Using Emotional Learning Analytics to Improve Students’ Engagement in Online Learning

• DOI: 10.14742/apubs.2022.129
• Publication Date: 2022-11-23T04:37:32Z
• Authors: Chaitali Samani; Amara Atif; Kaska Musial-Gabrys
• Keywords: Boredom; Surprise; Learning Analytics; Curiosity; Leverage (statistics)

Online learning is enabled by technology as students’ interact with subject content, access and download resources, watch videos, participate in online quizzes. Epistemic Emotions (EE), e.g., curiosity, surprise, confusion, frustration or boredom has a complex impact when due to the lack of real-time feedback loop. It becomes vital detect such emotions timely accurately before they starts impacting leaner adversely (e.g., D’Mello et al., 2017; Kosasi 2020). EE activate confines learning. can be positive (surprise, enjoyment) negative (frustration, boredom), two folded (confusion). Positive contribute achievement, while cause anxiety, leading impaired are ambiguous dynamic, without innovations analytics (LA) Educational Data Mining, it challenging predict (D’Mello, S., 2017). This extended abstract focuses on how LA help surprise confusion learning, ultimately accomplishing achievement emotion(s) learners. provides tools techniques measure, collect, analyse, report interaction data students. Deeper analysis all kinds digital traces collected platforms potential emotional cognitive states (Karaoglan & Yilmez 2022; Han et. 2021; Silvola 2021). Innovations leverage voluminous heterogeneous available propose adopting data-driven analytical approaches study engagement cognition. like helps learners discover new knowledge, improve memory focus cognition (Foster Keane, 2019). adds "novelty factor" nourish students' curiosity motivation (Hayden 2011; Roesch 2012). Whilst, Confusion causes disequilibrium incongruent information (Arguel When students overcome gains, but prolonged leave student frustrated bored, impeding their (Atapattu 2019; Baker 2010). The activity log valuable non-intrusive source detecting important epistemic identify difficulties. Such promotes unbiased assessment technically easy deploy simultaneously. Text analysed isolate Ai 2006; Lee 2011). quiz using fuzzy logic inferences (Author, 2021) multilayer perceptron Both induced promote meaningful for example, some studies have added conflicting text triggered high-confidence errors experimentation. Emotion Scales (EES) were then used even analysing unexpected text. (e.g Pekrun 2017, Vogl We show this study, that utilised foster interventions, hence providing great research potential.