• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

计算机工程与科学 ›› 2023, Vol. 45 ›› Issue (07): 1282-1291.

• 人工智能与数据挖掘 • 上一篇    下一篇

基于自编码器和局部嵌入的无监督特征选择

赵瑞平,降爱莲   

  1. (太原理工大学信息与计算机学院,山西 晋中 030600)
  • 收稿日期:2021-12-23 修回日期:2022-02-16 接受日期:2023-07-25 出版日期:2023-07-25 发布日期:2023-07-11
  • 基金资助:
    山西省回国留学人员科研资助项目(2017-051)

Unsupervised feature selection based on autoencoder and local embedding

ZHAO Rui-ping,JIANG Ai-lian   

  1. (College of Information and Computer,Taiyuan University of Technology,Jinzhong 030600,China)
  • Received:2021-12-23 Revised:2022-02-16 Accepted:2023-07-25 Online:2023-07-25 Published:2023-07-11

摘要: 为了能够在学习特征之间深层非线性关系的同时,保持特征局部几何结构,提出一种利用单层自编码器作为特征选择和流形学习的算法。首先,利用单层自编码器的重建能力剔除对重建样本贡献微弱的单个特征,学习特征深层非线性关系,并在特征权重矩阵上进行稀疏正则化;然后,通过改进局部线性嵌入算法保持特征之间局部结构,得到一个最优特征子集;最后,设计一个新的目标损失函数,并采用L-BFGS优化算法进行迭代优化。在6个数据集上与其他6种无监督特征选择算法进行对比,实验结果表明,该算法的聚类性能和分类性能要优于其他无监督特征选择算法的。

关键词: 特征选择, 自编码器;局部线性嵌入;非线性关系;局部几何结构

Abstract: In order to maintain the local geometric structure of features while learning the deep nonlinear relationship between features, this paper proposes a single-layer autoencoder as a joint framework for feature selection and manifold learning. Firstly, the reconstruction capability of single-layer autoencoder is used to eliminate the single feature with weak contribution to the reconstructed sample, learn the deep nonlinear relationship of the feature, and carry out sparse regularization on the feature weight matrix. Secondly, an optimal feature subset is obtained by improving the local linear embedding algorithm to preserve the local structure among features. Finally, a new target loss function is designed and the L-BFGS algorithm is used for iterative optimization. Compared with other six unsupervised feature selection algorithms on six data sets, the experimental results show that this algorithm is superior to other unsupervised feature selection algorithms in clustering performance and classification performance.

Key words: feature selection, autoencoder, local linear embedding, nonlinear relationship, local geo-metric structure