Enhancing User Receptivity in E-Learning Environments through Gesture-Based Emotion Tracking as a Pathway to Mental Well-being and Health Equity (Preprint)

Abstract

In the realm of e-learning, where optimizing user engagement is paramount, this study explores the efficacy of gesture-based emotion tracking in enhancing user receptivity. Departing from traditional multimodal approaches, we focus on non-intrusive methods to measure emotions solely through gesture analysis. By leveraging this innovative approach, we aim to dissect the intricate dynamics of user engagement and propose actionable recommendations for improving online educational experiences. Our investigation not only sheds light on the effectiveness of gesture-based emotion tracking but also offers valuable insights for stakeholders in the e-learning domain, contributing to the advancement of mental well-being and health equity.

This research endeavors to advance our understanding of user receptivity in e-learning environments by employing gesture-based emotion tracking as a non-intrusive approach. Through rigorous analysis and experimentation, we seek to uncover the nuanced factors influencing user engagement and satisfaction on e-learning platforms. Our goal is to develop actionable recommendations informed by gesture analysis insights, empowering stakeholders to create more immersive and effective educational experiences. By adopting this innovative approach, we aim to contribute to the promotion of mental well-being and health equity in digital learning environments.

The methods undertaken in our study encompass data collection, feature selection, training and testing of the AI model. For data collection, real-time emotion data from a live e-learning website is collected using the Emaww API. The data is obtained with user consent, and privacy is ensured. The dataset consists of 6,026 users across 35 countries over 53 weeks. Feature selection involves calculating users’ receptivity and extracting web pages' properties. Users are considered receptive (towards the content of a web page) if they are focused for at least 70% of the time. The properties of web pages are related to their content and style. The training and testing utilize various machine learning techniques, and their performance is evaluated using accuracy, recall, F1-score, precision, and ROC curve.

The models were trained and validated using 10-fold cross-validation, with a diverse set of algorithms selected for analysis. The results showed that the Light Gradient Boosting Machine (LightGBM) performed the best, with an accuracy of 83.20%. Time was found to be an important input feature, as excluding it led to overfitting issues. Incorporating time as a feature enhanced the model's ability to generalize and make accurate predictions.

This research underscores the imperative features essential for integration into a web page to ensure user concentration on the content. The research findings, particularly the receptivity score derived from web page attributes, serve as a valuable resource for UI/UX designers. By providing insights into user engagement levels, these findings empower designers to craft web pages with a meticulous layout, thereby enhancing the ability to captivate and maintain user focus on the content. Consequently, these findings can be integrated into design practices as a guiding principle to optimize user experience and overall web page effectiveness.

N/A