Explainable student recommendations
Using modern natural language processing algorithms, student recommendations can be equipped by explanations of why they are advised as they are.
In this way, it is expected students will follow the advice more closely and with more motivation.
In this way, it is expected students will follow the advice more closely and with more motivation.
Example:
Students receive explanations such as "Since your rate of correct answers on matrix inverse is relatively low, we advise you to do more cases on matrix computation."
Reference:
Yongfeng Zhang and Xu Chen (2020 ), "Explainable Recommendation: A Survey and New Perspectives", Foundations and Trends® in Information Retrieval: Vol. 14: No. 1, pp 1-101.
Author of the tip:
Andrej Košir
University of Ljubljana