Data Classification and Demand Prediction Methods Based on Semi-Supervised Agricultural Machinery Spare Parts Data

Because of the continuous improvement technology, mechanization has emerged in various fields. Due to different suitable seasons for growth agricultural plants, faces problems from other industries. That is, machinery and equipment may be used frequently a period time, or idle long time. This leads aging no longer becoming regular, maintenance time spare parts is not fixed, number stored warehouse cannot too large occupy funds, small meet needs, so prediction become particularly important. lack information, difficulty labeling, imbalance positive negative sample classification, this paper semi-supervised learning algorithm solve problem data classification. In order forecast demand machinery, compared IPSO-BP neural network BP algorithm. It was found that error between predicted value actual met accuracy requirements.