Enhancing Agri-Sustainability: Multimodal Recommendations for Demand-Supply Alignment

The objective of the research is to develop an innovative approach for supply-demand balancing for the agriculture goods for sustainability of farmers. The proposed model considers the Demand-Supply in market to determine which crop farmer should grow according to their soil profile and type to align Demand-Supply in the market. The system uses an advanced predictive analytics approach and recommends the personalized crops based on textual and image input by accounting for characteristics such as soil composition, soil type, climate data such as rainfall, temperature etc., and market trends. The developed predictive system uses the multimodal machine learning algorithm comprises with random forest, and CNN model. Using multimodal machine learning, the system recommends the top five crops to farmers for their specific field based on input parameters, allowing farmers to improve crop selection, develop sustainable farming practices, and achieve demand-supply alignment in the agriculture sector. The multimodal hybrid model achieves 95% crop forecast accuracy, ensuring the model's versatility.