Most virtual learning environment fails to recognize that students have different needs when it comes to learning. With the evolving characteristics and tendencies of students, these learning environments must provide adaptation and personalization features for adaptive learning materials, course content and navigational designs to support student's learning styles. Based from the data mining results of learner behavioral features of five hundred seven (507) tertiary students, an accurate model for classification of students' learning styles were derived using J48 decision tree algorithm. The model was implemented in a prototype using a framework and a proposed system architectural design of an adaptive virtual learning environment. The study resulted in the development of an adaptive virtual learning environment prototype where learners' preferences are dynamically diagnosed to intelligently personalize the course content design and user interfaces for them.