Manual music classification is a slow and costly process. Most recent works about music auto-classification such as genre or emotions make this process easier, but are focused on a single task. In this work, a music multi-classification platform is presented. This platform is based on multi-agent systems, allowing to distribute the extraction, classification, and service tasks among agents. The platform performs a musical genre and emotional classification and provides context information of songs from social networks such as Twitter and Last.fm. The methods chosen based on meta-classifiers to perform single-label and multi-label classification obtain great results. In the case of multi-label classification, better results are obtained than in other previous works.
%0 Journal Article
%1 Pérez-Marcos2020
%A Perez-Marcos, Javier
%A Jimenez-Bravo, Diego M.
%A De Paz, Juan F.
%A Villarrubia Gonzalez, Gabriel
%A Lopez, Vivian F.
%A Gil, Ana B.
%D 2020
%J Knowledge and Information Systems
%K Auto-classification Social Performance Multi-label analysis intelligence classification Meta-classifiers Single-label Distributed Emotional Music processing Genre information Artificial Twitter recommendation learning networks from:kamber Machine Multi-agent systems Contextual Last.fm evaluation Multi-classification
%N 1
%P 401--422
%R 10.1007/s10115-018-1319-2
%T Multi-agent system application for music features extraction, meta-classification and context analysis
%U https://doi.org/10.1007/s10115-018-1319-2
%V 62
%X Manual music classification is a slow and costly process. Most recent works about music auto-classification such as genre or emotions make this process easier, but are focused on a single task. In this work, a music multi-classification platform is presented. This platform is based on multi-agent systems, allowing to distribute the extraction, classification, and service tasks among agents. The platform performs a musical genre and emotional classification and provides context information of songs from social networks such as Twitter and Last.fm. The methods chosen based on meta-classifiers to perform single-label and multi-label classification obtain great results. In the case of multi-label classification, better results are obtained than in other previous works.
@article{Pérez-Marcos2020,
abstract = {Manual music classification is a slow and costly process. Most recent works about music auto-classification such as genre or emotions make this process easier, but are focused on a single task. In this work, a music multi-classification platform is presented. This platform is based on multi-agent systems, allowing to distribute the extraction, classification, and service tasks among agents. The platform performs a musical genre and emotional classification and provides context information of songs from social networks such as Twitter and Last.fm. The methods chosen based on meta-classifiers to perform single-label and multi-label classification obtain great results. In the case of multi-label classification, better results are obtained than in other previous works.},
added-at = {2024-03-27T09:25:06.000+0100},
author = {Perez-Marcos, Javier and Jimenez-Bravo, Diego M. and De Paz, Juan F. and Villarrubia Gonzalez, Gabriel and Lopez, Vivian F. and Gil, Ana B.},
biburl = {https://www.bibsonomy.org/bibtex/2cf06b2a1320b8c317491949cdcf1b261/sop2-ffzg},
day = 01,
doi = {10.1007/s10115-018-1319-2},
interhash = {24178c54da3b1366db95ef27af56241c},
intrahash = {cf06b2a1320b8c317491949cdcf1b261},
issn = {0219-3116},
journal = {Knowledge and Information Systems},
keywords = {Auto-classification Social Performance Multi-label analysis intelligence classification Meta-classifiers Single-label Distributed Emotional Music processing Genre information Artificial Twitter recommendation learning networks from:kamber Machine Multi-agent systems Contextual Last.fm evaluation Multi-classification},
month = jan,
number = 1,
pages = {401--422},
timestamp = {2024-03-27T09:25:06.000+0100},
title = {Multi-agent system application for music features extraction, meta-classification and context analysis},
url = {https://doi.org/10.1007/s10115-018-1319-2},
volume = 62,
year = 2020
}