Abstract
<jats:p>This work examines the potential of Learning Analytics as a strategy for personalizing learning in virtual higher education, moving beyond a purely technocentric approach. Through a theoretical and methodological articulation, it analyzes how the digital traces generated by students in virtual learning environments can be transformed into interpretable pedagogical evidence to optimize learning pathways, strengthen student retention, and mitigate dropout risks. The text establishes important conceptual distinctions among Learning Analytics, Educational Data Mining (EDM), and Academic Analytics, defining the operational scope of each discipline within the context of university management. It also provides an in-depth examination of the dimensions that underpin the data lifecycle, ranging from data collection sources and technological interoperability to the development of valid indicators of student engagement, regularity, and academic performance. Through a comprehensive approach, the book describes the implementation of descriptive, predictive, and prescriptive models, detailing the operation of adaptive recommendation systems, automated academic tutoring, and evidence-based instructional design. Beyond computational and algorithmic components, the authors place strong emphasis on ethical considerations, transparency, institutional governance, and algorithmic fairness, warning against the risks of excessive surveillance, automated bias, and the stigmatizing labeling of students. In conclusion, the work is proposed as a reference framework for educators, researchers, and university administrators committed to digital transformation. It demonstrates that Learning Analytics achieves its true value only when it is placed at the service of human development, socio-educational equity, and the continuous improvement of academic quality in virtual learning environments.</jats:p>