Sustainable crop recommendation system using seasonally adaptive recursive spectral convolutional neural network for responsible agricultural production
Integrating deep learning in agriculture offers new pathways for sustainable crop recommendation and yield prediction, aiding decision-making processes such as optimal planting times and crop selection. Agriculture is a vital sector in Tamil Nadu, where environmental factors, including Humidity, rai...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2025-12-01
|
| Series: | Geomatics, Natural Hazards & Risk |
| Subjects: | |
| Online Access: | https://www.tandfonline.com/doi/10.1080/19475705.2025.2509619 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Integrating deep learning in agriculture offers new pathways for sustainable crop recommendation and yield prediction, aiding decision-making processes such as optimal planting times and crop selection. Agriculture is a vital sector in Tamil Nadu, where environmental factors, including Humidity, rainfall, sunlight, soil type, and temperature, significantly influence crop yield. However, selecting the ideal crop based on season remains challenging, often resulting in inefficient resource use and reduced yield. This research proposes a novel Sustainable Crop Recommendation System using a Recursive Spectral Convolutional Neural Network (RSCN2) to optimize the resource to improve the Sustainable Crop Recommendation System. The model incorporates feature subsets derived from high-impact factors through minimal Redundancy and Maximum Weight (mRmW) and Recursive Fisher Score Feature Selection (RFSFS), eliminating irrelevant data and enhancing accuracy. Seasonal crop suitability is assessed through multi-scale clustering of key attributes, and final classification employs the RSCN2 model with a Softmax activation function to provide precise crop recommendations by season. The evaluation shows that this method achieves superior precision, Recall, and f-measure metrics over traditional approaches, supporting sustainable agricultural practices by promoting efficient resource use and minimizing environmental impact. |
|---|---|
| ISSN: | 1947-5705 1947-5713 |