Integrating phenotypic traits, multivariate statistics, and machine‐learning models to predict yield in Himalayan kiwifruit ( Actinidia deliciosa ) orchards
Journal of the Science of Food and Agriculture · 30.08.2026 · 106 (11) · 6582-6595
DOI: 10.1002/jsfa.70689 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of the Science of Food and Agriculture, online erschienen am 29.04.2026. Verfasst von Shilpa Sharma und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Basic Sciences, College of Forestry, Dr. Yashwant Singh Parmar University of Horticulture and Forestry, Solan, India. Hauptthemen laut MEDLINE: Actinidia, Fruit. Die Arbeit nennt 44 Literaturangaben.
Abstract (Originalsprache)
BACKGROUND Understanding the morphological and physiological determinants of kiwifruit ( Actinidia deliciosa A. Chev.) yield is essential to improve orchard productivity in Himalayan hill agro‐ecosystems. However, most previous studies have focused on individual agronomic factors and have rarely integrated multivariate phenotypic traits with advanced predictive modeling approaches. The aim of this study was to evaluate whether combining multivariate statistical techniques with machine‐learning algorithms could improve identification of key traits influencing yield variability in Himalayan kiwifruit orchards. RESULTS Thirty vegetative, floral, phenological, and fruit biometric traits were recorded from 210 vines of Actinidia deliciosa ‘Allison’ across three major kiwifruit‐growing districts of Himachal Pradesh, India. Descriptive analysis revealed substantial phenotypic variability in canopy vigor, flowering intensity, and fruit morphology. Correlation analysis showed that fruit number per shoot, fruit width, and flowering duration were associated positively with total fruit yield, whereas leaf‐to‐fruit ratio showed a negative association. Principal component analysis identified five components explaining 79.1% of total phenotypic variance, representing canopy vigor, floral load, fruit morphology, and source–sink balance. Among machine‐learning models, random forest achieved the highest predictive accuracy (R 2 = 0.84; RMSE = 6.20 kg), outperforming gradient boosting and support vector regression. Feature importance analysis consistently identified fruits per shoot, fruit width, vine girth, flowering duration, and leaf‐to‐fruit ratio as the most influential predictors. CONCLUSION The integration of multivariate statistical analysis with machine learning provides a robust framework for identifying key trait‐based indicators of yield variability. These findings can support clonal selection, optimized pruning strategies, and site‐specific orchard management to increase productivity in Himalayan kiwifruit systems. © 2026 Society of Chemical Industry.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Basic Sciences, College of Forestry, Dr. Yashwant Singh Parmar University of Horticulture and Forestry, Solan, India.
- Mittal School of Business, Lovely Professional University, Phagwara, India.
- Department of Fruit Science, College of Horticulture, Dr. Yashwant Singh Parmar University of Horticulture and Forestry, Solan, India.
- Department of Forestry, School of Agriculture, Dev Bhoomi Uttarakhand University, Dehradun, India.
ORCID-Profile: Umesh Sharma
Zitieren
APA
Sharma, S., Gupta, R. K., Rana, V. S., & Sharma, U. (2026). Integrating phenotypic traits, multivariate statistics, and machine‐learning models to predict yield in Himalayan kiwifruit ( Actinidia deliciosa ) orchards. Journal of the Science of Food and Agriculture, 106(11), 6582-6595. https://doi.org/10.1002/jsfa.70689
Vancouver
Sharma S, Gupta RK, Rana VS, Sharma U. Integrating phenotypic traits, multivariate statistics, and machine‐learning models to predict yield in Himalayan kiwifruit ( Actinidia deliciosa ) orchards. Journal of the Science of Food and Agriculture. 2026;106(11):6582-6595. doi:10.1002/jsfa.70689
BibTeX
@article{Sharma2026,
author = {Sharma, Shilpa and Gupta, Rakesh Kumar and Rana, Vishal Singh and Sharma, Umesh},
title = {Integrating phenotypic traits, multivariate statistics, and machine‐learning models to predict yield in Himalayan kiwifruit ( Actinidia deliciosa ) orchards},
journal = {Journal of the Science of Food and Agriculture},
year = {2026},
volume = {106},
number = {11},
pages = {6582--6595},
doi = {10.1002/jsfa.70689}
}
EUVIMED