Loading [MathJax]/extensions/MathZoom.js
Toward high accuracy and visualization: An interpretable feature extraction method based on genetic programming and non-overlap degree | IEEE Conference Publication | IEEE Xplore

Toward high accuracy and visualization: An interpretable feature extraction method based on genetic programming and non-overlap degree


Abstract:

Genetic programming (GP) has shown promising results in interpretable feature extraction, but few works considered both classification accuracy and data visualization as ...Show More

Abstract:

Genetic programming (GP) has shown promising results in interpretable feature extraction, but few works considered both classification accuracy and data visualization as objectives. Evaluating the extracted features based on the combination of accuracy measures and visualization measures can help to achieve the two objectives simultaneously. However, the exploitation of improper visualization measures and combination methods will decrease the classification accuracy. In this paper, a novel feature extraction method based on GP and non-overlap degree is proposed to extract interpretable features for high accuracy and visualization. And a novel function that maximizes the product of the accuracy of a linear classifier and the non-overlap degree is proposed to evaluate the extracted features. The proposed method, named GP-ANO, is compared with other methods on five medical datasets by six common machine learning methods. The experimental results demonstrate that the GP-ANO method outperforms other compared methods in terms of both classification accuracy and data visualization.
Date of Conference: 16-19 December 2020
Date Added to IEEE Xplore: 13 January 2021
ISBN Information:
Conference Location: Seoul, Korea (South)

Funding Agency:


I. Introduction

Over the past decades, machine learning methods have already been widely applied in many fields. In general, feature extraction is critical to improve the accuracy and it is also helpful for visualizing and interpreting the data. While, the multi-objective characteristic of feature extraction has usually been ignored. Principal component analysis (PCA) [1], kernel principal component analysis (KPCA) [2], independent component analysis (ICA) [3] are the representatives of feature extraction methods. They project the original feature space into a new subspace to improve the classification accuracy and reduce the feature dimensions. The features extracted by these methods are not interpretable, while the interpretation of the extracted features can contribute to further research in the application.

References

References is not available for this document.