You are invited to submit papers to the special session on “Computational Intelligence in Cyber Threat Analytics" 18 - 21 November, 2018, Bengaluru, INDIA
In the last couple of years, the growing demand for digitalization in society has led to the increase in computational capabilities, volume of data, and as a result to more cyber threats. Artificial and computational intelligence has attained a huge popularity in providing solutions to the numerous challenges arising in the field of computer vision, natural language processing, knowledge management, decision making, etc. But, less has been contributed towards the area of cyber threat intelligence (CTI) to uncover cyber attacks and cyber espionage to the information and communication technology in real-time. Recently, cyber security experts and digital forensic analysts have been facing big challenges in providing adaptive and timely response for new and unknown threats to the decision makers. To address this problem, computational intelligence algorithms can be used to design effective and efficient methods to collect, structure and analyze sheer volume of threat data enabling identification of threat actors, their behaviors, resources, and attack methodologies.
By combining data from multiple sources (also termed as data fusion or data enrichment), computational intelligence can further boost the performance of traditional cyber security protection and response methods. The main objective of applying computational intelligence in cyber threat analytics is to establish correlation between different cyber events, utilize historical information to detect threat patterns, trends, and anomalies in the threat data. This helps in providing more sophisticated, targeted, and tailored threat alerts, by adding context to these alerts. Further, it enables industry and government organizations to increase the quality of their decisions regarding particular threat alerts. As result, performance and reliability of decision support systems in cyber incidents handling can be considerably improved.
Only technical papers describing previously unpublished, original, state-of-the-art research, and not currently under review by a conference or a journal will be considered for publication. Extended work must have a significant number of ”new and original” contributions along with more than 60 percent brand ”new” material. All accepted papers must be presented by at least one of the authors to be published in the electronic proceedings of the conference in IEEE Xplore Digital Library.
Topics for this special issue
This special issue focuses on sharing recent advances in computational intelligence, machine learning techniques, algorithms, methods and tools to perceive, reason, learn and act on a wide range of cyber threat data available from multiple sources such as malware campaigns, network attacks, HackForums , LeakForums , Darknets etc. Topics appropriate for this special issue include novel supervised, unsupervised, semisupervised and reinforcement machine learning algorithms, new formulations, and applications in cyber threat intelligence (but are not necessarily limited to):
- Open source threat intelligence data-driven methods.
- Methods and models for cyber threat graphs representation learning.
- Heterogeneous data and machine learning model fusion.
- Machine learning based recommender systems for CTI.
- Open source social media intelligence (OSSMINT) for government and law enforcement agencies. (i.e. Illegal trading of products & services, online radicalization, revealing terrorist threats and extremism etc.)
- Malware entity identification and classifications.
- Potential threat actors and their suspicious behaviour modelling.
- Machine learning in intrusion detection systems, mitigation and response techniques.
- Cyber threat predictive analysis on virtual security products.
- Algorithms/Platforms for sharing and exchange cyber threat knowledge.
- Privacy-preserving approaches to cyber threat intelligence information release.
- Threat actors, tactics, techniques, relationships and behaviour modelling using PRE-ATT&CK Matrix, MISP Galaxy clusters
Paper Submission Guidelines
The submission guidelines are adopted from the main conference. Authors are required to submit their manuscripts for this special session at the regular paper submission website. Papers should not exceed a maximum of 8 pages (including abstract, body, tables, figures, and references), and should be submitted, written in English, as a pdf in 2-column IEEE format. Template in both LaTex and Word are available at IEEE website Detailed instructions for submitting papers are provided on the conference at home page.
July 23, 2018 : Papers submission deadline (Extended)
August 15, 2018: Notification of paper acceptance to authors
September 15, 2018: Early bird registration
November 18-21, 2018: Special Session