With the existing challenge in education industry reinforced by theconsistent pandemic situation, adapting the educational resources in order to help to personalize and accelerate the usage of digital content used by teacher is an emerging topic in this sector. New generation of student needs evolving multimedia resources that are easy to use, assemble and personalize, adapted to the subject and time of learningin a digital and mobile world. In addition, the quality and relevancy of resources is one of the key factors to increase the value of education and speed up digital learning for all kind of students. Also, identify the best resources for a specific knowledge domain and educational level is another challenge. Machine Learning could be efficiently used to increase the knowledge assessment for a specific student with the most valuable digital resources. In this paper, we proposed MLM-based educational recommender system (ERS) named EKRAM. The Educational Knowledge Resources Assessment using Machine Learning &linked Networks (Part I), whose objective is to assess content using machine learning models that analyze educational and classification metadata in order to identify the most relevant content and organize them to produce an educational resource for a specific usage in a set of progressive levels. Using simulation prototypes, we tried to demonstrate that EKRAM may improves accuracy and efficiencyof the educational process. This article is the firstpaper of Educatio project using EKRAM.