Efficient imbalanced data set classification method

Abstract

The invention discloses an efficient imbalanced data set classification method, which includes the steps of adding boundary samples and isolated points into consideration based on a traditional SMOTE method, so as to acquire an approximately balanced data set; and then making designs on the type of data based on a sub space selection and tree model scheme of ensemble random forests, to clearly classify two types of data. In this way, the accuracy and recall ratio are improved, the result is close to the actual situation, and the invention can be applied to the actual industry analysis.

Description

technical field [0001] The invention relates to the technical field of data processing, in particular to an efficient method for classifying unbalanced data sets. Background technique [0002] The classification problem is one of the most important problems in data analysis, and the actual data sets often have the problem of unbalanced number of categories. For unbalanced data sets, traditional classification methods such as decision trees, SVM, and Bayesian networks are less effective because traditional algorithms contain assumptions about sample balance. When the data set is unbalanced, the minority class samples are not easy to be identified because the multi-class samples occupy an absolute advantage in the classification model training process. Data classification problems are usually concerned with minority samples, such as the prediction of telecommunication customer churn in practical applications, abnormal detection of credit card transactions, network intrusion p...

AI Technical Summary

Problems solved by technology

Although the traditional random forest algorithm guarantees randomness, it does not take into account the particularity of the unbalanced data set, and does not give more resources to the minority samples. When forming a decision forest, all trees are selected to be included in the voting tree, which is not conducive to unbalanced Classification of Datasets

Images