interactive-music-recommendation

A personalized and interactive music recommendation tool with bandit approach. A Bandit Approach to Music Recommendation notebook contains an overview of model.

Contributors

• Barış Can Esmer
• Beyza Gül Gürbüz
• H. Esra Aydemir
• Gökçe Uludoğan

Datasets

• The datasets utilized for finding out user behaviours and audio features:

  • LastFM User Listening History
  • MSD genre dataset

• To prepare data follow the instructions in data-preprocessing.sh and Music Dataset Preparation notebook.

Methods

The methods implemented to solve the model are:

• Epsilon-greedy
• LinUCB
• BayesUCB (with variational inference)

Experiments

References

[1] Wang, Xinxi, et al. "Exploration in interactive personalized music recommendation: a reinforcement learning approach." ACM Transactions on Multimedia Computing, Communications, and Applications (TOMM) 11.1 (2014): 7.

[2] Li, Lihong, et al. "A contextual-bandit approach to personalized news article recommendation." Proceedings of the 19th international conference on World wide web. ACM, 2010.