This book aims to fully demonstrate the burnout of learners in online learning processes. The authors propose a series of feasible and reliable solutions to sufficiently obtain and analyze massive instances of online learning behavior.

In order to flexibly perceive and intervene in the "burnout state" and improve online learning processes and learning effectiveness, the authors design and construct various novel data analysis models and decision prediction methods using technological means and data-driven learning strategies. Their innovative methods, techniques, and decisions would benefit autonomous learning behavior tracking and stimulate the learning interest of online learning processes enabled by predictive learning analytics. By employing behavioral science research strategies, they build adaptive prediction and optimization measures for positive online learning patterns, improve learning behaviors, optimize learning states, and establish dynamic and sustainable knowledge tracing paths and behavior scheduling methods, enabling users to achieve self-organization and self-mobilization in their overall learning processes.

The book will appeal to scholars and learners in Europe, North America, and Asia, especially those majoring in educational statistics and measurement, educational big data, learning analytics, educational psychology, artificial intelligence in education, computer science, and online collaborative learning.