This study examines the research landscape of smart learning environments by conducting a comprehensive bibliometric analysis of the field over the years. The study focused on the research trends, scholar’s productivity, and thematic focus of scientific publications in the field of smart learning environments. A total of 1081 data consisting of peer-reviewed articles were retrieved from the Scopus database. A bibliometric approach was applied to analyse the data for a comprehensive overview of the trend, thematic focus, and scientific production in the field of smart learning environments. The result from this bibliometric analysis indicates that the first paper on smart learning environments was published in 2002; implying the beginning of the field. Among other sources, “Computers & Education,” “Smart Learning Environments,” and “Computers in Human Behaviour” are the most relevant outlets publishing articles associated with smart learning environments. The work of Kinshuk et al., published in 2016, stands out as the most cited work among the analysed documents. The United States has the highest number of scientific productions and remained the most relevant country in the smart learning environment field. Besides, the results also showed names of prolific scholars and most relevant institutions in the field. Keywords such as “learning analytics,” “adaptive learning,” “personalized learning,” “blockchain,” and “deep learning” remain the trending keywords. Furthermore, thematic analysis shows that “digital storytelling” and its associated components such as “virtual reality,” “critical thinking,” and “serious games” are the emerging themes of the smart learning environments but need to be further developed to establish more ties with “smart learning”. The study provides useful contribution to the field by clearly presenting a comprehensive overview and research hotspots, thematic focus, and future direction of the field. These findings can guide scholars, especially the young ones in field of smart learning environments in defining their research focus and what aspect of smart leaning can be explored.

The increasing attention to Machine Learning (ML) in K-12 levels and studies exploring a different aspect of research on K-12 ML has necessitated the need to synthesize this existing research. This study systematically reviewed how research on ML teaching and learning in K-12 has fared, including the current area of focus, and the gaps that need to be addressed in the literature in future studies. We reviewed 43 conference and journal articles to analyze specific focus areas of ML learning and teaching in K-12 from four perspectives as derived from the data: curriculum development, technology development, pedagogical development, and teacher training/professional development. The findings of our study reveal that (a) additional ML resources are needed for kindergarten to middle school and informal settings, (b) further studies need to be conducted on how ML can be integrated into subject domains other than computing, (c) most of the studies focus on pedagogical development with a dearth of teacher professional development programs, and (d) more evidence of societal and ethical implications of ML should be considered in future research. While this study recognizes the present gaps and direction for future research, these findings provide insight for educators, practitioners, instructional designers, and researchers into K-12 ML research trends to advance the quality of the emerging field.

Imbalanced Data (ID) is a problem that deters Machine Learning (ML) models from achieving satisfactory results. ID is the occurrence of a situation where the quantity of the samples belonging to one class outnumbers that of the other by a wide margin, making such models’ learning process biased towards the majority class. In recent years, to address this issue, several solutions have been put forward, which opt for either synthetically generating new data for the minority class or reducing the number of majority classes to balance the data. Hence, in this paper, we investigate the effectiveness of methods based on Deep Neural Networks (DNNs) and Convolutional Neural Networks (CNNs) mixed with a variety of well-known imbalanced data solutions meaning oversampling and undersampling. Then, we propose a CNN-based model in combination with SMOTE to effectively handle imbalanced data. To evaluate our methods, we have used KEEL, breast cancer, and Z-Alizadeh Sani datasets. In order to achieve reliable results, we conducted our experiments 100 times with randomly shuffled data distributions. The classification results demonstrate that the mixed Synthetic Minority Oversampling Technique (SMOTE)-Normalization-CNN outperforms different methodologies achieving 99.08% accuracy on the 24 imbalanced datasets. Therefore, the proposed mixed model can be applied to imbalanced binary classification problems on other real datasets.

Virtual Reality (VR) and educational games are emerging technologies mediating a rapid transformation in the educational world. However, few studies have systematically analyzed Educational Virtual Reality Games (EVRGs) and how they have been applied in educational settings. This study reviewed 31 articles published in high impact journals and educational conference proceedings to unravel thetechnological, pedagogical, and gaming characteristics of contemporaryEVRGs. The results show the predominance of Oculus Rift headsets and HTC Vive as the main technology used in EVRGs. Moreover, the analysis revealed that the pedagogical application of the majority of EVRGs was developed for all levels of education (e.g. tertiary, K-12, lifelong learning), with the specific target audience of each game based on the desired learning outcome. Furthermore, the application of EVRGs has primarily focused on out of class use, with healthcare education topics dominating the topics taught using EVRGs. Based on our findings, we highlight some key implications and suggestions to advance the field of EVRGs.

The field of drug discovery has experienced a remarkable transformation with the advent of artificial intelligence (AI) and machine learning (ML) technologies. However, as these AI and ML models are becoming more complex, there is a growing need for transparency and interpretability of the models. Explainable Artificial Intelligence (XAI) is a novel approach that addresses this issue and provides a more interpretable understanding of the predictions made by machine learning models. In recent years, there has been an increasing interest in the application of XAI techniques to drug discovery. This review article provides a comprehensive overview of the current state-of-the-art in XAI for drug discovery, including various XAI methods, their application in drug discovery, and the challenges and limitations of XAI techniques in drug discovery. The article also covers the application of XAI in drug discovery, including target identification, compound design, and toxicity prediction. Furthermore, the article suggests potential future research directions for the application of XAI in drug discovery. This review article aims to provide a comprehensive understanding of the current state of XAI in drug discovery and its potential to transform the field.