Learning From Hidden Traits: Joint Factor Analysis and Latent Clustering
Authors:
- Bo YangBYBo Yang
- Xiao FuXFXiao Fu
- Nicholas D. SidiropoulosNSNicholas D. Sidiropoulos
Preprint
Abstract
Dimensionality reduction techniques play an essential role in data analytics, signal processing and machine learning. Dimensionality reduction is usually performed in a preprocessing stage that is separate from subsequent data analysis, such as clustering or classification. Finding reduced-dimension representations that are well-suited for the intended task is more appealing. This paper proposes a joint factor analysis and latent clustering framework, which aims at learning cluster-aware low-dimensional representations of matrix and tensor data. The proposed approach leverages matrix and tensor factorization models that produce essentially unique latent representations of the data to unravel latent cluster structure -- which is otherwise obscured because of the freedom to apply an oblique transformation in latent space. At the same time, latent cluster structure is used as prior information to enhance the performance of factorization. Specific contributions include several custom-built problem formulations, corresponding algorithms, and discussion of associated convergence properties. Besides extensive simulations, real-world datasets such as Reuters document data and MNIST image data are also employed to showcase the effectiveness of the proposed approaches.Keywords
All Related Versions
- Version 1, 2016-05-22, ArXiv
- Published version: IEEE Transactions on Signal Processing, 65 (1), 256.
This publication has 22 references indexed in Scilit:
- Location based Social Network analysis using Tensors and Signal Processing tools2015
- Putting nonnegative matrix factorization to the test: a tutorial derivation of pertinent cramer—rao bounds and performance benchmarkingIEEE Signal Processing Magazine, 2014
- A Signal Processing Perspective on Hyperspectral Unmixing: Insights from Remote SensingIEEE Signal Processing Magazine, 2013
- Graph Regularized Nonnegative Matrix Factorization for Data RepresentationIEEE Transactions on Pattern Analysis and Machine Intelligence, 2010
- Locally Consistent Concept Factorization for Document ClusteringIEEE Transactions on Knowledge and Data Engineering, 2010
- Total variation norm-based nonnegative matrix factorization for identifying discriminant representation of image patternsNeurocomputing, 2008
- ICE: a statistical approach to identifying endmembers in hyperspectral imagesIEEE Transactions on Geoscience and Remote Sensing, 2004
- Factorial k-means analysis for two-way dataComputational Statistics & Data Analysis, 2001
- Nearest q-Flat to m PointsJournal of Optimization Theory and Applications, 2000
- Learning the parts of objects by non-negative matrix factorizationNature, 1999