Category weight combined integrated learning classifying method

Abstract

The invention relates to a category weight combined integrated learning classifying method. The method comprises the steps of preprocessing original data; converting into data formats that can be processed by the classification method so as to obtain a training data set and a data set to be classified; initializing the training data set sample weight; re-iterating and training M base classifiers according to the training data set and the sample weight thereof; calculating the category weight; integrating all base classifies; classifying the data set to be classified through a determining classifier according to the category weight; storing the classifying result into a file to obtain classification predication reference. With the adoption of the method, the problem of unbalancing category training under multi-category and multi-classification condition of integrated learning can be solved, the excessive learning is inhibited well, and moreover, the model predication precision is improved, and reliable reference is provided for classification predication.

Description

technical field [0001] The invention relates to a method for classifying data of unknown categories, in particular to an integrated learning classification method combined with category weights. Background technique [0002] Ensemble learning has become an important research direction in machine learning. Because integrated learning has a certain theoretical basis, is simple to implement, and has good classification and prediction accuracy, it has been widely recognized and applied. As technology advances make data collection easier and easier, it is becoming more common to use ensemble learning to classify multiple categories of data. [0003] The use of integrated learning classification is to use a series of base classifiers for learning, and use some rules to integrate the results of these base classifiers, so as to obtain an integrated classifier with better learning effect and generalization ability than these base classifiers. When the number of categories is known,...

AI Technical Summary

Problems solved by technology

However, cost-sensitive ensemble learning methods are only suitable for unbalanced data, not balanced data

In addition, during model training, the influence of each base classifier on the degree of "over-learning" of different categories is not really considered, and the problem of imbalance between categories still exists

Images