Learner engagement is a critical concept that can lead to satisfaction, motivation, and success in e-learning courses. It covers contextual, emotional, behavioral, cognitive, and social aspects. The instructors have difficulties identifying who is involved in the courses and the lack of face-to-face interaction with a learning resource to act upon and reduce the dropout rate. This paper presents a novel approach that aims to predict learner engagement in online courses and quantify the relationship between the learners’ success and their engagement. For this purpose, we used the traces gathered from 1 356 learners’ reactions in e-learning courses during the winters of 2020, 2021, and 2022, to implement this approach. To model the learning engagement, a variety of features were considered such as the total number of posts made in forums and the total time spent on the e-learning platform. This study uses the BiLSTM method with FastText word embedding to detect learners’ emotions from the forum discussions. Then, an unsupervised clustering technique based on the new dataset was used to cluster learners into groups according to their engagement level. Several supervised classification algorithms were trained and their performance was evaluated using cross-validation techniques and diverse precision metrics. The findings indicate that the decision tree rule model is more relevant than others, with an accuracy of 98% and an AUC score of 0.97. The conclusions of this research reveal that most learners are observers and that there is a nonlinear correlation between learning success and learning engagement.

In recent years, social media have been at the heart of new communication technologies. They are no longer used only to facilitate interaction between family, friends, and professional relationships, but tend to become an increasingly used communication channel to address public opinion. Research involving data sources from social media are relatively a recent and expanding area of research, nevertheless, the literature remains limited regarding the complex issue of how to assess and ensure the quality of social media data significantly and adequately. Our goal in this study is to provide a clearer and deeper understanding and a comprehensive overview of the existing state of research pertaining to the assessment of data quality in social media context. We performed a systematic literature review (SLR) on the quality of social media data to collect, analyze, and discuss data on the accuracy and value of prior literature that has focused on this area, has addressed a variety of topics, and has been published between 2016 and 2021. We followed a predefined review process to cover all relevant research papers published during this period. Our results demonstrate and strengthen the significance and the importance of data quality especially in the context of social media.

The learning management system (LMS) is an e-learning software that raised the interest of disparate learners’ groups. However, learners have difficulties in finding learning resources tailored to their preferences in the best way at the right time. Making the learning process more efficient and pleasant for learners can be achieved by using context and learning styles such as customizing aspects. This study proposes a new data-driven approach to retrieve learners' characteristics using traces of their activities based on the Felder-Silverman Learning Style Model (FSLSM). In this research, the traces of 714 learners who enrolled in three agronomy courses taught at IAV HASSAN II (winter session 2019, 2020, and 2021) were analyzed. Learners are categorized into clusters by their preference level for global/sequential learning styles, using an unsupervised clustering method. Then a classifier model tailored to our requirements was trained and based on the learner's learning style and their current context, a learning object recommendation list is proposed for them. The results revealed that the k-means algorithm performed well in identifying learning styles (LS) and the use of context features defined from the learners' adaptive close environments improved learning performance with an accuracy of over 96% given that most of the learners preferred a global learning style.

Defining and evaluating data quality can be a complex task as it varies depending on the specific purpose for which the data is intended. To effectively assess data quality, it is essential to take into account the intended use of the data and the specific requirements of the data users. It is important to recognize that a standardized approach to data quality assessment (DQA) may not be suitable in all cases, as different uses of data may have distinct quality criteria and considerations. In order to advance research in the field of data quality, it is useful to determine the current state of the art by identifying, evaluating, and analyzing relevant research conducted in recent years. In light of this objective, the study proposes a systematic literature review (SLR) as a suitable approach to examine the landscape of data quality and investigate available research specifically pertaining to DQA. The findings of our SLR clearly reveal and demonstrate the criticality of data quality and point to new directions for future study and have consequences for researchers and practitioners interested in defining and assessing data quality.

The CODATA Data Science Journal is a peer-reviewed, open access, electronic journal, publishing papers on the management, dissemination, use and reuse of research data and databases across all research domains, including science, technology, the humanities and the arts. The scope of the journal includes descriptions of data systems, their implementations and their publication, applications, infrastructures, software, legal, reproducibility and transparency issues, the availability and usability of complex datasets, and with a particular focus on the principles, policies and practices for open data. All data is in scope, whether born digital or converted from other sources.