IoT 23(1): e4

Research Article

Robustness of NMF Algorithms Under Different Noises

Download204 downloads
  • @ARTICLE{10.4108/eetiot.v9i1.3271,
        author={Mengyao Kang and Jiawei  Zhao and Zheng Han},
        title={Robustness of NMF Algorithms Under Different Noises},
        journal={EAI Endorsed Transactions on Internet of Things},
        volume={9},
        number={1},
        publisher={EAI},
        journal_a={IOT},
        year={2023},
        month={6},
        keywords={Machine Learning, Nonnegative matrix factorization, Robustness of algorithm, NMF},
        doi={10.4108/eetiot.v9i1.3271}
    }
    
  • Mengyao Kang
    Jiawei Zhao
    Zheng Han
    Year: 2023
    Robustness of NMF Algorithms Under Different Noises
    IOT
    EAI
    DOI: 10.4108/eetiot.v9i1.3271
Mengyao Kang1,*, Jiawei Zhao1, Zheng Han1
  • 1: University of Sydney
*Contact email: xiao7sky@outlook.com

Abstract

In machine learning, datasets are often disturbed by different noises. The Nonnegative Matrix Factorization (NMF) algorithm provides a robust method to deal with noise, which will significantly improve the efficiency of machine learning. In this investigation, the standard NMF algorithm and L2,1-Norm Based NMF algorithm are studied by designing experiments on different noise types, noise levels, and datasets. Furthermore, Relative Reconstruction Errors (RRE), accuracy, and Normalized Mutual Information (NMI) are used to evaluate the robustness of the two algorithms. In this experiment, there is no significant difference in performance between the two algorithms, while L2,1-Norm Based NMF algorithm shows relatively small advantages.