Teachers are instrumental in advancing the cognitive and motor skills of children with autism. Despite their importance, the incorporation of both educators and robotic aids in the educational frameworks of specialized schools and centers is infrequent. Extensive research has been conducted to evaluate the impact of robotic assistance on the learning outcomes for children with autism. This study investigates the effects of the Furhat robot on the educational experiences of autistic children in schools, analyzing its utility both with and without the presence of teachers. Interviews with educators were carried out to gauge the effectiveness of implementing Furhat robots in these settings. Data collected from sessions with autistic children were analyzed using ANOVA tests, offering insights into the Furhat Social Robot’s potential as a significant tool for fostering engagement and interaction. The findings highlight the robot’s effectiveness in enhancing social interaction and engagement, thereby contributing to the ongoing discussion on how social robots can improve the developmental progress and well-being of children with autism. Moreover, this paper underlines the innovative aspects of our proposed model and its wider implications. By presenting specific quantitative outcomes, our aim is to extend the reach of our findings to a broader audience. Ultimately, this research delineates significant contributions to the understanding of social robots, such as Furhat, in improving the overall well-being and developmental trajectories of children with autism.

In learning, one of the fundamental motivating factors is self-efficacy. Therefore, it is crucial to understand the level of students’ self-efficacy in learning programming. This article presents a quantitative study on undergraduate students’ perceived programming self-efficacy. 110 undergraduate computing students took part in this survey to assess programming self-efficacy. Before being given to the respondents, the survey instrument, which included a 28-item self-efficacy assessment and 30 multiple-choice programming questions, was pilot-tested. The survey instrument had a reliability of 0.755. The study results show that the students’ self-efficacy was low when they solved complex programming tasks independently. However, they felt confident when there was an assistant to guide them through the tasks. From this study, it could be concluded that self-efficacy is an essential achievement component in programming courses and can avoid education dropouts.

E-learning has become an integral part of higher education, significantly influencing the teaching and learning landscape. This study investigates the impact of student characteristics such as gender, grade, and major on E-learning satisfaction. Utilizing Structural Equation Modeling (SEM) and collecting data through 527 valid questionnaires from Nanjing Normal University students, this research reveals the nuanced relationships between these variables and E-learning satisfaction. The findings indicate that gender, grade, and major significantly and positively impact student satisfaction with E-learning, highlighting the need for tailored E-learning resources to meet diverse student needs. The study underscores the importance of continuous improvement in E-learning resources and platforms to enhance student satisfaction. This research contributes to the understanding of effective E-learning strategies in higher education institutions.

In this study, we consider the extended Brinkman's-Darcy model for a triple diffusive convection system which consists of some parameters such as Taylor number (Ta), Solutal Rayleigh numbers (RC1 , RC2 ), and Prandtl number (Pr). To investigate the range of these parameters, a dynamical system of the Ginzburg-Landau equation is developed. The parametric analysis and comparative study of the model for the three Rayleigh numbers which leads to the clear fluid layer, sparsely packed porous layer, and densely packed porous layer is done with the help of bifurcation maps and the Lyapunov exponents. It is found that for a certain range of parameters, the system exhibits a chaotic behaviour.

Accurate drug-drug interaction (DDI) prediction is essential to prevent adverse effects, especially with the increased use of multiple medications during the COVID-19 pandemic. Traditional machine learning methods often miss the complex relationships necessary for effective DDI prediction. This study introduces a deep learning-based classification framework to assess adverse effects from interactions between Fluvoxamine and Curcumin. Our model integrates a wide range of drug-related data (e.g., molecular structures, targets, side effects) and synthesizes them into high-level features through a specialized deep neural network (DNN). This approach significantly outperforms traditional classifiers in accuracy, precision, recall, and F1-score. Additionally, our framework enables real-time DDI monitoring, which is particularly valuable in COVID-19 patient care. The model’s success in accurately predicting adverse effects demonstrates the potential of deep learning to enhance drug safety and support personalized medicine, paving the way for safer, data-driven treatment strategies.

The purpose of Vehicular Ad Hoc Network (VANET) is to provide users with better information services through effective communication. For this purpose, IEEE 802.11p proposes a protocol standard based on enhanced distributed channel access (EDCA) contention. In this standard, the backoff algorithm randomly adopts a lower bound of the contention window (CW) that is always fixed at zero. The problem that arises is that in severe network congestion, the backoff process will choose a smaller value to start backoff, thereby increasing conflicts and congestion. The objective of this paper is to solve this unbalanced backoff interval problem in saturation vehicles and this paper proposes a method that is a deep neural network Q-learning-based channel access algorithm (DQL-CSCA), which adjusts backoff with a deep neural network Q-learning algorithm according to vehicle density. Network simulation is conducted using NS3, the proposed algorithm is compared with the CSCA algorithm. The find is that DQL-CSCA can better reduce EDCA collisions.