Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{180058,
        author = {Imran Ahmad and Rohit Khubnani and Saurabh Patle and Saurabh Madan and DS Karthik},
        title = {Saffron Crop Yield Prediction Using XGBoost & SVM Hybrid Machine Learning Models},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {1},
        pages = {354-363},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=180058},
        abstract = {Saffron (Crocus sativus L.), one of the most 
valuable spices, is very susceptible to the environment, 
cultivation practices, and soil properties. Traditional 
yield forecasting methods sometimes fail to capture the 
complex, non-linear dynamics influencing saffron 
production. To tackle this challenge, this study proposes 
a hybrid machine learning method that combines 
Extreme Gradient Boosting (XGBoost) and Support 
Vector Machine (SVM) to reliably estimate saffron 
production. The proposed model utilises XGBoost's 
ability to handle large datasets and capture feature 
importance, as well as SVM to increase forecast 
accuracy in high-dimensional domains. The dataset 
includes important climatic parameters such as soil 
type, temperature, precipitation, and historical yield 
records. Mean Absolute Error (MAE), Mean Absolute 
Percentage Error (MAPE), and Root Mean Squared 
Error (RMSE) are used to compare the model's results 
with those of the independent SVM and XGBoost 
models in order to evaluate its performance. 
Experimental results show that the hybrid XGBoost
SVM model outperforms individual models and 
generates more reliable yield estimations. This study 
shows how farmers can improve saffron production 
techniques and more effectively allocate resources by 
using advanced machine learning techniques in 
precision agriculture. By encouraging data-driven 
decision-making, the outcomes guarantee increased 
production and financial efficiency in sustainable 
saffron cultivation. We gathered and combined several 
publicly accessible environmental and agronomic 
datasets because there aren't many saffron yield 
datasets. We extracted and standardized important 
factors including rainfall, temperature, soil pH, 
daylight hours, and altitude using sophisticated feature 
engineering to produce a single dataset. Furthermore, 
we examined each feature's weight to ascertain how 
each one affected the forecast of saffron yield.},
        keywords = {Support Vector Machine (SVM), XGBoost,  hybrid model, machine learning (ML), and saffron yield  forecast.},
        month = {May},
        }