No abstract

As the Industrial Internet of Things (IIoT) develops rapidly, cloud computing and fog computing become effective measures to solve some problems, e.g., limited computing resources and increased network latency. The Industrial Control Systems (ICS) play a key factor within the development of IIoT, whose security affects the whole IIoT. ICS involves many aspects, like water supply systems and electric utilities, which are closely related to people’s lives. ICS is connected to the Internet and exposed in the cyberspace instead of isolating with the outside recent years. The risk of being attacked increases as a result. In order to protect these assets, intrusion detection systems (IDS) have drawn much attention. As one kind of intrusion detection, anomaly detection provides the ability to detect unknown attacks compared with signature-based techniques, which are another kind of IDS. In this paper, an anomaly detection method with a composite autoencoder model learning the normal pattern is proposed. Unlike the common autoencoder neural network that predicts or reconstructs data separately, our model makes prediction and reconstruction on input data at the same time, which overcomes the shortcoming of using each one alone. With the error obtained by the model, a change ratio is put forward to locate the most suspicious devices that may be under attack. In the last part, we verify the performance of our method by conducting experiments on the SWaT dataset. The results show that the proposed method exhibits improved performance with 88.5% recall and 87.0% F1-score.

To assess the influence of education on the performance of Chinese version of Montreal cognitive assessment (C-MoCA) in relation to the mini-mental state examination (MMSE) in detecting amnesic mild cognitive impairment (aMCI) among rural-dwelling older people C-MoCA and MMSE was administered and diagnostic interviews were conducted among community-dwelling elderly in two villages in Beijing. The performance of C-MoCA and MMSE in detecting aMCI was evaluated by the area under the ROC curve (AUC). Effect size of education on variations in C-MoCA scores was estimated with general linear model. Among 172 study participants (24 cases of aMCI and 148 normal controls), the AUC of C-MoCA was 0.72 (95% CI = 0.62–0.81, cutoff = 20/21), compared to AUC of MMSE of 0.74 (95% CI = 0.64–0.84, cutoff = 26/27). The performance of both C-MoCA and MMSE was especially poorer among those with low (0–6 years) education. After controlling for gender and age, education (η 2 = 0.204) had a surpassing effect over aMCI diagnosis (η 2 = 0.052) on variations in C-MoCA scores. Among rural older people, the MoCA showed modest accuracy and was no better than MMSE in detecting aMCI, especially in those with low education, due to the overwhelming effect of education relative to aMCI diagnosis on variations in C-MoCA performance.