Unsupervised Machine Learning Based on Recommendation of Pedagogical Resources

摘要

In E-learning context, we can recommend pedagogical resources to help learners. In this context, the recommender proposes the nearest resource(s) in term of similarity, but the scarcity of resources may affects seriously the quality of predictions. To make accurate predictions we begin in determining the scarce resources to be taken into account in the recommendation process. To achieve this objective we use the unsupervised neural network I2GNG (Improved Incremental Growing Neural Gas).