Computational intelligence applied to cybersecurity

This special issue compiles recent applications of Computational Intelligence in the security domain. It is aimed at both researchers and practitioners from academia industry who are engaged deploying intelligent solutions for securing information systems. Four papers included this issue, covering a wide variety case studies, ranging assisted driving to explainable robotics. Additionally, cutting-edge techniques proposed, such as blockchain autoencoders among others. All these works contribute scientific progress domain, going one step further. In first contribution by Barreno et al., it proposed fuzzy expert system identifying classifying conventional two-lane roads based on their geometric characteristics expected vehicle performance (passenger cars, trucks, buses). The road features were measured sensors an equipped vehicle, travelling real-life located Madrid region (Spain). Fuzzy neuro-fuzzy systems applied order classify sections current characteristics. risk identification presented. Its main target assess whether with inappropriate speed. do that, model some variables type, longitudinal gradient, angle covered each horizontal curve, existence or not additional lane traffic. A index verify relatively unsafe sections, where more accidents prone happen. next contribution, Wilusz Wojtowicz, proposes architecture cryptocurrency insurance framework along set corresponding communication protocols smart contract template that jointly reduce various risks which cryptocurrency-based transactions exposed. Some threats unintentional credential leaks, fraudulent exchanges broken private keys Authors analyse present solution can potentially enable new business models technology. framework, holder transfers insurer applies technological countermeasures secure unit. Funds be transferred bilateral agreement unit thanks use multi-signatures required contracts. analysed standpoint, guaranteeing deployed several practical scenarios domains applications. Zayas-Gato al. address challenging up-to-date detecting anomalies industrial automatically detect anomalies, hybrid proposed. First, data clustered Density-Based Spatial Clustering Applications With Noise (DBSCAN) algorithm. Then, eight one-class benchmarked normal operation plants, namely: NCBoP, Autoencoders, Gauss, K-Centers, Minimum Spanning Trees, Parzen Density Estimator, Principal Component Analysis, Support Vector Data Description. Authors' proposal validated using two gathered during different operating points plants. controlling liquid level tank while second manufacturing wind generator blades made carbon fibre material. approach used improve classification when scattered clusters groups. As result, early detection anomalous situations feasible, increasing optimisation. last Rodríguez-Lera focus accountability robots. implies any robot charge logging its activities verifiable evidence so all actions traceable events triggering action identified. contributions study overview, modelling, formalization three perspectives autonomous robots; GUI tool graphically illustrate knowledge supported Conceptual Graphs; finally, proof concept summarizing dumped innovative debugging tools. findings research support idea verbose drives serious issues platform, detailed provided far explaining robot's behaviour. Besides, large volumes might managed properly specific situations; also, access sacrifices privacy human-robot interaction. could incorporate mechanisms help manufacturers, client/user developers know reasons trigger certain guest editors wish thank Prof. Jon G. Hall (Editor-in-Chief Wiley-Blackwell Journal Expert Systems: Knowledge Engineering) providing opportunity edit issue. Additional Lucia Rapanotti, her managing editing process. Finally, would also like referees have thoroughly evaluated editorial staff support. authors declare no potential conflict interest.