A Machine Learning Model for Crop Yield Prediction

Abstract

India is one of the largest global agricultural countries. Instead of that, the farmers face a lot of challenges in crop production, despite its crucial economic value. These issues have an immediate impact on the livelihoods of farmers making it imperative to enhance agricultural output to meet the global food demand. To enhance the productivity, the farmers need to determine which crop are best suited for their specific fields. A proposed solution is the crop recommendation system which forecasts the idea type of crop for cultivation by monitoring various parameters and constraints. Previous research has focused on numerous agricultural issues such as cost, water, and soil management fruit maturity, plant diseases, soil mining, and other farm sectors. These suggestions are dependable and based on, measurable evidence including soil pH, humidity, rainfall, temperature, and soil nutrient concentration (N, P, and K). However, they lacked a thorough evaluation on machine learning role in crop recommendation. By using the Support Vector Machine (SVM) approach, the proposed model can determine the most profitable crop under some current conditions. The model is first trained on a dataset using pre-recorded values, enabling it to predict the crop type independently. Selecting the appropriate crop type can enhance total yield, promoting agricultural sustainability and addressing the global food demand. Incorporating real-time data from IoT sensors, satellite images improve the efficiency of the model, using advanced machine learning techniques such as SVM helps in precise crop prediction, assisting in optimizing resource usage, also increasing the economic viability of the farmers.

Keywords: Machine Learning Model, Yield Prediction, Support Vector Machine (SVM), Crop Recommendation Model, Agricultural Factors

View PDF