Machine Learning Fusion for Crop Yield Estimation: A Comparative Approach

The backbone of India's economy and population is agriculture. Unfortunately, unpredictable weather patterns and outdated farming practices, such as manual surveys and weather-based models, often hinder optimal yield and sustainability. These practices can limit productivity and resilience in the face of changing environmental conditions. The proposed Hybrid-DX model, a powerful combination of the deep learning algorithm DNN and the ensemble method XGBoost, aims to address these issues by integrating advanced data analytics with traditional agricultural practices. This integration enhances decision-making and optimizes yield in the face of various challenges. We compared and evaluated the performance of several algorithms, like Linear Regression, XGBoost, Random Forest, DNN, LSTM, CNN, along with the proposed Hybrid-DX. The insights from our research can assist farmers in making informed decisions by providing better predictions. With the proposed Hybrid-DX model, farmers can overcome the challenges of agriculture and ensure a brighter future for themselves and the economy.