.png){kind=icon}
.png){kind=icon}
Return
l2,1 norm regularized multi-kernel based joint nonlinear feature selection and over-sampling for imbalanced data classification
.png){kind=icon}
OA
AI
DOI：10.1016/j.neucom.2016.12.036.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
High dimensionality and classification of imbalanced data sets are two of the most interesting machine learning challenges. Both issues have been independently studied in the literature. In order to simultaneously explore the both issues of feature selection and oversampling, we efficiently combine two different methodological approaches in an unified kernel framework. Specifically, we proposed a novel l(2,1) norm balanced multiple kernel feature selection (l(2,1) MKFS), and designed a proximal based optimization algorithm for efficiently learning the model. Moreover, multiple kernel oversampling (MKOS) was developed to generate synthetic instances in the optimal kernel space induced by l(2,1) MKFS, so as to compensate for the class imbalanced distribution. Our experimental results on multiple UCI data and two real medical application demonstrate that jointly operating nonlinear feature selection and oversampling with l(2,1) norm multi-kernel learning framework (l(2,1) MKFSOS) can lead to a promising classification performance.
Keywords:
Imbalanced data learning
Feature selection
Classification
Multi-kernel learning
Proximal method
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
6.5
Papers:
2.5W
Citations:
6.5W