Robust Structure Preserving Nonnegative Matrix Factorization for Dimensionality Reduction

Abstract

As a linear dimensionality reduction method, nonnegative matrix factorization (NMF) has been widely used in many fields, such as machine learning and data mining. However, there are still two major drawbacks for NMF: (a) NMF can only perform semantic factorization in Euclidean space, and it fails to discover the intrinsic geometrical structure of high-dimensional data distribution. (b) NMF suffers from noisy data, which are commonly encountered in real-world applications. To address these issues, in this paper, we present a new robust structure preserving nonnegative matrix factorization (RSPNMF) framework. In RSPNMF, a local affinity graph and a distant repulsion graph are constructed to encode the geometrical information, and noisy data influence is alleviated by characterizing the data reconstruction term of NMF withl2,1-norm instead ofl2-norm. With incorporation of the local and distant structure preservation regularization term into the robust NMF framework, our algorithm can discover a low-dimensional embedding subspace with the nature of structure preservation. RSPNMF is formulated as an optimization problem and solved by an effective iterative multiplicative update algorithm. Experimental results on some facial image datasets clustering show significant performance improvement of RSPNMF in comparison with the state-of-the-art algorithms.

Research on Intelligent Diagnosis of Fault Data of Large and Medium-Sized Pumping Stations Under Information Evaluation System

“…In that invention, the fault data set information acquisition node distribution model of the large and medium-sized pump station is constructed, and a multi-sensor fusion sampling method is adopted to carry out fault data sampling and characteristic decomposition of the large and medium-sized pump station, In this paper, the intelligent diagnosis algorithm for the fault data of large and medium-sized pump station based on the information-based evaluation system is presented. The abnormal working condition characteristic quantity of the fault data of the large and medium-sized pump station is excavated, the difference distribution characteristic of the abnormal working condition of the fault data set of the large and medium-sized pump station is analyzed [ 13 ]. The abnormal working condition component of the fault data of the large and medium-sized pump station is decomposed by the singular value, so that the abnormal working condition component of .…”

Research on Resource Equilibrium Scheduling of Massive MOOC Teaching Resources Information Management System

“…The task priority is optimized and the fuzzy K-means clustering method is designed to deal with the adjacent tasks of teaching resource scheduling [6]. The clustering model is described as follows:…”

Construction and Optimization of Ideological and Political Education for College Students Based on Mobile Network Platform

“…Third, to overcome the shortcomings of the traditional mode of ideological and political education for college students to adopt new ways, such as interactive, experiential, guided, and permeable, etc. In order to realize the harmony and unity of virtual space and real space, the ideological and political education of college students should be carried out with the help of mobile phone network platform (Li, Tang, & Han, 2017).…”