A Support Vector Machine Grid Service Classification Method Based on Feature Weight Learning

Abstract

The present invention relates to a support vector machine power grid business classification method based on feature weight learning, which uses the feature weight learning method combined with the multi-classification characteristics of support vector machines to process the business classification problem, including: dividing the collected data into a training set and a test set , using the 1‑a‑1SVM classification method to decompose the multi-classification problem into a series of SVM binary classification problems, perform parameter optimization and feature subset selection for each SVM binary classifier, and introduce the idea of ​​​​feature learning to assign different features Different weights to represent their importance. According to the selected optimal feature subset and optimal parameter training model, the model obtained is the classification model, and then the test set samples are classified; the method provided by the invention, by selecting each SVM two classifiers respectively according to their respective characteristics The optimal parameters and feature subsets retrain the SVM classification model, which fully considers the differences between different sub-classifiers and has better classification accuracy.

Description

technical field [0001] The invention relates to the field of data processing and classification, in particular to a support vector machine power grid service classification method based on feature weight learning. Background technique [0002] With the deepening of the smart grid and the construction of "three sets and five majors", the types of business carried by the grid are increasing and tend to be complex and changeable. In order to better manage and control the business, optimize the allocation of network resources, and customize personality for different businesses In order to meet the needs of customization, the business must be classified and processed. [0003] The Support Vector Machine (SVM) method has advantages in small sample, nonlinear and high-dimensional classification problems. The key to SVM classification is to seek the optimal classification hyperplane. Generally, the data is divided into training set and test set, the optimal classification boundary ...

AI Technical Summary

Problems solved by technology

However, the distribution of samples in each binary classifier is not the same, and each classifier has its own characteristics. Using the same parameters and feature subsets for each sub-classifier will lead to inaccurate division of classification boundaries, so it is necessary to Improve

Images