Weighted sampling and forecast model using data of pig farming management system

Finding high predictive modeling has been an important task in agriculture. Recently, a machine learning technique using big data achieved high prediction model performance. Following this trend, the goal of this study was to predict the number of pig shipments by using production data collected from pig management systems. This study used weighted sampling to prevent system user sample bias and inconsistency and compared the performance of a model that used a machine learning technique to apply a weighted value and one that did not apply a weighted value. The results indicated that the model that used the machine learning technique to apply a weighted value had higher prediction performance than the model without the applied weighted value.