Music Recommendations Based on User's Mood Using Convolutional Neural Networks
This paper proposes a method for music recommendations using emotions, using deep learning techniques. The method is composed of two modules. The emotion detection module, which utilizes a hybrid architecture involving a Convolutional Neural Network (CNN) and a Reccurent Neural Network using Long-S...
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Babes-Bolyai University, Cluj-Napoca
2022-07-01
|
| Series: | Studia Universitatis Babes-Bolyai: Series Informatica |
| Subjects: | |
| Online Access: | http://193.231.18.162/index.php/subbinformatica/article/view/1204 |
| _version_ | 1797322225691394048 |
|---|---|
| author | Andrei PETRESCU |
| author_facet | Andrei PETRESCU |
| author_sort | Andrei PETRESCU |
| collection | DOAJ |
| description |
This paper proposes a method for music recommendations using emotions, using deep learning techniques. The method is composed of two modules. The emotion detection module, which utilizes a hybrid architecture involving a Convolutional Neural Network (CNN) and a Reccurent Neural Network using Long-Short Term Memory (LSTM) Cells. We compared individual architectures of CNNs and LSTMs against our hybrid approach, outperforming them during experiments. We evaluated the modules on our own data set, created using Spotify’s API and containing 2028 songs from different genres and linguistic families, labeled with valence and arousal values. The model also outperforms other related approaches, however we did not evaluate them on the same data set. The predictions are used by the second module, for which we proposed a simple method of ordering the results based on the similarity to user’s input.
Received by the editors: 10 October 2021.
2010 Mathematics Subject Classification. 68T45.
1998 CR Categories and Descriptors. I.4.8 [Image Processing and Computer Vision]:Scene Analysis – Object recognition; I.2.6 [Artificial Intelligence]: Learning – Connectionism and neural nets; I.2.10 [Artificial Intelligence]: Vision and Scene Understanding – Intensity, color, photometry, and thresholding.
|
| first_indexed | 2024-03-08T05:11:06Z |
| format | Article |
| id | doaj.art-64b16edc0d724b648fddcaceaf3d653f |
| institution | Directory Open Access Journal |
| issn | 2065-9601 |
| language | English |
| last_indexed | 2024-03-08T05:11:06Z |
| publishDate | 2022-07-01 |
| publisher | Babes-Bolyai University, Cluj-Napoca |
| record_format | Article |
| series | Studia Universitatis Babes-Bolyai: Series Informatica |
| spelling | doaj.art-64b16edc0d724b648fddcaceaf3d653f2024-02-07T10:03:31ZengBabes-Bolyai University, Cluj-NapocaStudia Universitatis Babes-Bolyai: Series Informatica2065-96012022-07-0167110.24193/subbi.2022.1.04Music Recommendations Based on User's Mood Using Convolutional Neural NetworksAndrei PETRESCU0Department of Computer Science, Babes-Bolyai University, 1, M. Kogalniceanu Street, 400084, Cluj-Napoca, Romania Email address: andrei.petrescu@stud.ubbcluj.ro This paper proposes a method for music recommendations using emotions, using deep learning techniques. The method is composed of two modules. The emotion detection module, which utilizes a hybrid architecture involving a Convolutional Neural Network (CNN) and a Reccurent Neural Network using Long-Short Term Memory (LSTM) Cells. We compared individual architectures of CNNs and LSTMs against our hybrid approach, outperforming them during experiments. We evaluated the modules on our own data set, created using Spotify’s API and containing 2028 songs from different genres and linguistic families, labeled with valence and arousal values. The model also outperforms other related approaches, however we did not evaluate them on the same data set. The predictions are used by the second module, for which we proposed a simple method of ordering the results based on the similarity to user’s input. Received by the editors: 10 October 2021. 2010 Mathematics Subject Classification. 68T45. 1998 CR Categories and Descriptors. I.4.8 [Image Processing and Computer Vision]:Scene Analysis – Object recognition; I.2.6 [Artificial Intelligence]: Learning – Connectionism and neural nets; I.2.10 [Artificial Intelligence]: Vision and Scene Understanding – Intensity, color, photometry, and thresholding. http://193.231.18.162/index.php/subbinformatica/article/view/1204mood, emotion, valence, energy, convolutional neural network,recurrent neural networks, long-short term memory, hybrid, regression, classification. |
| spellingShingle | Andrei PETRESCU Music Recommendations Based on User's Mood Using Convolutional Neural Networks Studia Universitatis Babes-Bolyai: Series Informatica mood, emotion, valence, energy, convolutional neural network,recurrent neural networks, long-short term memory, hybrid, regression, classification. |
| title | Music Recommendations Based on User's Mood Using Convolutional Neural Networks |
| title_full | Music Recommendations Based on User's Mood Using Convolutional Neural Networks |
| title_fullStr | Music Recommendations Based on User's Mood Using Convolutional Neural Networks |
| title_full_unstemmed | Music Recommendations Based on User's Mood Using Convolutional Neural Networks |
| title_short | Music Recommendations Based on User's Mood Using Convolutional Neural Networks |
| title_sort | music recommendations based on user s mood using convolutional neural networks |
| topic | mood, emotion, valence, energy, convolutional neural network,recurrent neural networks, long-short term memory, hybrid, regression, classification. |
| url | http://193.231.18.162/index.php/subbinformatica/article/view/1204 |
| work_keys_str_mv | AT andreipetrescu musicrecommendationsbasedonusersmoodusingconvolutionalneuralnetworks |