Hybrid projective nonnegative matrix factorization based on α-divergence and the alternating least squares algorithm

作者：

Highlights：

• Developed a new hybrid Projective NMF algorithm based on α-divergence and ALS.

• The new algorithm exhibits better reconstruction accuracy than other PNMF algorithms.

• The new algorithm provides very sparse and highly “orthogonal” basis factors.

• The new algorithm extracts better and distinctive localized features.

摘要

•Developed a new hybrid Projective NMF algorithm based on α-divergence and ALS.•The new algorithm exhibits better reconstruction accuracy than other PNMF algorithms.•The new algorithm provides very sparse and highly “orthogonal” basis factors.•The new algorithm extracts better and distinctive localized features.

论文关键词：Hybrid projective nonnegative matrix factorization,α-Divergence,Feature extraction,Clustering,Orthogonality,Sparsity

论文评审过程：Received 30 January 2019, Revised 24 May 2019, Accepted 10 October 2019, Available online 30 October 2019, Version of Record 30 October 2019.

论文官网地址：https://doi.org/10.1016/j.amc.2019.124825