Huge and increasing amount of online educational data presents both substantial opportunities and exposed challenges to enhance informative offering for study in machine learning. To automatically sense the prerequisite dependencies amongst gigantic quantities of online course, and to upkeep decision making such as curriculum planning for students, regardless of the language that student want to study in, and also to support curriculum and course design by educators based on in effect course offerings is one of the utmost essential tasks. Colleges take care of this issue in the outdated way, through scholastic guides, yet it is not clear how to address this issue with regards to MOOC’s then again cross-college offerings where courses don’t have distinctive IDs and are not depicted in a worldwide controlled or steady vocabulary. What can be done, so that an information rich concept graph can be obtained? Manual Measurement is apparently not scalable when it comes to thousands of concepts. The novel contribution of our paper is to address this open experiment with principle algorithmic results we target to achieve in this paper. We state our new method Concept Graph Learning (CGL). © 2016, International Journal of Pharmacy and Technology. All rights reserved.