Learning Navigator goes beyond codifying concepts, to structure learning as a progression space (metric space of concepts with partial ordering of concept-dependencies) of competencies that we refer to as the Navigator Competency Framework (NCF). NCF enables AI techniques to be used to personalize learning pathways and provide real-time dashboards to stakeholders.
The Navigator Competency Framework (NCF) provides a framework for organizing all learning concepts. The NCF details the competency, embeds a variety of curated learning activities, captures every learner profile and relates to local norms. We use LDA and word2/doc2vec embeddings to compute competency models from a set of documents, use a number of machine learning techniques on activity stream data to curate learning activities and understand the learner, and then use principles of learning to personalize pathways.
The vastness of the learning space is structured as a progression space of competencies. Creating the progression space makes it possible for mathematical measures such as route and mean-time-to-learn to be computed. The structure also allows for the implementation of learning sciences research and best practices in teaching and learning.
Each competency is multidimensional and can be characterized by factors, such as depth of knowledge, common struggles on concept, and decay functions. The competencies are structured within the Navigator Competency Model (NCM).
There are many facets to learning that relate to the many disciplines including curriculum, skills and non-cognitive aspects. These facets are often related. For e.g., learner’s performance in science is often related to their proficiency in Math and their non-cognitive skills. Navigator has detailed and consistent representation of the learner across all of their facets. This enables the suggestion of personalized pathways based on an understanding of the learner.