Characterising Confounding Effects in Music Classification Experiments through Interventions
We address the problem of confounding in the design of music classification experiments, that is, the inability to distinguish the effects of multiple potential influencing variables in the measurements. Confounding affects the validity of conclusions at many levels, and so must be properly accounte...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Ubiquity Press
2019-08-01
|
| Series: | Transactions of the International Society for Music Information Retrieval |
| Subjects: | |
| Online Access: | https://transactions.ismir.net/articles/24 |
| _version_ | 1818973273667403776 |
|---|---|
| author | Francisco Rodríguez-Algarra Bob L. Sturm Simon Dixon |
| author_facet | Francisco Rodríguez-Algarra Bob L. Sturm Simon Dixon |
| author_sort | Francisco Rodríguez-Algarra |
| collection | DOAJ |
| description | We address the problem of confounding in the design of music classification experiments, that is, the inability to distinguish the effects of multiple potential influencing variables in the measurements. Confounding affects the validity of conclusions at many levels, and so must be properly accounted for. We propose a procedure for characterising effects of confounding in the results of music classification experiments by creating regulated test conditions through interventions in the experimental pipeline, including a novel resampling strategy. We demonstrate this procedure on the GTZAN genre collection, which is known to give rise to confounding effects. |
| first_indexed | 2024-12-20T15:21:33Z |
| format | Article |
| id | doaj.art-cac31a26cc9141a3ab5de2729054bc24 |
| institution | Directory Open Access Journal |
| issn | 2514-3298 |
| language | English |
| last_indexed | 2024-12-20T15:21:33Z |
| publishDate | 2019-08-01 |
| publisher | Ubiquity Press |
| record_format | Article |
| series | Transactions of the International Society for Music Information Retrieval |
| spelling | doaj.art-cac31a26cc9141a3ab5de2729054bc242022-12-21T19:36:01ZengUbiquity PressTransactions of the International Society for Music Information Retrieval2514-32982019-08-012110.5334/tismir.2415Characterising Confounding Effects in Music Classification Experiments through InterventionsFrancisco Rodríguez-Algarra0Bob L. Sturm1Simon Dixon2Centre for Digital Music, School of Electronic Engineering and Computer Science, Queen Mary University of LondonKTH Royal Institute of Technology, StockholmCentre for Digital Music, School of Electronic Engineering and Computer Science, Queen Mary University of LondonWe address the problem of confounding in the design of music classification experiments, that is, the inability to distinguish the effects of multiple potential influencing variables in the measurements. Confounding affects the validity of conclusions at many levels, and so must be properly accounted for. We propose a procedure for characterising effects of confounding in the results of music classification experiments by creating regulated test conditions through interventions in the experimental pipeline, including a novel resampling strategy. We demonstrate this procedure on the GTZAN genre collection, which is known to give rise to confounding effects.https://transactions.ismir.net/articles/24Music ClassificationEvaluationExperimental DesignConfoundingResampling |
| spellingShingle | Francisco Rodríguez-Algarra Bob L. Sturm Simon Dixon Characterising Confounding Effects in Music Classification Experiments through Interventions Transactions of the International Society for Music Information Retrieval Music Classification Evaluation Experimental Design Confounding Resampling |
| title | Characterising Confounding Effects in Music Classification Experiments through Interventions |
| title_full | Characterising Confounding Effects in Music Classification Experiments through Interventions |
| title_fullStr | Characterising Confounding Effects in Music Classification Experiments through Interventions |
| title_full_unstemmed | Characterising Confounding Effects in Music Classification Experiments through Interventions |
| title_short | Characterising Confounding Effects in Music Classification Experiments through Interventions |
| title_sort | characterising confounding effects in music classification experiments through interventions |
| topic | Music Classification Evaluation Experimental Design Confounding Resampling |
| url | https://transactions.ismir.net/articles/24 |
| work_keys_str_mv | AT franciscorodriguezalgarra characterisingconfoundingeffectsinmusicclassificationexperimentsthroughinterventions AT boblsturm characterisingconfoundingeffectsinmusicclassificationexperimentsthroughinterventions AT simondixon characterisingconfoundingeffectsinmusicclassificationexperimentsthroughinterventions |