Course recommendation model based on Knowledge Graph Embedding

The use of graphs as a method of storing data has begun to rise significantly in recent years, due to the new way of representing data in graphs. This is leveraged by the structure of graphs that facilitate modeling interactions between real-world entities. With the rapid development of technologies such as machine learning and deep learning, the digitalization of education has entered the era with artificial intelligence as the main feature. As an important part of artificial intelligence technology, knowledge graph as a format of data storing provides possibilities for smart education and promotes the innovation and development of smart education. In this context, we present in this communication a model that we built to facilitate reaching the appropriate lesson for each learner among a large group of lessons in an eLearning graph, this model is practically divided into two parts, the first in which we sort the learners by defining the profile of each of them, which will facilitate their classification into groups, and in the second part, we connect each learner with his appropriate path to achieve the desired lesson, by building a nodal path through sequential prediction until reaching the target lesson, relying mainly on embedding graph models.