Enhanced Clustering-based MOOC Recommendations using LinkedIn Profiles (MR-LI)

With the rapid development of massive open online courses (MOOCs), the interest of learners in MOOCs has increased significantly. MOOC platforms offer thousands of varied courses with many options. These options make it difficult for learners to choose courses that suit their needs and compatible with their interests. So, they become exposed to many courses on all topics. Therefore, there is an urgent need for personalized recommendation systems that assist learners in filtering courses according to their interests. Therefore, in this research, we target learners on the professional platform, LinkedIn, to be the basis for user modeling; the number of extracted profiles equals 5,039. Then, skill-based clustering algorithms were applied to LinkedIn users. Subsequently, we applied the similarity measurement between the vector features of the resulting clusters and the extracted course vectors. In the experiment result, four clusters were provided with the top-N course recommendations. Ultimately, the proposed approach was evaluated, and the F1-score of the approach was .81.

@article{Alruwaili2021,
  title     = {Enhanced Clustering-based MOOC Recommendations using LinkedIn Profiles (MR-LI)},
  journal   = {International Journal of Advanced Computer Science and Applications},
  volume    = {12},
  number    = {8},
  year      = {2021},
  publisher = {The Science and Information Organization},
  author    = {Fatimah Alruwaili and Dimah Alahmadi},
  doi       = {10.14569/IJACSA.2021.0120818},
  url       = {https://doi.org/10.14569/IJACSA.2021.0120818}
}