search for


Knowledge graph-based music recommender system using lyrics keyword extraction
Journal of the Korean Data & Information Science Society 2022;33:937-49
Published online November 30, 2022;
© 2022 Korean Data and Information Science Society.

Jaegwon Lee1 · Bonggyun Ko2

12Department of Mathematics and Statistics, Chonnam national University
Correspondence to: This research was supported by the BK21 FOUR (Fostering Outstanding Universities for Research, NO.5120200913674) funded by the Ministry of Education(MOE, Korea) and National Research Foundation of Korea(NRF).
This research was supported by the National Research Foundation of Korea(NRF) grant funded by the Korea government(MSIT) (No. 2021R1F1A1060049).
1 Master course, Department of Mathematics and Statistics, 77, Yongbong-ro, Buk-gu, Gwangju 61186, Korea.
2 Professor, Department of Mathematics and Statistics, 77, Yongbong-ro, Bukgu, Gwangju 61186, Korea. E-mail:
Received September 22, 2022; Revised October 25, 2022; Accepted October 30, 2022.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License ( which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
In the content industry, a recommender system is being used as a solution to the problem of information overload. However, existing recommender systems have problems with data sparsity and cold start. To alleviate this, methods of using various side information are being studied. The purpose of this study is to improve the performance of the music recommender system by suggesting a method of using song lyrics information as side information in the music recommender system. As a method for improving the performance of the music recommender system, we propose a knowledge graphbased music recommender system that extracts keywords from lyrics information and utilizes them as side information. The music recommender system of this study used Sentence-BERT, a natural language embedding model, for extracting lyrics keywords, and KPRN, which can complexly consider user information and item features for music recommendation. As a result of the experiment, the recommendation model using the lyrics keyword extraction method of this study showed superior recommendation accuracy than the existing model. Therefore, through these results, it was concluded that the lyrics keyword information improves the performance of the knowledge graph-based music recommender system as side information.
Keywords : Deep learning, keyword extraction, knowledge graph, natural language processing, recommender system.