Leveraging Machine Learning for Song and Artist Success Prediction: A Multimodal Approach

Ozan Demirel; Tolga Kaya

doi:10.1007/978-3-031-98565-2_53

Leveraging Machine Learning for Song and Artist Success Prediction: A Multimodal Approach

Ozan Demirel^*
, Tolga Kaya

^*Corresponding author for this work

Department of Management Engineering

Istanbul Technical University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

The pursuit of understanding the science behind the success of a song has been a challenge for decades. Hit Song Prediction (HSP), a subfield of Music Information Retrieval, helps artists, labels, and talent scouts predict song performance and streamline market-driven song selection. The purpose of this study is to suggest a new prediction model which has the ability to detect the top 10 songs out of Billboard Hot 100 songs, using a multi-model approach. Using a dataset of 300 charted songs of last 2 years, we have developed a range of ML models including Gradient Boosting, Multi-Layer Perceptron (MLP) and Decision Trees. Analysis incorporated lyrics, audio characteristics, and artist-related data including social media metrics. Results reveal that combining audio, lyrics and social media data is a promising strategy in HSP.

Original language	English
Title of host publication	Intelligent and Fuzzy Systems - Artificial Intelligence in Human-Centric, Resilient and Sustainable Industries, Proceedings of the INFUS 2025 Conference
Editors	Cengiz Kahraman, Selcuk Cebi, Basar Oztaysi, Sezi Cevik Onar, Cagri Tolga, Irem Ucal Sari, Irem Otay
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	491-498
Number of pages	8
ISBN (Print)	9783031985645
DOIs	https://doi.org/10.1007/978-3-031-98565-2_53
Publication status	Published - 2025
Event	7th International Conference on Intelligent and Fuzzy Systems, INFUS 2025 - Istanbul, Turkey Duration: 29 Jul 2025 → 31 Jul 2025

Publication series

Name	Lecture Notes in Networks and Systems
Volume	1530 LNNS
ISSN (Print)	2367-3370
ISSN (Electronic)	2367-3389

Conference

Conference	7th International Conference on Intelligent and Fuzzy Systems, INFUS 2025
Country/Territory	Turkey
City	Istanbul
Period	29/07/25 → 31/07/25

Bibliographical note

Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

Keywords

Hit Song Prediction
Machine Learning
Natural Language Processing
Statistical Analysis on Music

Access to Document

10.1007/978-3-031-98565-2_53

Cite this

Demirel, O., & Kaya, T. (2025). Leveraging Machine Learning for Song and Artist Success Prediction: A Multimodal Approach. In C. Kahraman, S. Cebi, B. Oztaysi, S. Cevik Onar, C. Tolga, I. Ucal Sari, & I. Otay (Eds.), Intelligent and Fuzzy Systems - Artificial Intelligence in Human-Centric, Resilient and Sustainable Industries, Proceedings of the INFUS 2025 Conference (pp. 491-498). (Lecture Notes in Networks and Systems; Vol. 1530 LNNS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-031-98565-2_53

Demirel, Ozan ; Kaya, Tolga. / Leveraging Machine Learning for Song and Artist Success Prediction : A Multimodal Approach. Intelligent and Fuzzy Systems - Artificial Intelligence in Human-Centric, Resilient and Sustainable Industries, Proceedings of the INFUS 2025 Conference. editor / Cengiz Kahraman ; Selcuk Cebi ; Basar Oztaysi ; Sezi Cevik Onar ; Cagri Tolga ; Irem Ucal Sari ; Irem Otay. Springer Science and Business Media Deutschland GmbH, 2025. pp. 491-498 (Lecture Notes in Networks and Systems).

@inproceedings{600e407c284847419c2247a3f8f57585,

title = "Leveraging Machine Learning for Song and Artist Success Prediction: A Multimodal Approach",

abstract = "The pursuit of understanding the science behind the success of a song has been a challenge for decades. Hit Song Prediction (HSP), a subfield of Music Information Retrieval, helps artists, labels, and talent scouts predict song performance and streamline market-driven song selection. The purpose of this study is to suggest a new prediction model which has the ability to detect the top 10 songs out of Billboard Hot 100 songs, using a multi-model approach. Using a dataset of 300 charted songs of last 2 years, we have developed a range of ML models including Gradient Boosting, Multi-Layer Perceptron (MLP) and Decision Trees. Analysis incorporated lyrics, audio characteristics, and artist-related data including social media metrics. Results reveal that combining audio, lyrics and social media data is a promising strategy in HSP.",

keywords = "Hit Song Prediction, Machine Learning, Natural Language Processing, Statistical Analysis on Music",

author = "Ozan Demirel and Tolga Kaya",

note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.; 7th International Conference on Intelligent and Fuzzy Systems, INFUS 2025 ; Conference date: 29-07-2025 Through 31-07-2025",

year = "2025",

doi = "10.1007/978-3-031-98565-2\_53",

language = "English",

isbn = "9783031985645",

series = "Lecture Notes in Networks and Systems",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "491--498",

editor = "Cengiz Kahraman and Selcuk Cebi and Basar Oztaysi and \{Cevik Onar\}, Sezi and Cagri Tolga and \{Ucal Sari\}, Irem and Irem Otay",

booktitle = "Intelligent and Fuzzy Systems - Artificial Intelligence in Human-Centric, Resilient and Sustainable Industries, Proceedings of the INFUS 2025 Conference",

address = "Germany",

}

Demirel, O & Kaya, T 2025, Leveraging Machine Learning for Song and Artist Success Prediction: A Multimodal Approach. in C Kahraman, S Cebi, B Oztaysi, S Cevik Onar, C Tolga, I Ucal Sari & I Otay (eds), Intelligent and Fuzzy Systems - Artificial Intelligence in Human-Centric, Resilient and Sustainable Industries, Proceedings of the INFUS 2025 Conference. Lecture Notes in Networks and Systems, vol. 1530 LNNS, Springer Science and Business Media Deutschland GmbH, pp. 491-498, 7th International Conference on Intelligent and Fuzzy Systems, INFUS 2025, Istanbul, Turkey, 29/07/25. https://doi.org/10.1007/978-3-031-98565-2_53

Leveraging Machine Learning for Song and Artist Success Prediction: A Multimodal Approach. / Demirel, Ozan; Kaya, Tolga.
Intelligent and Fuzzy Systems - Artificial Intelligence in Human-Centric, Resilient and Sustainable Industries, Proceedings of the INFUS 2025 Conference. ed. / Cengiz Kahraman; Selcuk Cebi; Basar Oztaysi; Sezi Cevik Onar; Cagri Tolga; Irem Ucal Sari; Irem Otay. Springer Science and Business Media Deutschland GmbH, 2025. p. 491-498 (Lecture Notes in Networks and Systems; Vol. 1530 LNNS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Leveraging Machine Learning for Song and Artist Success Prediction

T2 - 7th International Conference on Intelligent and Fuzzy Systems, INFUS 2025

AU - Demirel, Ozan

AU - Kaya, Tolga

N1 - Publisher Copyright: © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

PY - 2025

Y1 - 2025

N2 - The pursuit of understanding the science behind the success of a song has been a challenge for decades. Hit Song Prediction (HSP), a subfield of Music Information Retrieval, helps artists, labels, and talent scouts predict song performance and streamline market-driven song selection. The purpose of this study is to suggest a new prediction model which has the ability to detect the top 10 songs out of Billboard Hot 100 songs, using a multi-model approach. Using a dataset of 300 charted songs of last 2 years, we have developed a range of ML models including Gradient Boosting, Multi-Layer Perceptron (MLP) and Decision Trees. Analysis incorporated lyrics, audio characteristics, and artist-related data including social media metrics. Results reveal that combining audio, lyrics and social media data is a promising strategy in HSP.

AB - The pursuit of understanding the science behind the success of a song has been a challenge for decades. Hit Song Prediction (HSP), a subfield of Music Information Retrieval, helps artists, labels, and talent scouts predict song performance and streamline market-driven song selection. The purpose of this study is to suggest a new prediction model which has the ability to detect the top 10 songs out of Billboard Hot 100 songs, using a multi-model approach. Using a dataset of 300 charted songs of last 2 years, we have developed a range of ML models including Gradient Boosting, Multi-Layer Perceptron (MLP) and Decision Trees. Analysis incorporated lyrics, audio characteristics, and artist-related data including social media metrics. Results reveal that combining audio, lyrics and social media data is a promising strategy in HSP.

KW - Hit Song Prediction

KW - Machine Learning

KW - Natural Language Processing

KW - Statistical Analysis on Music

UR - https://www.scopus.com/pages/publications/105013078585

U2 - 10.1007/978-3-031-98565-2_53

DO - 10.1007/978-3-031-98565-2_53

M3 - Conference contribution

AN - SCOPUS:105013078585

SN - 9783031985645

T3 - Lecture Notes in Networks and Systems

SP - 491

EP - 498

BT - Intelligent and Fuzzy Systems - Artificial Intelligence in Human-Centric, Resilient and Sustainable Industries, Proceedings of the INFUS 2025 Conference

A2 - Kahraman, Cengiz

A2 - Cebi, Selcuk

A2 - Oztaysi, Basar

A2 - Cevik Onar, Sezi

A2 - Tolga, Cagri

A2 - Ucal Sari, Irem

A2 - Otay, Irem

PB - Springer Science and Business Media Deutschland GmbH

Y2 - 29 July 2025 through 31 July 2025

ER -

Demirel O, Kaya T. Leveraging Machine Learning for Song and Artist Success Prediction: A Multimodal Approach. In Kahraman C, Cebi S, Oztaysi B, Cevik Onar S, Tolga C, Ucal Sari I, Otay I, editors, Intelligent and Fuzzy Systems - Artificial Intelligence in Human-Centric, Resilient and Sustainable Industries, Proceedings of the INFUS 2025 Conference. Springer Science and Business Media Deutschland GmbH. 2025. p. 491-498. (Lecture Notes in Networks and Systems). doi: 10.1007/978-3-031-98565-2_53

Leveraging Machine Learning for Song and Artist Success Prediction: A Multimodal Approach

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this