A Multi-Criteria Decision Support Framework for Agricultural Fuel Demand Forecasting: Comparative Analysis of Machine Learning Approaches with Domain-Specific Feature Engineering

View Full Paper

Ask AI

Bookmark

Share

Key Points

The research aims to develop a framework for forecasting agricultural fuel demand using machine learning techniques and feature engineering.
Utilized real operational data from Tarnet A.S., including 21,187 daily transactions from agricultural cooperatives.
Developed an Agricultural Fuel Demand Feature Taxonomy categorizing 844 features into five groups.
Evaluated eight machine learning algorithms based on criteria like accuracy, statistical significance, and performance stability.
XGBoost algorithm achieved 78.01% classification accuracy, indicating strong predictive power.
Statistical tests confirmed the superiority of XGBoost over other algorithms, with significant results from the Friedman test.

We only have this paper's title and publication details.

Comparative analysis demonstrates improved fuel demand forecasting in agriculture, suggesting effective machine learning applications.

No takes yet. Share an insight, caveat, or question.

Çoban et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark

Share