Multi-target prediction of wheat flour quality parameters with near infrared spectroscopy
Near Infrared (NIR) spectroscopy is an analytical technology widely used for the non-destructive characterisation of organic samples, considering both qualitative and quantitative attributes. In the present study, the combination of Multi-target (MT) prediction approaches and Machine Learning algori...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2020-06-01
|
| Series: | Information Processing in Agriculture |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2214317318304554 |
| _version_ | 1797765406144856064 |
|---|---|
| author | Sylvio Barbon Junior Saulo Martielo Mastelini Ana Paula A.C. Barbon Douglas Fernandes Barbin Rosalba Calvini Jessica Fernandes Lopes Alessandro Ulrici |
| author_facet | Sylvio Barbon Junior Saulo Martielo Mastelini Ana Paula A.C. Barbon Douglas Fernandes Barbin Rosalba Calvini Jessica Fernandes Lopes Alessandro Ulrici |
| author_sort | Sylvio Barbon Junior |
| collection | DOAJ |
| description | Near Infrared (NIR) spectroscopy is an analytical technology widely used for the non-destructive characterisation of organic samples, considering both qualitative and quantitative attributes. In the present study, the combination of Multi-target (MT) prediction approaches and Machine Learning algorithms has been evaluated as an effective strategy to improve prediction performances of NIR data from wheat flour samples. Three different Multi-target approaches have been tested: Multi-target Regressor Stacking (MTRS), Ensemble of Regressor Chains (ERC) and Deep Structure for Tracking Asynchronous Regressor Stack (DSTARS). Each one of these techniques has been tested with different regression methods: Support Vector Machine (SVM), Random Forest (RF) and Linear Regression (LR), on a dataset composed of NIR spectra of bread wheat flours for the prediction of quality-related parameters. By combining all MT techniques and predictors, we obtained an improvement up to 7% in predictive performance, compared with the corresponding Single-target (ST) approaches. The results support the potential advantage of MT techniques over ST techniques for analysing NIR spectra. |
| first_indexed | 2024-03-12T20:09:43Z |
| format | Article |
| id | doaj.art-764cc66afda64805a95677d6ec992d11 |
| institution | Directory Open Access Journal |
| issn | 2214-3173 |
| language | English |
| last_indexed | 2024-03-12T20:09:43Z |
| publishDate | 2020-06-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Information Processing in Agriculture |
| spelling | doaj.art-764cc66afda64805a95677d6ec992d112023-08-02T01:46:53ZengElsevierInformation Processing in Agriculture2214-31732020-06-0172342354Multi-target prediction of wheat flour quality parameters with near infrared spectroscopySylvio Barbon Junior0Saulo Martielo Mastelini1Ana Paula A.C. Barbon2Douglas Fernandes Barbin3Rosalba Calvini4Jessica Fernandes Lopes5Alessandro Ulrici6Computer Science Department, Londrina State University (UEL), Londrina 86057-970, Brazil; Corresponding author.Computer Science Department, Londrina State University (UEL), Londrina 86057-970, BrazilAnimal Science Department, Londrina State University (UEL), Londrina 86057-970, BrazilDepartment of Food Engineering, Campinas State University (UNICAMP), Campinas 13083-862, BrazilDepartment of Life Science, University of Modena and Reggio Emilia (UNIMORE), 42122 Reggio Emilia, ItalyComputer Science Department, Londrina State University (UEL), Londrina 86057-970, BrazilDepartment of Life Science, University of Modena and Reggio Emilia (UNIMORE), 42122 Reggio Emilia, ItalyNear Infrared (NIR) spectroscopy is an analytical technology widely used for the non-destructive characterisation of organic samples, considering both qualitative and quantitative attributes. In the present study, the combination of Multi-target (MT) prediction approaches and Machine Learning algorithms has been evaluated as an effective strategy to improve prediction performances of NIR data from wheat flour samples. Three different Multi-target approaches have been tested: Multi-target Regressor Stacking (MTRS), Ensemble of Regressor Chains (ERC) and Deep Structure for Tracking Asynchronous Regressor Stack (DSTARS). Each one of these techniques has been tested with different regression methods: Support Vector Machine (SVM), Random Forest (RF) and Linear Regression (LR), on a dataset composed of NIR spectra of bread wheat flours for the prediction of quality-related parameters. By combining all MT techniques and predictors, we obtained an improvement up to 7% in predictive performance, compared with the corresponding Single-target (ST) approaches. The results support the potential advantage of MT techniques over ST techniques for analysing NIR spectra.http://www.sciencedirect.com/science/article/pii/S2214317318304554Random forestSupport Vector MachineNear-infrared spectroscopyMachine learningMTRSDSTARS |
| spellingShingle | Sylvio Barbon Junior Saulo Martielo Mastelini Ana Paula A.C. Barbon Douglas Fernandes Barbin Rosalba Calvini Jessica Fernandes Lopes Alessandro Ulrici Multi-target prediction of wheat flour quality parameters with near infrared spectroscopy Information Processing in Agriculture Random forest Support Vector Machine Near-infrared spectroscopy Machine learning MTRS DSTARS |
| title | Multi-target prediction of wheat flour quality parameters with near infrared spectroscopy |
| title_full | Multi-target prediction of wheat flour quality parameters with near infrared spectroscopy |
| title_fullStr | Multi-target prediction of wheat flour quality parameters with near infrared spectroscopy |
| title_full_unstemmed | Multi-target prediction of wheat flour quality parameters with near infrared spectroscopy |
| title_short | Multi-target prediction of wheat flour quality parameters with near infrared spectroscopy |
| title_sort | multi target prediction of wheat flour quality parameters with near infrared spectroscopy |
| topic | Random forest Support Vector Machine Near-infrared spectroscopy Machine learning MTRS DSTARS |
| url | http://www.sciencedirect.com/science/article/pii/S2214317318304554 |
| work_keys_str_mv | AT sylviobarbonjunior multitargetpredictionofwheatflourqualityparameterswithnearinfraredspectroscopy AT saulomartielomastelini multitargetpredictionofwheatflourqualityparameterswithnearinfraredspectroscopy AT anapaulaacbarbon multitargetpredictionofwheatflourqualityparameterswithnearinfraredspectroscopy AT douglasfernandesbarbin multitargetpredictionofwheatflourqualityparameterswithnearinfraredspectroscopy AT rosalbacalvini multitargetpredictionofwheatflourqualityparameterswithnearinfraredspectroscopy AT jessicafernandeslopes multitargetpredictionofwheatflourqualityparameterswithnearinfraredspectroscopy AT alessandroulrici multitargetpredictionofwheatflourqualityparameterswithnearinfraredspectroscopy |