Page 269 - Artificial Intelligence in the Age of Neural Networks and Brain Computing

References 261

Embedding (t-SNE) [23]. In order to identify a particular type of cyberattack, clas-
sifiers must be designed, for example, Support Vector Machine (SVM) and random
forests.
The above mentioned 17 features intentionally eliminate detailed information such
as port numbers and header information of a TCP packet. This type of abstraction is
important to define a broad feature space covering unknown malicious activities. Once
unknown darknet traffic patterns are identified using an anomaly detection method; a
more precise analysis such as identifying a specific malware type is conducted. In a
way what is done within the cyberattack protection intelligent functionality is similar
to some steps introduced by FDSs where, at first malware is detected, secondly, iden-
tified and, third, mitigation strategies are taken into account.

7. CONCLUSIONS
This chapter discusses how machine learning and computational intelligent tech-
niques could be important elements of the next generation of cyber-physical systems
(CPS) and Internet of Things (IoT). Intelligent mechanisms embedded on devices
and systems are aimed to work with appropriate performances under an evolving,
time-variant environment, optimally harvest the available energy and manage the
residual energy, reduce the energy consumption of the whole system, identify and
mitigate occurrence of faults, and provide a continuously secure platform against
cyberattacks. All these efforts are still ongoing and there are many open problems
to be solved. A huge amount of CPS and IoT devices have been deployed and
supporting our life, and our dependence on CPS and IoT technologies will be contin-
uously increasing. Without intelligent technologies, it would be quite difficult for us
to manage such big systems operating in a good shape. We hope that this chapter
would provide some insights on a prospective research direction for computational
intelligence community.

ACKNOWLEDGMENTS
The authors thank Dr. Ban Tao, Dr. Junji Nakazato (National Institute of Information and
Communications Technology, Japan), and Mr. Jumpei Shimamura (Clwit Inc.) for providing
the darknet traffic data and their expert knowledge on cybersecurity. This chapter partially
contains the experimental results in the research project, supported by the Ministry of Educa-
tion, Science, Sports and Culture, Grant-in-Aid for Scientific Research (B) 16H02874.

REFERENCES
[1] International Data Corporation (IDC), Final Study Report: Design of Future Embedded
Systems, 2012.