The network has come a long way with the advances of Software Defined Network (SDN) and Network Function Virtualization (NFV). It generates a huge amount of data (big data) in daily life. Therefore, it is very difficult to manually analyze all the data by the network professionals and to decide whether the network is good enough to manage all the data or an adjustment is required. Moreover, security issues are continuously increasing in the network due to the presence of numerous hackers and malicious users. Therefore, there should be a strong security technique to protect data against hackers and malicious users. Currently, Machine Learning (ML) algorithms are used for network management by many researchers. Machine learning is the study of mathematical model-based algorithms that improve automatically through past experience. ML algorithms are based on data to make decisions without being explicitly programmed to do so. There are many applications of ML in daily life, such as smart email categorization, chatbot, marketing, healthcare, gaming, plagiarism check, autonomous vehicle and many more. Nowadays, ML is used in industry and academia due to the data driven feature for achieving high performance of a network. ML algorithms can be run throughout the network without the need of any external hardware to predict the potential problems of the network prior to occurring. In addition, ML algorithms can also recognize the problems of a network and can make a recommendation to fix them. Thus, the network professionals can take a decision to manage the network efficiently. Nowadays, new attacks are being developed every day by the attackers and it is very difficult to detect them by using the traditional intrusion detection techniques. ML algorithms can be developed to train a network for detecting sophisticated attacks, which are similar to the already defined known attacks. It is important to improve the algorithms, so that there is a good trade-off between learning cost and detection accuracy. Recent research has also shown the negative impact of ML as these advanced fields support new attack tools by using the adversarial machine learning techniques to develop new attacks. Attackers and malicious users can also hack ML algorithms by altering the training data and modifying the classification function of ML, which can directly affect the detection accuracy of a network. These types of threats are very critical. Therefore, novel techniques of cybersecurity must be developed to protect the network.
This workshop gives a platform for the researchers, academicians and industry professionals to present their research work on ML in network management. This workshop aims to address the challenges and issues of applying ML in network.
Theoretical as well as experimental research works on the following topics (but are not limited to) are within the scope of this workshop:
- Learning from network data
- Issues to apply ML in network
- ML for network managementML for service placement
- ML for predictive analysis in network
- ML for network slicing optimization
- ML for predicting user behavior in network
- Pattern recognition and classification for network
- Analysis, modelling and visualization of network using ML
- ML for 5G
- ML in network security and privacy
- Security threats, intrusions and malware detection exploiting ML methods
- Challenges of black-box attacks in ML methods
- ML driven attack model generation and specification
- ML based cryptographic protocols in network
- ML based identity management in network
- ML for big data security/cloud security/IoT security
- Emerging technologies in network management
All registered papers will be submitted for publishing by Springer and made available through SpringerLink Digital Library.
Proceedings will be submitted for inclusion in leading indexing services, such as Web of Science, EI Engineering Index (Compendex and Inspec databases), DBLP, EU Digital Library, Google Scholar, IO-Port, MathSciNet, Scopus, Zentralblatt MATH.
Selected papers of this workshop will be invited to submit an extended version to:
Community Review is a service offered to Program Committees and submitting Authors of all EAI conferences designed to improve the speed and the quality of the review process.
Abstracts of all authors who opt in to Community Review during submission will be published and available for Bidding here.
We invite workshop participation through contributions that respond to one or more of the mentioned research questions in (6-11) pages. Short papers should be submitted through EAI ‘Confy+‘ system, and have to comply with the Springer format (see Author’s kit section).
Paper Submission deadline
10 July 2021 30 July 2021
27 August 2021
27 September 2021
- Suyel Namasudra, PhD
National Institute of Technology Patna, Bihar, India
- Prof. Pascal Lorenz
University of Haute-Alsace, Colmar, France
- Uttam Ghosh, PhD
Vanderbilt University, Tennessee, USA
Dr. Deepal Tosh, University of Texas at El paso, TX, USA
Prof. Jalel Ben-othman, University of Paris 13, France
Dr. M. D. Borah, National Institute of Technology Silchar, India
Dr. Gautam Srivastava, Brandon University, Canada
Dr. Jaime Lloret Mauri, Polytechnic University of Valencia, Spain
Dr. Maanak Gupta, Tennessee Tech University, TN, USA
Dr. Sanjeet Kumar Nayak, Indian Institute of Information Technology, Design and
Manufacturing, Kancheepuram, India
Prof. Denis A. Pustokhin, State University of Management, Russia
Prof. Jerry Chun-wei Lin, Western Norway University of Applied Sciences, Norway
Dr. Ganesh Chandra Deka, Ministry of Skill Development & Entrepreneurship, Govt. of