The increasing utilization of surveillance cameras in smart cities, coupled with the surge of online video applications, has heightened concerns regarding public security and privacy protection, which propelled automated Video Anomaly Detection (VAD) into a fundamental research task within the Artificial Intelligence (AI) community. With the advancements in deep learning and edge computing, VAD has made significant progress and advances synergized with emerging applications in smart cities and video internet, which has moved beyond the conventional research scope of algorithm engineering to deployable Networking Systems for VAD (NSVAD), a practical hotspot for intersection exploration in the AI, IoVT, and computing fields. In this article, we delineate the foundational assumptions, learning frameworks, and applicable scenarios of various deep learning-driven VAD routes, offering an exhaustive tutorial for novices in NSVAD. In addition, this article elucidates core concepts by reviewing recent advances and typical solutions and aggregating available research resources accessible at https://github.com/fdjingliu/NSVAD . Last, this article projects future development trends and discusses how the integration of AI and computing technologies can address existing research challenges and promote open opportunities, serving as an insightful guide for prospective researchers and engineers.

Soft sensors play a vital role in industrial process monitoring and control by estimating difficult-to-measure quality variables. While significant progress has been made in improving the accuracy of soft sensor models, challenges remain in ensuring their interpretability and stability in dynamic industrial environments. From a measurement science perspective, ensuring transparency and reliable performance under varying process conditions is becoming increasingly critical, particularly in high-stakes industrial applications. This paper provides a comprehensive review of methodologies to enhance the interpretability and stability of soft sensor models. To address interpretability, we analyze various interpretable machine learning techniques applicable to soft sensors and discuss open-source projects that facilitate the implementation of these techniques. For improving stability, we emphasize the role of causal machine learning, detailing methods for causal discovery in industrial processes and highlighting relevant open-source tools. By highlighting current limitations and identifying areas for improvement, we aim to provide valuable insights and practical tools for researchers and practitioners. These insights will guide the development of more transparent and reliable soft sensors, ultimately enhancing industrial process monitoring and control.

Aiming at the problem that the vibration signal characteristics of rolling bearings are not obvious and the diagnosis accuracy of traditional intelligent fault diagnosis algorithms is not high, a rolling bearing fault diagnosis method based on VMD feature energy reconstruction and ADE-ELM is proposed. Firstly, the VMD is used to process the bearing vibration signal to obtain the intrinsic mode function. Then, each IMF component is calculated, and the fault feature vector is reconstructed based on the feature energy ratio. Finally, the adaptive differential evolution (ADE) algorithm is introduced to optimize the hidden layer input weights and biases of the extreme learning machine. The simulation shows that the VMD feature energy reconstruction can accurately extract fault features, and the ADE-ELM model can effectively improve the fault classification accuracy. Compared with other methods, this method is more accurate and reliable.