Latent Skill Embedding

A probabilistic model of student knowledge and educational content that can be used to recommend personalized lesson sequences in online courses

Abstract

Students in online courses generate large amounts of data that can be used to personalize the learning process and improve quality of education. We propose the Latent Skill Embedding, a probabilistic model of students and educational content that can be used to recommend personalized sequences of lessons with the goal of helping students prepare for specific assessments. Akin to collaborative filtering for recommender systems, the algorithm does not require students or content to be described by features, but it learns a representation using access traces. We formulate this problem as a regularized maximum-likelihood embedding of students, lessons, and assessments from historical student-content interactions. An empirical evaluation on large-scale data from Knewton, an adaptive learning technology company, shows that this approach predicts assessment results competitively with benchmark models and is able to discriminate between lesson sequences that lead to mastery and failure.

L@S 2016

Accepted as a Work-in-Progress at ACM Conference on Learning at Scale (L@S) 2016.

ICML 2015

Accepted as a workshop paper and invited talk at the Machine Learning for Education workshop at International Conference on Machine Learning (ICML) 2015.