Abstract In recent decades, fog computing has contributed significantly to the expansion of smart cities. It generated numerous real‐time data and coped with time‐constraint applications. They use sensors, physical objects, and network standards to monitor health imaging, traffic surveillance, industrial management, and so forth. Interactive applications have been proposed for the Internet of Things (IoT) to control wireless channels and improve communication. However, most of the existing lack of handing network interference and a reliable monitoring process. Moreover, many solutions are vulnerable to external threats, resulting in inconsistent and untrustworthy information for end users. Thus, this article proposes a framework that considers possible shortest paths to provide the most reliable and low‐latency healthcare decision system using Q‐learning. In addition, fog devices offer a trusted transmission interference system and are kept secure. The proposed framework is specially designed for rapid real‐time medical data processing while enforcing robust security throughout the IoT‐based transmission process. To identify the health sensors in pairwise objects with the initial computing cost, the proposed framework applies graph theory. It also extracts the most effective and least loaded communication edges by examining the behaviour of devices. Moreover, the identities of devices are verified using lightweight timestamps and secret information, accordingly, it decreases the privacy threats.

ABSTRACT Digital textbooks (DTs) have evolved from DT 1.0, which simply converted paper textbooks to PDF format, to DT 2.0, which provides various multimedia content, for example, video and audio content. DTs have now advanced to DT 3.0, which enhances learner engagement through gamification and simulations. Recently, with the advancement of cloud computing technology and digital devices, for example, tablets, DT 4.0, which supports personalised learning through artificial intelligence (AI) tutors and chatbots, has been realised. South Korea is actively implementing a policy to distribute artificial intelligence–based DTs, equivalent to DT 4.0, to all schools under national leadership. For artificial intelligence–based DTs (AIDTs) in South Korea to develop into a sustainable education system, reliance on cloud computing alone is insufficient. It is also necessary to build layers of fog computing and edge computing from the initial stage. There are concerns that AIDTs may exacerbate the learning gap because they are more likely to be utilised actively by high‐performing students with established self‐directed learning habits rather than struggling students. Thus, it is essential to enhance usage monitoring and explore strategies that provide educational scaffolding to prevent differences in the level of AIDT utilisation from leading to a widening learning gap.

Abstract Traditional healthcare systems have suffered from different data communication, security, data processing, and compliance issues. The traditional systems are also not well equipped to handle the new technologies like Artificial Intelligence (AI) by enabling more accurate diagnostics, personalized treatment plans, and improved patient outcomes. The existing data communication and security protocols and compliance are also not fully implemented to tackle the system's challenges. This article proposes a Tri‐Tier architecture by using data communication, AI data generative, and regulation and compliance tiers. The data communication tier is based on advanced sensing and monitoring technologies like cloud and edge‐based systems integrated with security detection mechanisms. The edge and cloud layer provides the all functions of the perception layer like smart sensing, visual sensing, and monitoring services, and can control the device's perception and behaviour. The second tier provides the AI data generative functionalities to handle real‐time synthetic medical images for predictive analytics to enhance patient care. This tier also automates routine tasks, such as administrative work and data analysis, which can free up healthcare professionals to focus on more complex tasks. The last regulation and compliance tier is responsible for handling the standards and compliance for healthcare systems. Experiments are conducted to test the data communication and security level of the proposed architecture. The results showed the suitability of existing solutions and synchronization with the proposed architecture.

As the performance of generative artificial intelligence (GAI), such as ChatGPT, improves, content created by GAI will be distributed in the social media space, and knowledge and writings from unknown sources will be disseminated and reproduced. Now that GAI is becoming widespread, it is necessary to distinguish GAI from human intelligence, which constitutes knowledge. The data, information, knowledge, and work (DIKW) hierarchy is a useful framework for teaching and for checking metacognitive and explainable artificial intelligence (XAI) literacy. There are two types of collaboration between GAI and human intelligence: a combined intelligence model and a parallel intelligence model. The combined intelligence model is a method of using GAI for creating works by collecting data, organizing information, and deriving knowledge from information. This model is suitable for GAI-assisted tasks (GAIATs). The parallel intelligence model is suitable for GAI-assisted learning (GAIAL); it is a method in which a person develops abilities by analyzing and comparing tasks created by GAI after going through the data-information-knowledge-work process. The zone of proximal development (ZPD) created by educational scaffolding is a quantitative framework that is appropriate for evaluating the effects of GAI. The ZPD generated by GAI that corresponds to scaffolding should be managed so as not to favor or disadvantage specific individuals.

Summary Popular songs have evolved from vinyl records (voice files) to music videos. Sound and video can be mixed and presented in such a manner that only sound is heard. Scenes from music programs require continuous editing to maximize audience immersion. Music program scene editing is developed through the expertise of a television program director (PD). The PD considers detailed editing methods by the analyzing lyrics and the singer's choreography to express a song in a rhythmic video. This study presents an algorithm for music program scene editing that determines the size of a shot, evaluating the psychological distance between singer and viewers after measuring the pitch contour of a song.

Abstract This study proposes a method for analyzing the level of student engagement in a non‐face‐to‐face learning situation by processing facial expressions. In previous studies, the level of learning engagement was determined through biosignals, such as galvanic skin response feedback. The method proposed in this study assesses engagement levels in an individual education situation by processing emotions revealed through a student's facial expressions in real time. The emotional state revealed through facial expressions can be a useful indicator for assessing the level of immersion in learning. The algorithm proposed in this study is effective because teachers can teach while simultaneously monitoring students' engagement.

Abstract In recent years, deep convolutional neural networks (CNNs) have achieved a lot of outstanding results in super‐resolution with superior ability. However, the majority of CNNs only use a series of convolution kernels with the same size to extract features. This will cause limited receptive fields. In this work, we propose a parallel convolution attention network (PCAN) to extract features in an effective way. Specifically, a pair of parallel convolutions (PCs) with different kernel sizes is used in one layer in our network, which can extract features within different receptive fields, thereby making full use of the multiscale information. Meanwhile, we apply a channel‐spatial attention (CSA) module in each parallel convolution block to calculate and fuse channel attention and spatial attention. The obtained attention maps emphasize useful features. Experimental results demonstrate the superiority of our PCAN in comparison with the state‐of‐the‐art methods.

Virtual reality drama has the benefit of enhancing immersion, which was lacking in original e-Learning systems. Moreover, dangerous and expensive educational content can be replaced by stimulating users’ interest. In this study, we investigate the effects of avatar performance in virtual reality drama. The hypothesis that the psychical distance between virtual characters and their viewers changes according to the size of video shots is tested with an autonomic nervous system function test. Eighty-four college students were randomly assigned to three groups. Virtual reality drama is used to train teachers concerning school bullying prevention, and deals with the dialogue between teachers and students. Group 1 was provided with full-shot video clips, Group 2 was shown various clips from full shots to extreme close-ups, and Group 3 was provided with close-up shots. We found that the virtual reality drama viewers’ levels of stimulation changed in relation to the size of the shots. The R-R (between P wave and P wave) intervals of the electrocardiograms (ECGs, bio-signal feedback) became significantly narrower as the shot size became smaller.