Universität Bielefeld › Technische Fakultät › NI
Recommender systems are being ubiquitously used to optimally couple users and products. The goal of this project is to employ standard methods used to produce recommendations [1,4]
in order to provide learners with suitable practice units while they are learning a motor skill, such as playing piano [2]. The resulting project should explore applications of available software [3].
Multiple work directions include the following building blocks:
1: Automatic creation of database with practice exercises from internet (e.g. based on pieces available in midi format); automatic labeling according to the complexity level
2:Web-based interface to gather learner-data
3: Testing different approaches to sequence-aware recommendation systems to accelerate learning (e.g. developed for amazon and netflix recommendation [5])
[1] https://developers.google.com/machine-learning/recommendation/overview/t...
[2] https://arxiv.org/abs/2106.12937
[3] https://github.com/microsoft/recommenders
[4] https://arxiv.org/pdf/1905.06874.pdf
[5] https://aws.amazon.com/blogs/media/whats-new-in-recommender-systems/
*Picture taken from https://realpython.com/build-recommendation-engine-collaborative-filtering/