In recent years, large language models (LLMs) and generative AI have revolutionized natural language processing (NLP), offering unprecedented capabilities in education. This chapter explores the transformative potential of LLMs in automated question generation and answer assessment. It begins by examining the mechanisms behind LLMs, emphasizing their ability to comprehend and generate human-like text. The chapter then discusses methodologies for creating diverse, contextually relevant questions, enhancing learning through tailored, adaptive strategies. Key prompting techniques, such as zero-shot and chain-ofthought prompting, are evaluated for their effectiveness in generating high-quality questions, including open-ended and multiple-choice formats in various languages. Advanced NLP methods like fine-tuning and prompt-tuning are explored for their role in generating task-specific questions, despite associated costs. The chapter also covers the human evaluation of generated questions, highlighting quality variations across different methods and areas for improvement. Furthermore, it delves into automated answer assessment, demonstrating how LLMs can accurately evaluate responses, provide constructive feedback, and identify nuanced understanding or misconceptions. Examples illustrate both successful assessments and areas needing improvement. The discussion underscores the potential of LLMs to replace costly, time-consuming human assessments when appropriately guided, showcasing their advanced understanding and reasoning capabilities in streamlining educational processes.

The functionality of a statistical framework for unified multivariate vital signs data fusion is seen in its flexibility for switching between the two data spaces. This paper develops data recovery methods for data fusion built on composite similarity measures using three distinct variable specific weighting methods - mean, median, and Orthogonalized Gnanadesikan-Kettenring (OGK). Using spatial covariance parameters derived from empirical patient data, the method successfully exhibits the ability to accurately recover eight different vital signs from its fused counterpart. Performance evaluation using Root Mean Squared Recovery Error (RMSRE), Root Mean Absolute Recovery Error (RMARE), and Root Standardized Mean Absolute Fusion Error (RSMARE) demonstrated that all three weighting approaches achieved comparable and highly accurate results, with parameters converging within ±0.01 of 1.0. The recovered signals closely matched the original data patterns, preserving both long-term trends and short-term fluctuations. Notably, the method proved computationally efficient and robust across different weighting approaches, suggesting broad applicability in clinical settings. This approach offers a promising solution for dimensionality reduction in complex physiological datasets while maintaining clinical relevance, with potential applications in patient monitoring systems and physiological modelling.

Risk analysis is a key process in project management, aiming to evaluate and mitigate risks by quantifying the probability and severity of potential events. A common tool for this purpose is the risk matrix, which offers a structured visualization of risks. Despite its effectiveness, constructing risk matrices traditionally depends on domain experts to define event probabilities and severities, leading to scalability and consistency challenges. This paper introduces RMGNN (Risk Matrix Generation with Graph Neural Networks), a semi-supervised and transductive method for automating risk matrix construction. RMGNN models historical events as graph structures, leveraging graph neural networks and language models to estimate event probabilities and severities. The method integrates labeled and unlabeled data, reducing reliance on manual annotation while maintaining accuracy. The main contributions of this paper include: (1) a graph-based framework for semi-supervised and transductive risk matrix learning, (2) a unified approach for estimating event probability and severity from graph representations, and (3) the use of graph topology to enhance interpretability in risk assessment. Experiments on real-world datasets demonstrate the effectiveness of RMGNN in addressing challenges associated with traditional risk matrix construction, supporting improved risk analysis and decision-making processes.

Soft magnetic composites cast with recycled ferrites and construction binders like cement, epoxy, or asphalt are highly robust and possess the lowest core up to 100 kHz in a wide temperature range. This paper shows how by using precision-enhanced magnetic testing and sophisticated hysteresis modeling, the performance of these materials could be significantly improved with respect to the stress effect on losses. A composite material grade with minimal losses was developed by benefiting from the behavior of ferrite compositions yielding zero crystalline anisotropy at temperatures below 0°C under compressive stress.

Power-centric research conducted in the past few decades has identified many influencing factors that affect and contribute to the development of high voltage surface flashover. Among them, the surface condition—morphology, roughness, and texture—is one such physical characteristic of gas-solid interfaces that is believed to be exploitable, relatively inexpensively, as a method to improve surface flashover strength. High voltage pulsed power systems and equipment face similar challenges, but there presently exists limited literature focused on impulse-driven flashover and the effects of surface roughness. In this work, the impulsive flashover strengths of five polymers relevant to pulsed power system design (PVC, Delrin, Ultem, Torlon, Perspex) are reported under two different (∼20 nanosecond and ∼100 microsecond risetime) impulse waveforms in atmospheric air. Samples of different surface conditions— “as received” and “machined”—were subjected to flashover tests, complemented with surface profilometry measurements to evaluate the effects of various roughness characteristics on the impulse-driven breakdown strengths and times-to-breakdown. The results obtained indicate a general enhancement of the impulsive flashover strength with increased roughness with a corresponding prolongation of the time-to-breakdown. Rougher “machined” surfaces were therefore found to outperform the smoother “as received” surfaces. A correlation analysis between the measured surface roughness parameters and the breakdown data suggests that the short-wavelength components of the surface profile contributes more towards the enhancement of the flashover strength compared to longer-wavelength “waviness” components. The consistency of this result with the theory of increased streamer path length and streamer inhibition is discussed, as are the potential consequences to insulator surface modification for flashover mitigation.

This paper addresses the modulation, design, and ZVS range extension of a single-stage current-fed fullbridge AC-DC converter with an active clamp circuit. A modulation scheme is proposed to achieve zero voltage turn-ON and zero current turn-OFF with reduced current stress as compared to the traditional phase-shifted modulation scheme. The impact of converter parameters, parasitic elements, and loading is considered while analyzing the switching transitions to understand the ZVS bounds of the converter. It is demonstrated that the ZVS range can be extended by a suitable choice of the dead time. The analysis is validated using switching circuit simulations that incorporate semiconductor models. Furthermore, experimental results from a 1.2 kW hardware prototype operating at 94.5 % efficiency at full load are presented to validate the findings.

This paper investigates the computation of the radar cross section (RCS) of equilateral triangular targets using Monte Carlo numerical integration. The analysis employs the Physical Theory of Diffraction (PTD) to model the scattering behavior and numerically evaluates the RCS for a canonical planar geometry. A Python-based implementation is provided for an equilateral triangle with side lengths of 2 meters, illuminated by a 10 GHz radar wave. The simulation yields an RCS value of approximately 0.0292 m², showcasing the effectiveness of the Monte Carlo approach for RCS estimation in simple geometries. This work contributes to the understanding of electromagnetic scattering and stealth design methodologies.

In the dynamic domain of optimization, by combining the advantages of several methods, hybrid algorithms have attracted a lot of interest. This task introduces and evaluates the Hybrid FOX-TSA algorithm, a new optimization technique that combines the Tree-Seed Algorithm's (TSA) exploitative power with the FOX algorithm's exploratory capabilities. The main objective is to evaluate the effectiveness and robustness of this hybrid approach across several CEC benchmark suites, which are widely documented as typical testbeds for optimization algorithms. The performance of the Hybrid FOX-TSA algorithm is accurately compared with some established optimization techniques, including Particle Swarm Optimization (PSO), Grey Wolf Optimizer (GWO), and the original FOX and TSA algorithms. A range of complex engineering design issues has been addressed in the comparison study, serving as real-world examples of optimization tasks. The results of experiments show that in terms of convergence speed and solution quality, the Hybrid FOX-TSA algorithm regularly performs better than other tested algorithms. The hybrid approach validates a remarkable capability to navigate complex search spaces, avoid premature convergence, and produce ideal or almost ideal results in a range of situations. Using the findings, it can be suggested that the Hybrid FOX-TSA algorithm is a robust and adaptable tool for attempting a wide range of optimization problems, especially those with complex landscapes and high dimensionality.

Reactive balance responses, which involve corrective and protective strategies, are highly dependent on rapid muscle activation to restore postural stability. Although electromyography (EMG) is commonly used to measure muscle activity, it has limitations such as signal interference, particularly during fast responses to external disturbances. Ultrasound imaging (US), in contrast, provides visualization of both superficial and deep muscles. Combining EMG and US imaging offers complementary insight into muscle behavior during reactive balance tasks. In this study, we investigated muscle activation and fascicle length changes in the medial gastrocnemius (MGS), lateral gastrocnemius (LGS), and soleus (SOL) muscles of the dominant (stepping) leg during stepping responses to unexpected low-amplitude (57.6 ± 5.8 N) and high-amplitude (123.4 ± 11.1 N) waist-pull perturbations in the anterior and posterior directions. Five young male adults (age: 25.2 ± 5.5 years) participated in the study. Results showed that perturbation amplitude significantly affected the EMG activation of both the MGS and SOL muscles in both directions, consistent with previous studies. Similarly, perturbation amplitude impacted fascicle length shortening in the LGS and SOL muscles. Significant differences in MGS and SOL activation were observed between high-amplitude and lowamplitude perturbations in both directions. Fascicle shortening in the LGS also differed significantly between perturbation amplitudes, whereas SOL fascicle shortening did not. By combining EMG and US imaging within the same participants, this study provides new insights into the neuromuscular mechanisms underlying balance control. These findings may inform the development of improved control strategies for neurorehabilitation devices and fall-prevention systems.

This paper presents the Modular Assistive Vertebra (MAV), a key component of the FleXo exoskeleton. FleXo is a passive, soft back-support exoskeleton designed to enhance ergonomic support by providing lifting support without compressing the user's vertebrae. This study systematically investigates the MAV's force transmission properties through theoretical and experimental methods. The results demonstrate the MAV's capability to convert linear forces exerted by a pulling cable into torque while minimizing the transmission of linear forces. Experimental setups with varying geometrical configurations validate the theoretical models, providing insights into the design and optimization of ergonomic support devices.

This paper presents a conformal aperture inverse design framework based on the inverse design method based on spectral analysis (IDMBSA) to address efficiency and applicability challenges in conformal aperture design. The framework initially employs IDMBSA to devise planar aperture field distributions, subsequently extending them to curved surfaces through full-wave simulations based on the equivalence principle and uniqueness theorem. Furthermore, we examine the influence of field truncation on the implementation of the equivalence principle. The results demonstrate that this framework effectively reduces the dimensionality of conformal array design problems, significantly improving efficiency for conformal aperture design. The feasibility of the method is validated through simulations, prototype fabrication, and testing for sparse conformal arrays. The innovation of this paper lies in combining IDMBSA with the equivalence principle, proposing a versatile and efficient conformal aperture inverse design framework. This approach not only handles various complex curved surfaces but also substantially enhances design efficiency, offering new solutions for conformal antenna applications in aerospace and other fields.

The exponential growth of global data traffic has underscored the limitations of traditional routing protocols such as BGP, OSPF, and MPLS, which rely on predefined policies and reactive measures. While BGP dynamically adapts to network changes, its convergence time and path selection mechanisms often lead to suboptimal routing, increased latency, and inefficient bandwidth utilization. As network infrastructures scale, these approaches struggle to adapt dynamically, resulting in suboptimal routing decisions, increased latency, and inefficient bandwidth utilization. This paper introduces AI-driven routing protocols that leverage machine learning (ML) and softwaredefined networking (SDN) to optimize real-time network traffic. I propose an autonomous routing framework integrating reinforcement learning-based path optimization, predictive congestion control, and self-healing network mechanisms to enhance network performance by up to 200%. My experimental simulations in Mininet and GNS3 demonstrate a 35% reduction in packet loss, 40% improvement in traffic predictability, and sub-second failure recovery. Additionally, I discuss the scalability of AI-enhanced routing in real-world network infrastructures, emphasizing its applicability in 5G/6G networks, cloud environments, and enterprise SDN deployments.

In this contribution, we formalize the general case of linguistic summaries with multiple summarizers and qualifiers. We extend the definitions present in the literature to handle cases in the following form: Q R1 ⋆. .. ⋆ R k y's are P1 ⋄. .. ⋄ P l where ⋆, ⋄ ∈ {AND, OR}. Moreover, we study the consistency of fuzzy summaries, focusing mainly on the double negation property (DN). We consider different negations for linguistic quantifiers and show which can be used for preserving the DN property.

The increasing prevalence of digital learning platforms and collaborative open-source software development has resulted in abundant data capturing learning behaviors. This paper investigates the intersection of process mining and learning analytics to understand how students and contributors engage in Massive Open Online Courses (MOOCs) and Free/Libre Open-Source Software (FLOSS) communities. Building on a diverse set of studies, we explore methods to extract, analyze, and interpret learning patterns from event logs, with a focus on bridging methodologies across these domains. We also incorporate perspectives from broader fields of data analytics and educational research to propose an integrated framework for optimizing engagement and retention in diverse online learning environments. By leveraging advanced process discovery, conformance checking, and machine learning techniques, this study offers novel insights into educational data mining and collaborative learning practices. Our findings have implications for the design of adaptive learning systems and the future of digital education.

The advent of sixth generation of cellular network technology (6G) technology fosters Healthcare 4.0, including advanced predictive diagnostics and elevated patient convenience. With the resource-constrained Internet of Things (IoTs), Electronic Health Records (EHR) are mandatory for Healthcare 4.0 patients. Transparency in transactions encourages the utility of decentralized architecture for attaining security and privacy in Healthcare 4.0 by using the IOTA Tangle. Limited IoTs demand storage efficiency while achieving security and privacy. The Healthcare 4.0 use case diminishes storage efficiency in IoTs as the conventional Merkle Tree implementation directs super-exponential growth. We exposition the growth rate of the Merkle tree for a use case adhering to Healthcare 4.0 and identify the dynamic growth on the left side with varying vertical levels. We introduce pruning on the dynamic side to balance the growing tail of the tree, presenting the dynamic balancing hybrid Merkle tree (DBHMT). We analyze its morphological and security characteristics and lay down closed-form expressions. As there is an optimal value between security and storage efficiency, we formulate the optimization problem for reducing the storage cost that is NP-Hard and non-convex in nature, so we apply the gradient descendant optimization algorithm by incorporating Adam's update mechanism to find the optimal pruning factor. We evaluate the behaviour of Merkle Tree, DBHMT without pruning and DBHMT in IOTA Tangle for 1 million transactions and found that DBHMT reduces the storage cost by the ratio of four in comparison with the conventional Merkle tree, and the trend continues for growing transactions. Finally, we discuss future research directions by examining prevailing problems.

Gastrointestinal (GI) tract cancers account for a substantial portion of the global cancer burden, where early diagnosis is critical for improved management and patient outcomes. The complex aetiologies and overlapping symptoms across GI cancers often delay diagnosis, leading to suboptimal treatment strategies. Cancer-related queries are crucial for timely diagnosis, treatment, and patient education, as access to accurate, comprehensive information can significantly influence outcomes. However, the complexity of cancer as a disease, combined with the vast amount of available data, makes it difficult for clinicians and patients to quickly find precise answers. To address these challenges, we leverage large language models (LLMs) such as GPT-3.5 Turbo to generate accurate, contextually relevant responses to cancer-related queries. Pre-trained with medical data, these models provide timely, actionable insights that support informed decision-making in cancer diagnosis and care, ultimately improving patient outcomes. We calculate two metrics: A1 (which represents the fraction of entities present in the model-generated answer compared to the gold standard) and A2 (which represents the linguistic correctness and meaningfulness of the model-generated answer with respect to the gold standard), achieving maximum values of 0.546 and 0.881, respectively.

Within the color space of 3-coloring, we build deterministic algorithms for plane graphs with no more than 4 to find base configurations that appear in the n-DJC group, and thus find new triplets bringing 3-inseparable vertices. This paper provides the basic theory of the n-DJC group, including the removable conflict group between DJC and the rotation of DJC in the group. We can identify 3-uncolorable vertices according to unreasonable removable conflicts or frozen rotations. And we point out that such n-DJC groups can appear in the color space of planar graph with degree no more than 4.

Building effective semantic consistency feature space is important in vision-language alignment learning. To do this cross-modal task, previous works train coarse-grained internetscale data and design complicated embedding feature-cross networks. These methods often logically ignore bridging the finegrained relevance between vision and language. In this paper, we propose a structured semantic representation that contains fine-grained semantic alignment information. This information consists of four semantic elements: Time, Location, Object, and Object State. Semantic elements in each video frame can be organized into a hierarchical graph. Furthermore, this paper introduces the semantic element consistency learning network (SECNet), which facilitates fine-grained semantic alignment for vision-languagerelated tasks. SECNet consists of several semantic element consistency modules and a video-text alignment module. The former learns fine-grained semantic element consistency across video frames. The latter learns coarse-grained alignment from video-text pairs. To train our network, we build a video-text dataset VSET. VSET contains fine-grained semantic alignment information and question-answer(QA) pairs based on DiDeMo. Specifically, there is 594k corresponding fine-grained semantic alignment information, which consists of 4k semantic elements, and 148k QA pairs. Our experiments demonstrate that SEC-Net achieves better performance than the other state-of-the-art methods. Code and the sample data of VSET can be seen in the supplementary file.