacks in the distributed IoT systems. The IoT security includes: (i) the information security of the cyber space, and (ii) the device and environmental security of the physical space. These challenges call for novel approaches to consider the parameters and elements from both spaces, and get enough knowledge for ensuring the IoT's security. As the data has been collecting in the IoT, and the data analytics has been becoming mature, it is possible to conquer this challenge with novel machine learning or deep learning methods to analyse the data which synthesize the information from both spaces.
Therefore, this special issue will explore the potentials of machine learning and big data analytics by going beyond the existing simple approaches and present more advanced practices with well authentic implementations and results both at academic and industrial level. We believe that machine learning and big data analytics will play a vital role to provide and enhance the security of IoT and enable organizations to make crucial changes to their security landscape.
Topics of Interest
This special issue seeks recent important contributions on the machine learning and big data analytics for IoT security, with an emphasis on interdisciplinary approaches. Some topics of interest include, but are not limited to:
Novel machine learning and big data analytics methods for IoT security
Big data analytics/machine learning/deep learning for IoT security such as smart grid security analytics
Data mining and statistical modelling for the secure IoT
Machine learning and big data analytics architectures for IoT security
Machine learning based security detecting protocols
Machine learning experiments, test-beds and prototyping systems for IoT security
Analytics and machine learning applications to IoT security
Data based metrics and risk assessment approaches for IoT
Data confidentiality and privacy in IoT
Authentication and access control for data usage in IoT
Data-driven co-design of communication, computing and control for IoT security
Big data analytics/machine learning/deep learning edge/fog security