Multimodality Data Analysis in Information Security
1Harbin Engineering University, Harbin, China
2Carnegie Mellon University, Pittsburgh, USA
3Nanjing University of Science and Technology, Nanjing, China
Multimodality Data Analysis in Information Security
Description
As modern embedded devices, communication manufacture, and Internet technology have been significantly developed in the last decade, massive amounts of multimodality data can be easily acquired from electronic sensors, computers, mobile terminals, and various networks made up by them (e.g. the Internet, IoT, etc). Correspondingly, issues of information security in data exploitation and analysis inevitably arise.
Generally, multimodality data contains much more potential information available and is capable of providing an enhanced analytical result compared to mono-source data. The way to combine the data acquired from diverse sources suitably plays a crucial role in multimodality data analysis and is worth investigating. In addition, considering that multimodality data usually belongs to big data in practice, researchers have developed some technologies based on multimodal learning to enhance human analysis effectively and quickly at a low cost. The study of multimodal learning in information security has been attracting increasing numbers of researchers and practitioners in both academia and industry.
Original research contributions and review articles on multimodality data analysis and derivative issues of information security are solicited for the Special Issue. Research devoted to the improvement and optimization of the existing multimodality data analysis methods in information security, as well as work on new models, theories, and approaches for multimodality data, are encouraged.
Potential topics include but are not limited to the following:
- Multimodal source or sensor data fusion theory and application in information security
- Multimodal learning in cyber security, e.g. threat detection, malicious attack detection and identification, malware detection and classification, network analysis, endpoint protection, and vulnerability assessment, etc.
- Multimodality data representation, alignment, fusion, and co-learning
- Multimodality machine learning and data analysis
- Multimodal adversarial training for attacks and defence