EUVIMEDEUVIMED entdecken
Crossref

AGRIPRODNET: A FRAMEWORK FOR PREDICTING AGRICULTURE PRODUCTIVITY FACTORS USING AGRO-CLIMATIC DATA

Acta Innovations · 30.10.2026 · 64 (1s)

DOI: 10.62441/actainnovations.vi.2772 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Acta Innovations, online erschienen am 30.10.2026.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Agricultural productivity is significantly influenced by environmental factors such as relative humidity, rainy days, sunshine duration, evaporation, temperature, wind speed, rainfall, and vapor pressure. These factors play an important role in deciding the harvesting of a crop along with its management, marketing, safety, and storage. The primary focus of the existing methodologies is on determining the suitability of crop yield. Moreover, the existing methods often struggle to capture the complex interactions between various agro-climatic factors in the context of agricultural productivity. To address these challenges, an AgriProdNet method has been proposed that integrates heterogeneous machine learning and deep learning models, including random forest, gradient boosting, AdaBoost, K-Nearest Neighbor, support vector regressor, XG-boost, customized convolutional neural network, artificial neural network, gated recurrent unit, bidirectional long short memory, transformer, and stacked autoencoder within a stacked ensemble framework. In this architecture, the predictions generated from individual base learners are aggregated and optimized using a ridge regression–based meta-learner. It also reduces overfitting and improves generalization capability. Furthermore, the Mish activation function has been employed at the dense layer of the DL-regressor for effective non-linear transformation and feature learning. The performance of the proposed AgriProdNet method is evaluated on a multi-year agro-climatic dataset comprising 360 observations and 11 major environmental features. The comparative quantitative assessment illustrates that the proposed method outperforms the existing methodologies by achieving an R² and MSE of 0.9748 and 0.0004, respectively. Furthermore, feature-importance analysis and Shapley Additive Explanations validate that minimum temperature has the strongest influence on agricultural productivity.

Zitieren

APA

(2026). AGRIPRODNET: A FRAMEWORK FOR PREDICTING AGRICULTURE PRODUCTIVITY FACTORS USING AGRO-CLIMATIC DATA. Acta Innovations, 64(1s). https://doi.org/10.62441/actainnovations.vi.2772

Vancouver

AGRIPRODNET: A FRAMEWORK FOR PREDICTING AGRICULTURE PRODUCTIVITY FACTORS USING AGRO-CLIMATIC DATA. Acta Innovations. 2026;64(1s). doi:10.62441/actainnovations.vi.2772

BibTeX
@article{euvimed2026,
  title = {AGRIPRODNET: A FRAMEWORK FOR PREDICTING AGRICULTURE PRODUCTIVITY FACTORS USING AGRO-CLIMATIC DATA},
  journal = {Acta Innovations},
  year = {2026},
  volume = {64},
  number = {1s},
  doi = {10.62441/actainnovations.vi.2772}
}
Kostenlos starten →