Development of a skill-based recommender system for online guitar learning guidance

Thumbnail Image

URL

Journal Title

Journal ISSN

Volume Title

School of Science | Master's thesis

Authors

Date

2025-04-28

Department

Major/Subject

Data Science

Mcode

Degree programme

Master's Programme in ICT Innovation

Language

en

Pages

64

Series

Abstract

This thesis explores the development of a skill-based recommender system for online guitar learning, focusing on personalization in educational technology. While platforms like Yousician provide engaging digital learning experiences, they lack the adaptive guidance of human instructors. To address this, we introduce a datadriven approach that models user skill progression and predicts exercise playability. We construct and leverage a user skill representation, computed from large-scale interaction data, and develop a machine learning-based Playability Prediction model. Our results show that guitar-related skill representations allow the outcome prediction of user-exercise interactions and can be instrumental in personalizing digital music learning. These findings contribute to the field of Educational Recommender Systems by demonstrating the effectiveness of knowledge tracing in music learning applications.

Description

Supervisor

Malmi, Lauri

Thesis advisor

Klapuri, Anssi

Keywords

online guitar learning, machine learning, educational technology, knowledge tracing, interpretable student representation, educational recommender systems

Other note

Citation

Permanent link to this item

https://urn.fi/URN:NBN:fi:aalto-202505193805

Collections

[dipl] Perustieteiden korkeakoulu / SCI