The field of education is experiencing a massive digitisation process that has been ongoing for the past decade. The role played by distance learning and Video-Based Learning, which is even more reinforced by the pandemic crisis, has become an established reality. However, the typical features of video consumption, such as sequential viewing and viewing time proportional to duration, often lead to sub-optimal conditions for the use of video lessons in the process of acquisition, retrieval and consolidation of learning contents. Video augmentation can prove to be an effective support to learners, allowing a more flexible exploration of contents, a better understanding of concepts and relationships between concepts and an optimization of time required for video consumption at different stages of the learning process. This thesis focuses therefore on the study of methods for: 1) enhancing video capabilities through video augmentation features; 2) extracting concept and relationships from video materials; 3) developing intelligent user interfaces based on the knowledge extracted. The main research goal is to understand to what extent video augmentation can improve the learning experience. This research goal inspired the design of EDURELL Framework, within which two applications were developed to enable the testing of augmented methods and their provision. The novelty of this work lies in using the knowledge within the video, without exploiting external materials, to exploit its educational potential. The enhancement of the user interface takes place through various support features among which in particular a map that progressively highlights the prerequisite relationships between the concepts as they are explained, ie, following the advancement of the video. The proposed approach has been designed following a user-centered iterative approach and the results in terms of effect and impact on video comprehension and learning experience make a contribution to the research in this field.