In response to the current situation where traditional air defense command information system intelligence processing software can only manually analyze the attribute information of aerial targets. A method based on the random forest decision tree is adopted to perform supervised learning training on target motion parameters, which can effectively improve the accuracy of automatic target type analysis. This technology can solve the problem of difficulty in target identification that relies solely on radar data in traditional combat. Through the data mining method of supervised learning, it effectively improves the accuracy of target classification and identification, making it applicable to the judgment of aircraft types in air defense systems, meeting the needs of commanders for situational analysis and decision-making support.