Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Controllable Music Playlist Generation Based on Knowledge Graph and Reinforcement Learning.

Keigo Sakurai1, Ren Togo2, Takahiro Ogawa2

  • 1Graduate School of Information Science and Technology, Hokkaido University, N-14, W-9, Kita-ku, Sapporo 060-0814, Hokkaido, Japan.

Sensors (Basel, Switzerland)
|May 28, 2022
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Surgical Video Understanding with Alignment-Preserving Temporal Adaptation and Action Triplet Text Alignment.

Bioengineering (Basel, Switzerland)·2026
Same author

Personalized federated learning for medical vision-language models via efficient fine-tuning and uncertainty-aware disentanglement.

Journal of biomedical informatics·2026
Same author

Transfer Learning Strategies for Pathological Foundation Models: A Systematic Evaluation in Brain Tumor Classification.

Pathology international·2026
Same author

Dual-model weight selection and self-knowledge distillation for medical image classification.

Computers in biology and medicine·2026
Same author

Privacy-Aware Continual Self-Supervised Learning on Multi-Window Chest Computed Tomography for Domain-Shift Robustness.

Bioengineering (Basel, Switzerland)·2026
Same author

Deep-learning-based automatic liver segmentation using computed tomography images in dogs.

Frontiers in veterinary science·2025

This study introduces a new music playlist generation method using knowledge graphs and reinforcement learning to better capture long-term user preferences and guide listeners to new music. The approach enhances music discovery by optimizing recommendations based on user history and customizable feedback.

Area of Science:

  • Artificial Intelligence
  • Music Information Retrieval
  • Recommender Systems

Background:

  • Music streaming platforms have revolutionized music consumption, making playlist generation a key multimedia technique.
  • Conventional methods struggle to capture users' long-term music preferences effectively.
  • Personalized music recommendations are crucial for user satisfaction on streaming services.

Purpose of the Study:

  • To propose a novel music playlist generation method leveraging knowledge graphs and reinforcement learning.
  • To address the limitations of existing methods in capturing long-term user preferences.
  • To enhance music discovery by guiding users toward new music tracks aligned with their unique tastes.

Main Methods:

  • Utilizing reinforcement learning to model and predict users' long-term music preferences.
Keywords:
knowledge graphmultimedia techniquesmusic playlist generationmusic recommendationpreference sensingreinforcement learning

Related Experiment Videos

  • Integrating knowledge graph data to facilitate efficient reinforcement learning optimization.
  • Implementing a flexible reward function allowing user-defined parameters for personalized music guidance.

    • Main Results:

    • Demonstrated effectiveness in predicting music tracks based on user listening history.
    • Showcased the method's ability to guide users toward novel music satisfying unique preferences.
    • Validated the proposed approach's superior performance in capturing long-term user preferences.

    Conclusions:

    • The proposed knowledge graph and reinforcement learning-based method significantly improves music playlist generation.
    • This approach offers a more personalized and effective way to discover new music on streaming platforms.
    • The flexible reward function empowers users to actively shape their music discovery journey.