Smart Farming Sensor for Yield Prediction Dataset

1) Data Introduction

• The Smart Farming Sensor Data for Yield Prediction Dataset is a structured agricultural dataset collected from 500 smart farms located in regions such as India, the USA, and Africa. It contains IoT-based crop cultivation environment data, including soil moisture, temperature, humidity, pH, NDVI, pesticide usage, irrigation method, and more. The target variable is crop_disease_status, categorized into four levels: None, Mild, Moderate, and Severe.

2) Data Utilization

(1) Characteristics of the Smart Farming Sensor Data for Yield Prediction Dataset: • The dataset includes 22 columns such as farm ID, location, environmental sensor values, crop type, cultivation period, NDVI, etc. The crop_disease_status column serves as a multi-class label indicating the crop's disease status in four severity levels. • With time-series-like IoT data and labeled disease conditions, the dataset is suitable for training advanced crop disease prediction models. (2) Applications of the Smart Farming Sensor Data for Yield Prediction Dataset: • Crop Disease Classification Model Training: Machine learning or deep learning-based classification models can be developed to predict the disease status of crops based on climatic and cultivation environment data. • Smart Farm Disease Management Systems: The dataset can be used to design systems for early disease detection and severity estimation, optimizing pesticide usage, preventing disease outbreaks, and supporting crop protection strategies.