Search

Students’ acquisition of generic skills is a key to their incorporation into the job world. However, teachers encounter several difficulties when measuring their students’ performance in generic skills. These difficulties increase in online courses based on learning management systems, where there is no direct contact between teachers and their students. To address this problem, this paper adopts a solution based on the use of activity records from learning management systems as indicators of the students’ performance in generic skills. After conducting a survey among teachers of online courses on the suitability of the solution, an artefact was developed and applied within four online courses. Interviews conducted with teachers and the results of a survey for mapping skills and activities provide positive evidence of this solution’s potential.

In this paper, the term "applicability domain" refers to the range of chemical compounds for which the statistical quantitative structure-activity relationship (QSAR) model can accurately predict their toxicity. This is a crucial concept in the development and practical use of these models. First, a multidisciplinary review is provided regarding the theory and practice of applicability domains in the context of toxicity problems using the classical QSAR model. Then, the advantages and improved performance of neural networks (NNs), which are the most promising machine learning algorithms, are reviewed. Within the domain of medicinal chemistry, nine different methods using NNs for toxicity prediction were compared utilizing 29 alternative artificial intelligence (AI) techniques. Similarly, seven NN-based toxicity prediction methodologies were compared to six other AI techniques within the realm of food safety, 11 NN-based methodologies were compared to 16 different AI approaches in the environmental sciences category and four specific NN-based toxicity prediction methodologies were compared to nine alternative AI techniques in the field of industrial hygiene. Within the reviewed approaches, given known toxic compound descriptors and behaviors, we observed a difficulty in being able to extrapolate and predict the effects with untested chemical compounds. Different methods can be used for unsupervised clustering, such as distance-based approaches and consensus-based decision methods. Additionally, the importance of model validation has been highlighted within a regulatory context according to the Organization for Economic Co-operation and Development (OECD) principles, to predict the toxicity of potential new drugs in medicinal chemistry, to determine the limits of detection for harmful substances in food to predict the toxicity limits of chemicals in the environment, and to predict the exposure limits to harmful substances in the workplace. Despite its importance, a thorough application of toxicity models is still restricted in the field of medicinal chemistry and is virtually overlooked in other scientific domains. Consequently, only a small proportion of the toxicity studies conducted in medicinal chemistry consider the applicability domain in their mathematical models, thereby limiting their predictive power to untested drugs. Conversely, the applicability of these models is crucial; however, this has not been sufficiently assessed in toxicity prediction or in other related areas such as food science, environmental science, and industrial hygiene. Thus, this review sheds light on the prevalent use of Neural Networks in toxicity prediction, thereby serving as a valuable resource for researchers and practitioners across these multifaceted domains that could be extended to other fields in future research.

Evaluating web traffic on a web server is highly critical for web service providers since, without a proper demand forecast, customers could have lengthy waiting times and abandon that website. However, this is a challenging task since it requires making reliable predictions based on the arbitrary nature of human behavior. We introduce an architecture that collects source data and in a supervised way performs the forecasting of the time series of the page views. Based on the Wikipedia page views dataset proposed in a competition by Kaggle in 2017, we created an updated version of it for the years 2018–2020. This dataset is processed and the features and hidden patterns in data are obtained for later designing an advanced version of a recurrent neural network called Long Short-Term Memory. This AI model is distributed training, according to the paradigm called data parallelism and using the Downpour training strategy. Predictions made for the seven dominant languages in the dataset are accurate with loss function and measurement error in reasonable ranges. Despite the fact that the analyzed time series have fairly bad patterns of seasonality and trend, the predictions have been quite good, evidencing that an analysis of the hidden patterns and the features extraction before the design of the AI model enhances the model accuracy. In addition, the improvement of the accuracy of the model with the distributed training is remarkable. Since the task of predicting web traffic in as precise quantities as possible requires large datasets, we designed a forecasting system to be accurate despite having limited data in the dataset. We tested the proposed model on the new Wikipedia page views dataset we created and obtained a highly accurate prediction; actually, the mean absolute error of predictions regarding the original one on average is below 30. This represents a significant step forward in the field of time series prediction for web traffic forecasting.

This special issue, Special Issue on Multisensor user tracking and analytics to improve education and other application fields, concentrates on the practical and experimental use of data mining and analytics techniques, specially focusing on the educational area. The selected papers deal with the most relevant issues in the field, such as the integration of data from different sources, the identification of data suitable for the problem analysis, and the validation of the analytics techniques as support in the decision making process. The application fields of the analytics techniques presented in this paper have a clear focus on the educational area (where Learning Analytics has emerged as a buzzword in the recent years) but not restricted to it. The result is a collection of use cases, experimental validations and analytics systems with a clear contribution to the state of the art.

Self-regulated learning (SRL) is a cyclical process through which individuals plan their objectives, execute them and self-evaluate their own behavior so as to obtain their proposed goals. SRL has been investigated by different authors such as Zimmerman, Boekaerts, Winne and Hadwin, Pintrich, Efklides and Hadwin, Järvelä and Miller and it's being applied in learning environments. This systematic review describes the current state of the art in terms of the support for SRL in Massive Online Open courses (MOOC) using technologies based on psychological models. 66 studies conducted between 2010 and 2020 were analyzed by searching three multidisciplinary databases: Scopus, Web of Science and Google Scholar. The review methodology steps were the review planning, the search, literature analysis and the results report. Results show SRL in MOOCs is an emerging study area incentivized by the high dropout rate of the participants in MOOC. Regarding models of SRL, the most representative author reported was identified as Zimmerman. The most prominent self-regulation strategies used by MOOCs participants are: Goal setting, Help Seeking, Time management, Self-evaluation and Strategic planning. The platforms with research on SRL in MOOCs that stand out are Coursera, Edx, Open Edx and Moodle. We identified tools which have been developed to support SRL in MOOC and a set of good practices useful to support SLR that can be used by MOOC designers and tutors. Finally, a series of open problems and challenges that could lead to new research on the topic of SRL in MOOCs have been identified.

This paper presents a desk research that analysed available recent studies in the field of Technology Enhanced Learning. The desk research is focused on work produced in the frame of FP6 and FP7 European programs, in the area of Information and Communication Technologies. It concentrates in technologies that support existing forms of learning, and also in technologies that enhance new learning paradigms. This approach includes already adopted and successfully piloted technologies. The elaboration of the desk research had three main parts: firstly, the collection of documents from CORDIS and other institutions related to TEL research; secondly, the identification of relevant terms appearing in those documents and the elaboration of a thesaurus; and thirdly, a quantitative analysis of each term occurrences. Many of the identified technologies belong to the fields of interactive multimedia, Human-computer Interaction and-or related to recommendation and learning analytics. This study becomes a thorough review of the current state of these fields through the actual development of R&D European projects. This research, will be used as a basis to better understand the evolution of the sector, and to focus future research efforts on these sectors and their application to education.