Music Recommendation System |
||
 |
 |
|
| © 2023 by IJCTT Journal | ||
| Volume-71 Issue-5 |
||
| Year of Publication : 2023 | ||
| Authors : Brijmohan Daga, Harsh Kadam, Sarthak Shrungare, Scott Fernandes, Ashwin Johsnon | ||
| DOI : 10.14445/22312803/IJCTT-V71I5P105 | ||
How to Cite?
Brijmohan Daga, Harsh Kadam, Sarthak Shrungare, Scott Fernandes, Ashwin Johsnon, "Music Recommendation System," International Journal of Computer Trends and Technology, vol. 71, no. 5, pp. 26-36, 2023. Crossref, https://doi.org/10.14445/22312803/IJCTT-V71I5P105
Abstract
Music recommendation systems have become a crucial aspect of the music industry, providing personalized recommendations to users based on their listening habits. This project aims to develop a music recommendation system using the K-Means clustering algorithm. The system will collect data from the user's listening history and use it to generate recommendations based on similar listening habits of other users.
K-Means is a commonly used unsupervised learning algorithm that is used for clustering and partitioning of data. The algorithm works by grouping similar data points into clusters, which can be used for various purposes, including recommendations. In this project, K-Means will be used to group similar users based on their listening habits and generate recommendations based on their preferences.
The system will be designed to collect data from multiple sources, including music streaming platforms such as Spotify, and process it to generate recommendations. The collected data will include the user's listening history, playlists, and favourite artists. The algorithm will process this data and group users with similar preferences into clusters, which will be used to generate recommendations for each user.
In addition to K-Means, the project will also evaluate the performance of the K-Medoids algorithm, a variation of K-Means, to determine its effectiveness in generating recommendations. The evaluation results will be compared to determine which algorithm provides better recommendations, and the results will be discussed in detail in the paper.
The project paper will discuss developing and implementing the music recommendation system using K-Means and K-Medoids, including a detailed description of the algorithms used, the data collected, and the results obtained. The paper will also provide a comprehensive evaluation of the system's performance, including its accuracy and efficiency, and suggest areas for future improvement.
In conclusion, this project aims to demonstrate the effectiveness of K-Means and K-Medoids in generating music recommendations. It provides valuable insights into the use of these algorithms in the development of music recommendation systems. The project results will be useful for researchers and practitioners interested in exploring the use of clustering algorithms in music recommendation systems. The project will begin with collecting and pre-processing the dataset, which will include a large number of songs along with their audio features. The audio features will then be used as inputs to the K-Means algorithm, which will partition the songs into K clusters. Each cluster will represent a unique musical style, and songs within a cluster will have similar audio features.
Once the songs have been clustered, the next step will be to create a recommendation system that suggests songs to users based on their past listening behaviour and preferences. This will be done by analysing the songs a user has listened to in the past, and finding the cluster most similar to the user's preferences. The system will then suggest songs from that cluster that the user has not listened to yet.
Keywords
Clustering, K-means, Music, Recommendation, Similarity.
Reference
[1] P. Lamere, “Playlist Generation at Pandora,” Proceedings of the International Conference on Music Information Retrieval, 2012.
[2] T. Bertin-Mahieux et al., “The Million Song Dataset,” Proceedings of the 12th International Conference on Music Information Retrieval, 2011.
[Google Scholar] [Publisher Link]
[3] J. H. Lee, and J. W. Lee, “A Collaborative Filtering Recommendation Algorithm Using Improved Item-based Nearest Neighbors,” Journal of the Korean Society for Quality Management, vol. 42, no. 1, pp. 95–106, 2014.
[4] D. P. W. Ellis, and G. Poliner, “The Quest for Ground Truth in Musical Artist Similarity,” Proceedings of the 8th International Conference on Music Information Retrieval, 2007.
[Google Scholar]
[5] B. McFee, “Hierarchical Verification of Music Metadata Using Content and Web Information,” Proceedings of the 14th International Society for Music Information Retrieval Conference, 2013.
[6] J. Schedl, C. Bauer, and M. Schedl, “Predicting User Ratings of Music with Audio and Context Features,” Proceedings of the 12th International Conference on Music Information Retrieval, 2011.
[7] Yue Shi et al., “CLiMF: Learning to Maximize Reciprocal Rank with Collaborative Less-is-More Filtering,” Proceedings of the 7th ACM Conference on Recommender Systems, 2013.
[CrossRef] [Google Scholar] [Publisher Link]
[8] S. Abdul-Jabbar, and R. V. Yampolskiy, “Hybrid Approach to Music Recommendation Based on User Preference and Audio Content,” Proceedings of the 12th International Conference on Music Information Retrieval, 2011.
[9] C. C. M. Yeh et al., “A Hybrid Music Recommendation System for Music Streaming Services,” Proceedings of the 8th International Conference on Ubiquitous Information Management and Communication, 2014.
[10] B. McFee, M. Rafferty, and G. R. Lanckriet, “Learning Content-Aware Music Recommendation,” Proceedings of the 12th International Society for Music Information Retrieval Conference, Miami, USA, 2011.
[11] Y. Yang, Y. Zhang, and X. Li, “A Survey of Collaborative Filtering-Based Recommendation Algorithms,” Journal of Computer and Communication, vol. 2, no. 5, pp. 1-9, 2012.
[12] Dr Brijmohan Daga et al., “Computer Science Career Recommendation System using Artificial Neural Network,” International Journal of Computer Trends and Technology, vol. 68, no. 3, pp.84-88, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[13] Rochelle Cordeiro et al., “Automated training for Job Interviews,” International Journal of Computer Trends and Technology, vol. 68, no. 3, pp. 74-79, 2020.
[CrossRef] [Publisher Link]
[14] B. S. Daga, A. A. Ghatol, and A. B. Daga, “Video Events Extraction Based on Mixed Feature Modeling,” International Journal of Computer Applications, vol. 157, no. 10, pp. 22-29, 2017.
[CrossRef] [Google Scholar] [Publisher Link]
[15] S. Sharma, and A. S. Jalal, “A Review of Music Recommendation Systems,” International Journal of Computer Applications, vol. 118, no. 6, pp. 18-22, 205.
[16] B. Logan, “Music Recommendation and Discovery Using Audio Content Analysis,” Proceedings of the 12th International Society for Music Information Retrieval Conference, Miami, USA, 2011.
[17] Y. Wang, and H. Gao, “Music Recommendation System Based on Association Rules Mining,” Journal of Computer Applications, vol. 35, no. 5, pp. 1272-1276, 2015.
[18] Yongqiang Bai et al., “A Personalized Music Recommendation System Based on User’s Emotion and Personality,” Proceedings of the 2013 IEEE International Conference on Systems, Man, and Cybernetics (SMC), Manchester, UK.
[19] L. Chen, Y. Chen, and C. Chen, “Personalized Music Recommendation System Based on User Emotion, Content and Social Interaction,” Multimedia Tools and Applications, vol. 73, no. 1, pp. 99-117, 2014.
[20] Y. Wang, and H. Gao, “A Music Recommendation Algorithm Based on Bayesian Network,” Journal of Computer Applications, vol. 36, no. 1, pp. 208-212, 2016.
[21] A. Schindler, A. Rauber, and M. Hübner, “A Comparative Study of Music Similarity Measures for Content-Based Recommendation,” Proceedings of the 12th International Society for Music Information Retrieval Conference (ISMIR 2011), Miami, USA.
[22] Y. Hu, and X. Chen, “A New Personalized Music Recommendation Algorithm Based on Genetic Algorithm,” Multimedia Tools and Applications, vol. 75, no. 5, pp. 2407-2418, 2016.