Article
Chemistry, Analytical
Keigo Sakurai, Ren Togo, Takahiro Ogawa, Miki Haseyama
Summary: This study proposes a novel music playlist generation method based on a knowledge graph and reinforcement learning. It aims to overcome the difficulty of capturing the target users' long-term preferences using a reinforcement learning algorithm and informative knowledge graph data, along with a flexible reward function. The effectiveness of the method is confirmed through verification of prediction performance and guidance performance.
Article
Computer Science, Artificial Intelligence
Ali Yurekli, Cihan Kaleli, Alper Bilge
Summary: The paper presents a multi-stage retrieval system utilizing user-generated playlist titles to address the cold-start problem in music recommender systems. The system employs clustering, latent semantic indexing, and singular value decomposition to enhance recommendation accuracy and user experience in music listening.
INTERNATIONAL JOURNAL OF MULTIMEDIA INFORMATION RETRIEVAL
(2021)
Article
Business
Zhi Li, Min Song, Shen Duan, Zhe Wang
Summary: This study investigates the features of playlist titles and covers that attract users and the consequences of playlist selection on music streaming platforms. The findings suggest that the linguistic style of titles and the image style of covers have moderating effects on the relationship between the number of comments and plays of playlists.
JOURNAL OF INNOVATION & KNOWLEDGE
(2022)
Article
Computer Science, Artificial Intelligence
Ali Yurekli, Alper Bilge, Cihan Kaleli
Summary: The study shows that playlist titles can serve as an auxiliary data source to address the cold-start problem, with titles related to specific themes providing highly accurate recommendations. However, the correlation between title and recommendation usability is weak.
JOURNAL OF AMBIENT INTELLIGENCE AND HUMANIZED COMPUTING
(2021)
Article
Computer Science, Artificial Intelligence
Anna Gatzioura, Joao Vinagre, Alipio Mario Jorge, Miquel Sanchez-Marre
Summary: This study introduces a hybrid recommender system HybA, which focuses on improving the quality of music playlist recommendations by considering semantic similarity at the recommendation moment, and providing support for dimensions beyond accuracy, such as coherence and diversity. Experiments have shown that this system outperforms other state of the art techniques in terms of accuracy, while balancing between diversity and coherence.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2021)
Article
Computer Science, Artificial Intelligence
Changfei Tang, Jun Zhang
Summary: Existing research on data analysis and personalized recommendation algorithms for music mainly focuses on improving traditional recommendation systems. However, there is a lack of efficient solutions for the cold start and sparse nature of rating matrices. This paper proposes a personalized recommendation algorithm based on a deep belief neural network, which can intelligently and efficiently support music teaching.
Article
Mathematics, Interdisciplinary Applications
Hong Kai
Summary: This paper proposes a tightly coupled fusion model based on deep learning and collaborative filtering to improve the prediction accuracy of video background music recommendation. By comprehensively considering user's interests and video background music characteristics, the optimized recommendation result is obtained.
Article
Computer Science, Hardware & Architecture
Tingrong Yin
Summary: Recommender systems using IoT and deep learning are crucial for enhancing the user experience on online music streaming platforms. However, building a music recommender system is challenging due to the short duration and contextual dependencies of tracks. Traditional algorithms fail to extract deep-level features from audio signals and effectively mine user preferences. Therefore, this paper proposes a deep learning-based model that preprocesses data, generates Mel spectrogram features, and utilizes a convolutional neural network to categorize music tracks. Experimental research and comparisons demonstrate the algorithm's effectiveness in music recommendation.
MOBILE NETWORKS & APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Yuxu Mao, Guoqiang Zhong, Haizhen Wang, Kaizhu Huang
Summary: This paper proposes a model called Music-CRN, which facilitates music classification and recommendation by learning the audio content features. Experimental results demonstrate that Music-CRN achieves better performance than previous methods in music classification and recommendation tasks.
COGNITIVE COMPUTATION
(2022)
Article
Computer Science, Information Systems
Linlin Peng
Summary: With the advancement of mobile Internet and streaming media technology, digital music has become increasingly popular. Traditional music indexing technology using keywords is insufficient for finding favorite music. Personalized music recommendation is the current focus, enabling users to quickly and accurately discover music tracks that match their interests.
MOBILE INFORMATION SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Mohamadreza Sheikh Fathollahi, Farbod Razzazi
Summary: This paper presents a music genre classification system based on convolutional neural networks, applied in an automatic music recommendation system. The results demonstrate significant accuracy in the 10-Best results of the recommendation system.
INTERNATIONAL JOURNAL OF MULTIMEDIA INFORMATION RETRIEVAL
(2021)
Article
Mathematical & Computational Biology
Juan Sun
Summary: This paper proposes a personalized recommendation algorithm based on the Spark platform to address the low accuracy and poor real-time performance of traditional recommendation algorithms in dealing with large-scale music data. The algorithm optimizes the initial centroids of K-means using an artificial fish swarm algorithm and applies collaborative filtering algorithm to calculate the correlation between users. Experimental results demonstrate that the algorithm achieves higher recommendation accuracy and real-time performance, especially in handling large-scale music data.
COMPUTATIONAL INTELLIGENCE AND NEUROSCIENCE
(2022)
Article
Information Science & Library Science
Carlos Fernandez-Loria, Foster Provost, Jesse Anderton, Benjamin Carterette, Praveen Chandar
Summary: This study presents a systematic comparison of different methods for individual treatment assignment and categorizes them into three metalearners. The study demonstrates that optimizing for prediction of outcomes or causal effects is not the same as optimizing for treatment assignments, and the A-learner may lead to better treatment assignments. The practical implications of the findings are shown in a real-world application, highlighting the value of large-scale A/B tests for learning treatment-assignment policies.
INFORMATION SYSTEMS RESEARCH
(2023)
Article
Computer Science, Information Systems
Tingting Zhang, Shengnan Liu
Summary: This paper proposes a hybrid recommendation algorithm based on music genes and improved knowledge graph to address the accuracy issue in music recommendation. The algorithm utilizes recommendation patterns of music genes, user and item label information, knowledge graphs, and deep learning to extract features and perform score prediction for recommendation.
SECURITY AND COMMUNICATION NETWORKS
(2022)
Article
Computer Science, Artificial Intelligence
Ling Dong
Summary: Music is an artistic form that expresses thoughts and emotions through harmony, rhythm, and melody. This research focuses on Sichuan unvoiced music and proposes a melody generation algorithm using deep learning and evolutionary algorithms. The algorithm utilizes RNN-LSTM for melody generation and GA for melody optimization. The proposed method is compared to existing approaches and found to be more effective in terms of accuracy and the average melody score.
