Avoiding the Inherent Limitations in Datasets Used for Measuring Aesthetics When Using a Machine Learning Approach
An important topic in evolutionary art is the development of systems that can mimic the aesthetics decisions made by human begins, e.g., fitness evaluations made by humans using interactive evolution in generative art. This paper focuses on the analysis of several datasets used for aesthetic predict...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2019-01-01
|
| Series: | Complexity |
| Online Access: | http://dx.doi.org/10.1155/2019/4659809 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832567813633474560 |
|---|---|
| author | Adrian Carballal Carlos Fernandez-Lozano Nereida Rodriguez-Fernandez Luz Castro Antonino Santos |
| author_facet | Adrian Carballal Carlos Fernandez-Lozano Nereida Rodriguez-Fernandez Luz Castro Antonino Santos |
| author_sort | Adrian Carballal |
| collection | DOAJ |
| description | An important topic in evolutionary art is the development of systems that can mimic the aesthetics decisions made by human begins, e.g., fitness evaluations made by humans using interactive evolution in generative art. This paper focuses on the analysis of several datasets used for aesthetic prediction based on ratings from photography websites and psychological experiments. Since these datasets present problems, we proposed a new dataset that is a subset of DPChallenge.com. Subsequently, three different evaluation methods were considered, one derived from the ratings available at DPChallenge.com and two obtained under experimental conditions related to the aesthetics and quality of images. We observed different criteria in the DPChallenge.com ratings, which had more to do with the photographic quality than with the aesthetic value. Finally, we explored learning systems other than state-of-the-art ones, in order to predict these three values. The obtained results were similar to those using state-of-the-art procedures. |
| format | Article |
| id | doaj-art-9e49ff49fe3e420f8a45a40697717367 |
| institution | Kabale University |
| issn | 1076-2787 1099-0526 |
| language | English |
| publishDate | 2019-01-01 |
| publisher | Wiley |
| record_format | Article |
| series | Complexity |
| spelling | doaj-art-9e49ff49fe3e420f8a45a406977173672025-02-03T01:00:27ZengWileyComplexity1076-27871099-05262019-01-01201910.1155/2019/46598094659809Avoiding the Inherent Limitations in Datasets Used for Measuring Aesthetics When Using a Machine Learning ApproachAdrian Carballal0Carlos Fernandez-Lozano1Nereida Rodriguez-Fernandez2Luz Castro3Antonino Santos4Computer Science Department, Faculty of Computer Science, University of A Coruña, A Coruña 15071, SpainComputer Science Department, Faculty of Computer Science, University of A Coruña, A Coruña 15071, SpainComputer Science Department, Faculty of Communication Science, University of A Coruña, A Coruña 15071, SpainComputer Science Department, Faculty of Communication Science, University of A Coruña, A Coruña 15071, SpainComputer Science Department, Faculty of Computer Science, University of A Coruña, A Coruña 15071, SpainAn important topic in evolutionary art is the development of systems that can mimic the aesthetics decisions made by human begins, e.g., fitness evaluations made by humans using interactive evolution in generative art. This paper focuses on the analysis of several datasets used for aesthetic prediction based on ratings from photography websites and psychological experiments. Since these datasets present problems, we proposed a new dataset that is a subset of DPChallenge.com. Subsequently, three different evaluation methods were considered, one derived from the ratings available at DPChallenge.com and two obtained under experimental conditions related to the aesthetics and quality of images. We observed different criteria in the DPChallenge.com ratings, which had more to do with the photographic quality than with the aesthetic value. Finally, we explored learning systems other than state-of-the-art ones, in order to predict these three values. The obtained results were similar to those using state-of-the-art procedures.http://dx.doi.org/10.1155/2019/4659809 |
| spellingShingle | Adrian Carballal Carlos Fernandez-Lozano Nereida Rodriguez-Fernandez Luz Castro Antonino Santos Avoiding the Inherent Limitations in Datasets Used for Measuring Aesthetics When Using a Machine Learning Approach Complexity |
| title | Avoiding the Inherent Limitations in Datasets Used for Measuring Aesthetics When Using a Machine Learning Approach |
| title_full | Avoiding the Inherent Limitations in Datasets Used for Measuring Aesthetics When Using a Machine Learning Approach |
| title_fullStr | Avoiding the Inherent Limitations in Datasets Used for Measuring Aesthetics When Using a Machine Learning Approach |
| title_full_unstemmed | Avoiding the Inherent Limitations in Datasets Used for Measuring Aesthetics When Using a Machine Learning Approach |
| title_short | Avoiding the Inherent Limitations in Datasets Used for Measuring Aesthetics When Using a Machine Learning Approach |
| title_sort | avoiding the inherent limitations in datasets used for measuring aesthetics when using a machine learning approach |
| url | http://dx.doi.org/10.1155/2019/4659809 |
| work_keys_str_mv | AT adriancarballal avoidingtheinherentlimitationsindatasetsusedformeasuringaestheticswhenusingamachinelearningapproach AT carlosfernandezlozano avoidingtheinherentlimitationsindatasetsusedformeasuringaestheticswhenusingamachinelearningapproach AT nereidarodriguezfernandez avoidingtheinherentlimitationsindatasetsusedformeasuringaestheticswhenusingamachinelearningapproach AT luzcastro avoidingtheinherentlimitationsindatasetsusedformeasuringaestheticswhenusingamachinelearningapproach AT antoninosantos avoidingtheinherentlimitationsindatasetsusedformeasuringaestheticswhenusingamachinelearningapproach |