| Popis: |
This paper builds a mathematical model based on data mining to study the related issues of fraud. This paper constructs a recognition model from two aspects of traditional recognition and intelligent algorithms, uses Logistic model, BP neural network model and decision tree model to identify fraudulent data in manufacturing, and uses confusion matrix to use classification accuracy, recall rate, and F value And accuracy and other indicators to evaluate the accuracy of recognition. Therefore, substitute manufacturing-related data into the recognition model, compare the recognition results of the three models, and select the model with the highest recognition accuracy to predict the sixth year of the manufacturing industry. |
