A Fast Nonnegative Autoencoder-Based Approach to Latent Feature Analysis on High-Dimensional and Incomplete Data

• Published in: IEEE transactions on services computing Vol. 17; no. 3; pp. 733 - 746
• Main Authors: Bi, Fanghui, He, Tiantian, Luo, Xin
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 01.05.2024; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Accuracy; Adaptation models; Algorithms; Big Data; Computational efficiency; Computational modeling; Data models; data science; Dimensional analysis; fast nonnegative AutoEncoder; Feature extraction; high-dimensional and incomplete data; Industrial applications; Knowledge acquisition; latent feature analysis; link prediction; Machine learning; network representation learning; neural network; Neural networks; Representation learning; Representations
• ISSN: 1939-1374; 2372-0204
• DOI: 10.1109/TSC.2023.3319713

High-Dimensional and Incomplete (HDI) data are frequently encountered in various Big Data-related applications. Despite its incompleteness, an HDI data repository contains rich knowledge and patterns concerning the complex interactions among numerous nodes. Recently, a Neural Network (NN)-based approach to Latent Feature Analysis (LFA) model becomes popular owing to its strong representation learning ability to HDI data. Nevertheless, existing NN-based LFA models neglect the inherent nonnegativity in most HDI data, resulting in representation accuracy loss. Motivated by this discovery, this study innovatively proposes a F ast N onnegative A uto E ncoder (FNAE)-based approach to LFA on HDI data, whose ideas are three-fold: a) constructing a multilayered autoencoder subject to nonnegativity constraints for high representation learning ability; b) incorporating the data density-oriented modeling mechanism into FNAE's input and output layers for high computational and storage efficiency; and c) implementing an Adam-based single latent factor-dependent, nonnegative and multiplicative update algorithm for efficient model training as well as fulfilling the nonnegativity constraints. Experimental results on eight commonly-adopted HDI matrices from industrial applications demonstrate that the proposed FNAE significantly outperforms several state-of-the-art NN-based LFA models in both estimation accuracy for missing links of an HDI matrix and computational efficiency.