An innovative artificial neural network model for smart crop prediction using sensory network based soil data.

Area of Science:

• Agricultural Science
• Computer Science
• Data Science

Background:

• Agricultural productivity is crucial for economic growth but is hampered by suboptimal crop selection based on environmental and soil factors.
• Accurate crop prediction is vital for maximizing yield and farmer income, requiring consideration of diverse environmental and soil parameters.
• Recommender systems (RS) and machine learning (ML) offer promising avenues for enhancing agricultural decision-making.

Purpose of the Study:

• To develop an innovative artificial neural network (ANN) based crop prediction system (CPS) to assist farmers in selecting the most suitable crops.
• To leverage sensor-based soil data, including nitrogen, phosphorus, potassium, temperature, humidity, pH, rainfall, electrical conductivity, and soil texture, for accurate crop forecasting.
• To validate the proposed CPS using machine learning techniques and hyperparameter optimization for enhanced reliability.

Main Methods:

• Development of a crop prediction system (CPS) utilizing an artificial neural network (ANN) architecture.
• Collection and analysis of sensor-based soil data encompassing key environmental and chemical parameters.
• Implementation in Python, with performance evaluation using accuracy, precision, recall, and F1-score, including hyperparameter optimization.

Main Results:

• The proposed artificial neural network (ANN) based crop prediction system (CPS) achieved a high accuracy of 99% on both real-time and benchmark datasets.
• The system effectively utilizes sensor-based soil data and environmental factors for precise crop recommendations.
• Hyperparameter optimization further refined the model's learning methods, ensuring robust performance.

Conclusions:

• The developed crop prediction system (CPS) significantly aids farmers in making informed crop selection decisions.
• The system's high accuracy translates to improved crop productivity and increased profitability for farmers.
• This data-driven approach represents a valuable advancement in precision agriculture and agricultural economics.