[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-126125-en":3,"doc-seo-126125-105":31,"detail-sidebar-cat-0-en-105":93},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":28,"seo_description":14,"update_tm":29,"read_time":30},126125,5909887254083,"Miles","https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c",8,"Research & Report","Bagging and boosting machine learning algorithms for modelling sensory perception from simple chemical variables - Wine mouthfeel as a case study","Aiming to predict sensory properties from chemical data, the application of bagging and boosting machine learning algorithms was investigated for modelling red wine mouthfeel using simple chemical measurements. Fifteen Australian winemakers described mouthfeel attributes for 30 commercial red wines via rate-all-that-apply sensory profiling. Linear sweep voltammetry and excitation-emission matrix/absorbance signals were collected in parallel, then analysed with PCA for interpretability and supervised regression with random forest and XGBoost. PCA revealed four independent sensory dimensions, while RF and XGBoost outperformed PLS on validated tests, achieving over 80% accuracy and benefiting from low computational cost and overfitting control.","Food Quality and Preference 129 (2025) 105494  \nContents lists available at ScienceDirect  \nFood Quality and Preference  \njournal [homepage:](homepage: www.elsevier.com/locate/foodqual)[ www.elsevier.com/locate/foodqual](homepage: www.elsevier.com/locate/foodqual)  \n| Bagging and boosting machine learning algorithms for modelling sensory perception from simple chemical variables: Wine mouthfeel as a case study María-Pilar S´aenz-Navajasa,*, Chelo Ferreira b, Susan E.P. Bastianc, David W. Jeffery c\u003Cbr>a Instituto de Ciencias de la Vid y del Vino (Universidad de La Rioja-Consejo Superior de Investigaciones Científicas-Gobierno de La Rioja), Departamento de Enología, Logro˜no, La Rioja, Spain\u003Cbr>b Instituto Universitario de Matem´aticas y Aplicaciones (IUMA-UNIZAR), Universidad de Zaragoza, c/ Pedro Cerbuna 12, 50009 Zaragoza, Spain\u003Cbr>c School of Agriculture, Food and Wine, and Waite Research Institute, The University of Adelaide, PMB 1, Glen Osmond, South Australia 5064, Australia |  |  |  |\n| --- | --- | --- | --- |\n| A R T I C L E I N F O |  | A B S T R A C T |  |\n| Keywords: Voltammetry Random Forest Regression XGBoost Mouthfeel Spectrofluorometry |  | Aiming to predict sensory properties from chemical data, the application of bagging and boosting machine learning (ML) algorithms was comprehensively investigated and applied to modelling of red wine mouthfeel from simple chemical measurements. A panel of 15 Australian winemakers described the mouthfeel properties of a total of 30 commercial red wines from Australia and Spain using rate-all-that-apply sensory methodology. In parallel, linear sweep voltammetry signals and excitation-emission matrix (EEM) and absorbance data were acquired for the wines. Data were analysed following unsupervised statistical strategies including principal component analysis (PCA with varimax rotation) to simplify the interpretation of sensory variables, along with supervised regression models based on ML, namely random forest (RF) and extreme gradient boosting (XGBoost). PCA results showed that four independent and uncorrelated sensory dimensions mainly related to perceptions of‘drying’, ‘full body’, ‘velvety’, and ‘gummy’ differentiated among the wines. The RF and XGBoost algorithms yielded superior validated regression models compared to classical PLS modelling. The ML algorithms exhibited strong predictive performance on test data, with an average value exceeding 80 % accuracy for any of the three sets of chemical variables employed. Although XGBoost provided slightly better models, the low computational effort required by RF is advantageous. Key variables included in the models are discussed along with the importance of controlling overfitting. Overall, absorbance, voltammetric or EEM signals coupled with RF or XGBoost algorithms are presented as cheap, easy-to-use, and rapid approaches to predicting sensory properties from chemical signals in complex matrices such as wine. |  |\n\n1. Introduction  \nIn recent years, new technologies have had a sudden and significant impact on disciplines related to food sciences. In particular, the fields of data science and machine learning (ML) are facilitating the development of novel paradigms pertaining to data modelling (Zatsu et al., 2024). Compared to classical modelling that associates dependent variables with explanatory variables, the emerging state ofthe art allows for much more when dealing with complex datasets. Consider the extensively utilised predictive models developed with partial least squares (PLS) regression, which have endeavoured to model human sensory perceptions from physical or chemical data for some time (Kvaal, Wold, Indahl, Baardseth, & Næs, 1998; Niimi et al., 2020), but not without controversy. The number of data points typically used to describe the sensory  \nspace of foods is often limited, with a relatively high number of predictor variables (Lee, Liong, & Jemain, 2018). This can result in an uncontrolled overfitting of the PLS","cbCaioTV6cgdyBfQ","https://ap.wps.com/l/cbCaioTV6cgdyBfQ","pdf",2803831,7,1,13,"English","en",105,"# Introduction\n## Machine learning as an alternative to classical modelling\n# Materials and methods\n## Sensory data collection\n## Chemical measurements and preprocessing\n## Unsupervised and supervised modelling approaches\n# Results\n## Sensory dimensions from PCA\n## Predictive performance of regression models\n# Discussion\n## Key variables and overfitting considerations\n## Practical implications for wine sensory prediction","[{\"question\":\"What sensory outcome was modelled in this study and from what data sources?\",\"answer\":\"The study modelled red wine mouthfeel. Sensory properties were obtained using rate-all-that-apply methodology, while chemical inputs came from linear sweep voltammetry and excitation-emission matrix plus absorbance data.\"},{\"question\":\"Which machine learning algorithms were used for regression and how did they compare with PLS?\",\"answer\":\"Random forest and XGBoost were used for supervised regression. Both produced superior validated regression models compared with classical partial least squares (PLS) modelling, with test accuracy above 80%.\"},{\"question\":\"What did PCA reveal about the sensory structure of the wines?\",\"answer\":\"PCA identified four independent, largely uncorrelated sensory dimensions. These mainly corresponded to perceptions of drying, full body, velvety, and gummy.\"}]","Bagging and boosting machine learning algorithms for modelling sensory perception from simple chemical variables - Wine mouthfeel as a case study | PDF",1785903291,33,{"code":4,"msg":32,"data":33},"ok",{"site_id":25,"language":24,"slug":34,"title":13,"keywords":35,"description":14,"schema_data":36,"social_meta":88,"head_meta":90,"extra_data":92,"updated_unix":29},"bagging-and-boosting-machine-learning-algorithms-for-modelling-sensory-perception-from-simple-chemical-variables-wine-mouthfeel-as-a-case-study","",{"@graph":37,"@context":87},[38,55,70],{"@type":39,"itemListElement":40},"BreadcrumbList",[41,45,49,52],{"item":42,"name":43,"@type":44,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":46,"name":47,"@type":44,"position":48},"https://docshare.wps.com/document/","Document",2,{"item":50,"name":12,"@type":44,"position":51},"https://docshare.wps.com/document/research-report/",3,{"item":53,"name":13,"@type":44,"position":54},"https://docshare.wps.com/document/bagging-and-boosting-machine-learning-algorithms-for-modelling-sensory-perception-from-simple-chemical-variables-wine-mouthfeel-as-a-case-study/126125/",4,{"url":53,"name":13,"@type":56,"author":57,"headline":13,"publisher":59,"fileFormat":62,"inLanguage":24,"description":14,"dateModified":63,"datePublished":64,"encodingFormat":62,"isAccessibleForFree":65,"interactionStatistic":66},"DigitalDocument",{"name":9,"@type":58},"Person",{"url":42,"name":60,"@type":61},"DocShare","Organization","application/pdf","2026-08-25","2026-08-05",true,{"@type":67,"interactionType":68,"userInteractionCount":20},"InteractionCounter",{"@type":69},"ViewAction",{"@type":71,"mainEntity":72},"FAQPage",[73,79,83],{"name":74,"@type":75,"acceptedAnswer":76},"What sensory outcome was modelled in this study and from what data sources?","Question",{"text":77,"@type":78},"The study modelled red wine mouthfeel. Sensory properties were obtained using rate-all-that-apply methodology, while chemical inputs came from linear sweep voltammetry and excitation-emission matrix plus absorbance data.","Answer",{"name":80,"@type":75,"acceptedAnswer":81},"Which machine learning algorithms were used for regression and how did they compare with PLS?",{"text":82,"@type":78},"Random forest and XGBoost were used for supervised regression. Both produced superior validated regression models compared with classical partial least squares (PLS) modelling, with test accuracy above 80%.",{"name":84,"@type":75,"acceptedAnswer":85},"What did PCA reveal about the sensory structure of the wines?",{"text":86,"@type":78},"PCA identified four independent, largely uncorrelated sensory dimensions. These mainly corresponded to perceptions of drying, full body, velvety, and gummy.","https://schema.org",{"og:url":53,"og:type":89,"og:title":13,"og:site_name":60,"og:description":14},"article",{"robots":91,"canonical":53},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":94},[95,99,103,107,112,117,121,124,129,132,136],{"id":21,"doc_module":4,"doc_module_name":47,"category_name":96,"show_sort_weight":97,"slug":98},"Story & Novel",90,"story-novel",{"id":48,"doc_module":4,"doc_module_name":47,"category_name":100,"show_sort_weight":101,"slug":102},"Literature",80,"literature",{"id":54,"doc_module":4,"doc_module_name":47,"category_name":104,"show_sort_weight":105,"slug":106},"Exam",70,"exam",{"id":108,"doc_module":4,"doc_module_name":47,"category_name":109,"show_sort_weight":110,"slug":111},5,"Comic",60,"comic",{"id":113,"doc_module":4,"doc_module_name":47,"category_name":114,"show_sort_weight":115,"slug":116},6,"Technology",50,"technology",{"id":20,"doc_module":4,"doc_module_name":47,"category_name":118,"show_sort_weight":119,"slug":120},"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":47,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":125,"doc_module":4,"doc_module_name":47,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":47,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":133,"doc_module":4,"doc_module_name":47,"category_name":134,"show_sort_weight":133,"slug":135},10,"Lifestyle","lifestyle",{"id":137,"doc_module":4,"doc_module_name":47,"category_name":138,"show_sort_weight":108,"slug":139},19,"General","general"]