| spelling |
Mirella Moura Morohttp://lattes.cnpq.br/6408321790990372Anisio Mendes LacerdaRenato VimieiroMichele Amaral Brandãohttp://lattes.cnpq.br/7922939906211697Mariana de Oliveira Santos Silva2022-08-25T16:05:34Z2022-08-25T16:05:34Z2020-03-30http://hdl.handle.net/1843/44585https://orcid.org/0000-0003-0110-9924Hit songs are more successful than average, where key factors make such songs qualitatively superior to others. Current techniques to predict hit songs exploit features that describe songs individually. We propose tackling this prediction problem through a multimodal form with songs’ features fused, together. Specifically, we describe songs through features from three modalities: music, artist and album. Initially, we identify collaboration profiles in a musical network composed of successful artists, unveiling how artists professionally connect can significantly impact their success. Then, to deepen such analyses, we use time series and the Granger Causality test for assessing whether there is a causal relationship between collaboration profiles and artists’ popularity. Finally, we model the Hit Song Prediction problem as two distinct tasks: classification and placement. The former is a classical machine learning binary classification problem and is a direct application of our fusion strategies. The later is a modeling approach that ranks a song relative to a given chart, then predicts hit songs and provides comparative popularity information of a set of songs. Furthermore, we emphasize collaboration artists’ profiles as important features when describing their songs. Extensive empirical studies using various features from the modalities confirm the effectiveness of our method that fuses heterogeneous data for both tasks.As músicas de sucesso são mais bem-sucedidas do que a média, onde fatores-chave tornam essas músicas qualitativamente superiores às outras. As técnicas atuais para prever músicas de sucesso exploram recursos que descrevem músicas individualmente. Propomos abordar esse problema de previsão através de uma forma multimodal, com a fusão de recursos musicais. Especificamente, descrevemos as músicas através de recursos de três modalidades: música, artista e álbum. Inicialmente, identificamos perfis de colaboração em uma rede musical composta por artistas de sucesso, revelando como os artistas se conectam profissionalmente pode impactar significativamente seu sucesso. Para aprofundar essas análises, usamos séries temporais e o teste de causalidade de Granger para avaliar se há uma relação causal entre perfis de colaboração e popularidade dos artistas. Finalmente, modelamos o problema de previsão de hits como duas tarefas distintas: classification e placement. A primeira é um problema clássico de classificação binária de aprendizado de máquina e é uma aplicação direta de nossas estratégias de fusão. A posterior é uma abordagem de modelagem que posiciona uma música em relação a um determinado ranking, prediz músicas de sucesso e fornece informações comparativas de popularidade de um conjunto de músicas. Além disso, enfatizamos os perfis dos artistas colaboradores como características importantes ao descrever suas músicas. Estudos empíricos extensos, usando diferentes features de cada modalidade, mostram a eficácia de nosso método que combina dados heterogêneos para ambas as tarefas.CNPq - Conselho Nacional de Desenvolvimento Científico e TecnológicoengUniversidade Federal de Minas GeraisPrograma de Pós-Graduação em Ciência da ComputaçãoUFMGBrasilICX - DEPARTAMENTO DE CIÊNCIA DA COMPUTAÇÃOhttp://creativecommons.org/licenses/by-nc-nd/3.0/pt/info:eu-repo/semantics/openAccessComputação – TesesCiência de dados – TesesAprendizado do Computador – TesesRedes complexas – TesesMineração de dados (Computação) – TesesHit song science – Teses.Hit Song ScienceData ScienceMachine LearningComplex NetworksMusic Data MiningCollaboration-aware hit song analysis and predictionAnálise e previsão de músicas de sucesso com base na colaboraçãoAnálisis y predicción de canciones de éxito con reconocimiento de colaboracióninfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesisreponame:Repositório Institucional da UFMGinstname:Universidade Federal de Minas Gerais (UFMG)instacron:UFMGORIGINAL_Disserta__o_Mariana___Vers_o_Final__Collaboration_Aware_Hit_Song_Analysis_and_Prediction.pdf_Disserta__o_Mariana___Vers_o_Final__Collaboration_Aware_Hit_Song_Analysis_and_Prediction.pdfDissertaçãoapplication/pdf19133094https://repositorio.ufmg.br/bitstream/1843/44585/3/_Disserta__o_Mariana___Vers_o_Final__Collaboration_Aware_Hit_Song_Analysis_and_Prediction.pdf968a647073e13732d9ff031427023773MD53CC-LICENSElicense_rdflicense_rdfapplication/rdf+xml; charset=utf-8811https://repositorio.ufmg.br/bitstream/1843/44585/4/license_rdfcfd6801dba008cb6adbd9838b81582abMD54LICENSElicense.txtlicense.txttext/plain; charset=utf-82118https://repositorio.ufmg.br/bitstream/1843/44585/5/license.txtcda590c95a0b51b4d15f60c9642ca272MD551843/445852022-08-25 13:05:35.348oai:repositorio.ufmg.br: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ório InstitucionalPUBhttps://repositorio.ufmg.br/oaiopendoar:2022-08-25T16:05:35Repositório Institucional da UFMG - Universidade Federal de Minas Gerais (UFMG)false
|