Year: 2025
Type: conference-paper
Abstract: This paper uses advanced data analysis and artificial intelligence methods to explore the relationships between metadata, lyrics and acoustic characteristics of music tracks. It creates a diverse dataset of songs consisting of audio features, lyrics and MP3 tracks. It uses a well-defined data collection methodology to study patterns and interactions related to song popularity, explicitness, genre identity and other song characteristics. This study also examines how these properties evolve. It aims to uncover hidden patterns and understand interactions present in the collected data using thorough exploratory data analysis, comprehensive feature engineering, natural language processing, machine learning techniques, and statistical methods.
DOI: 10.7148/2025-0604
Authors: Krystian Pielat, Daniel Grzonka
Institution: Cracow University of Technology
Language: English
Topic: Music and Audio Processing
Subfield: Signal Processing
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
OpenAlex ID: W4412603436