In the landscape of higher education, a significant portion of student learning occurs within digital environments, facilitated by interconnected networks such as Learning Management Systems (LMS), Massive Open Online Courses (MOOCs), and various online platforms. This digital landscape generates a vast repository of educational data, offering valuable insights for educational stakeholders. However, evaluating student engagement in online learning presents a critical challenge. This study addresses this challenge by focusing on the evaluation of student engagement using Learning Analytics (LA). We introduce DDS-Eng, a Data-Driven Decision Support System designed to analyze students’ engagement. Our approach involves thorough requirement analysis to define decision-makers’ needs and selected data sources, followed by the development of a multidimensional model capturing engagement aspects and associated metrics. We then implement a dedicated interface utilizing Online Analytical Processing (OLAP) queries and a Learning Analytics Dashboard (LAD) to analyze individual student behavior and overall student engagement comprehensively. Finally, we present an empirical evaluation of our LAD conducted in a controlled environment. This research underscores the importance of student engagement in online learning and emphasizes the need for innovative approaches to evaluate learner engagement in multimodal and informal learning environments.
Crowdfunding has emerged as a vital alternative funding source, transforming how creative projects and startups secure financing by directly connecting creators to backers. However, persistent trust issues and information asymmetry between creators and backers significantly hinder its growth and development. Existing trust-enhancement mechanisms, such as third-party endorsements and basic expert validation often lack objectivity and robustness, leaving backers vulnerable to biased signals and project failures. This paper addresses these limitations by introducing a novel trust-enhancement mechanism, referred to as Double-Score Voting. This approach refines expert validation systems by integrating two critical dimensions: firstly, a granular score-based vote from experts on a project's potential, moving beyond simple binary approval; and secondly, a weighted score representing the expert's credibility and level of expertise. This dual-layered evaluation provides a more nuanced, objective, and reliable assessment of project viability. The mechanism is formalised mathematically, and its practical implementation is demonstrated through CertiFund, a prototype crowdfunding platform developed to test and validate the concept. The findings of this study demonstrate that the Double-Score Voting mechanism can significantly mitigate information asymmetry, thereby increasing the credibility of projects and fostering a more trustworthy ecosystem for both creators and backers.