Method, device, equipment and medium for selecting partial multi-label features

By updating the tag confidence matrix, aggregating tag granularity and selecting important features in the partially multi-label data processing, the problems of inaccurate feature selection and high computational cost in the prior art are solved, and more efficient and transparent model training is achieved.

CN119312070BActive Publication Date: 2025-03-25HUAQIAO UNIVERSITY
2 Cites 0 Cited by

Patent Information

Application Number
CN202411855277.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-12-17
Publication Date
2025-03-25
Estimated Expiration
2044-12-17

AI Technical Summary

Technical Problem

When processing multi-label data, existing data processing methods fail to fully adapt to the uncertainty and incompleteness of the label, resulting in inaccurate feature selection, reduced model performance, and high calculation cost.

Method used

By obtaining the raw training data with a partially multi-label, initializing the tag confidence matrix, and updating the matrix through the nearest neighbor distribution, disambiguating untrusted and uncertain tags. Then, the labels are aggregated from fine-grained to coarse-grained, a virtual coarse-grained label is constructed, and important features are selected using a forward greedy search algorithm.

Benefits of technology

Improves the accuracy of feature selection and generalization capabilities of the model, reduces computational costs, and provides a more transparent label disambiguation process.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The present invention provides a method, device, equipment and medium for selecting features with too many labels, and relates to the technical field of data preprocessing. The method comprises S1, obtaining an original training data set with too many labels, and initializing a label confidence matrix and constructing a distribution area of ​​labels based on the original training data set. S2, updating the label confidence matrix based on the distribution of neighbors, disambiguating the distribution area of ​​untrustworthy labels, and iteratively updating the distribution area of ​​uncertain labels until the maximum number of iterations is reached, and obtaining the disambiguated labels. S3, based on the disambiguated labels, combining granular computing and graph theory, aggregate the labels from fine granularity to coarse granularity, and construct virtual coarse granular labels. S4, based on the virtual coarse granular labels, combined with multiple features obtained from sample data, construct a neighborhood rough set model suitable for over-labeled data, and use a forward greedy search algorithm to select the most important features.
Need to check novelty before this filing date? Find Prior Art