Classification Method of Big Data for Non-formatted Equipment Combat Training Support

With the development of network technology and the continuous expansion of its application in the military, there is a large amount of complex and non-formatted data in daily combat training support and aviation ammunition equipment storage and transportation, which can be processed or unprocessed. These data structures are irregular or incomplete, without a predetermined data model, and are not convenient to represent using two-dimensional database logic. This article starts from the classification problem of non formatted aviation ammunition equipment training support big data, defines the form and characteristics of formatted training support big data, draws on and improves convolutional neural networks to improve the accuracy of text and image non formatted data classification. By establishing an information entropy mathematical model, non-formatted monitoring data is classified, greatly improving the efficiency of non formatted training support big data classification.