Abstract
An undirected weighted network (UWN) is constantly encountered in various big data-related applications. A UWN’s topological information can be expressed by a Symmetric, High-dimensional and Incomplete (SHDI) matrix, upon which the representation learning task is essential for knowledge acquisition. However, existing models mostly fail in modeling its intrinsic symmetry or low-data density, resulting in the model’s weak representation learning ability to its numerical features. For addressing this vital issue, this study presents a Proximal Symmetric Non-negative Latent-factor-analysis (PSNL) model with three-fold ideas: a) building a proximal term-incorporated, symmetry-aware, and data density-oriented learning objective subjected to the non-negativity constraints for ensuring its high representation learning ability; b) designing an Alternating Direction Method of Multipliers (ADMM)-based learning scheme for solving the learning objective on the premise of fast convergence; c) implementing self-adaptation of the model’s multiple hyper-parameters via the Tree-structured of Parzen Estimators (TPE) algorithm, thus enabling its high scalability. Empirical studies on four UWNs from real applications demonstrate that the proposed PSNL model achieves higher representation accuracy than state-of-the-art models do, as well as promising computational efficiency.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Tang, X., Zhao, B., Qiu, X.: A novel algorithm for prioritizing disease candidate genes from the weighted PPI network. In: Proceedings of 2019 IEEE International Conference on Bioinformatics and Biomedicine, pp. 219–222 (2019)
Tian, L., Luo, P., Wang, H., Zheng, H., Wu, F.: CASNMF: a converged algorithm for symmetrical non-negative matrix factorization. Neurocomputing 275, 2031–2040 (2018)
He, Z., Xie, S., Zdunek, R., Zhou, G., Cichocki, A.: Symmetric non-negative matrix factorization: algorithms and applications to probabilistic clustering. IEEE Trans. Neural Netw. 22(12), 2117–2131 (2011)
Newman, M.: Scientific collaboration networks. II. Shortest paths, weighted networks, and centrality. Phys. Rev. E. 64(1) (2001)
Davis, T., Hu, Y.: The University of Florida sparse matrix collection. ACM Trans. Math. Softw. 38(1) (2011)
Zhang, T., Wang, M., Xi, J., Li, A.: LPGNMF: predicting long non-coding RNA and protein interaction using graph regularized nonnegative matrix factorization. IEEE-ACM Trans. Comput. Biol. Bioinform. 17(1), 189–197 (2020)
He, X., Deng, K., Wang, X., Li, Y., Zhang, Y., Wang, M.: LightGCN: Simplifying and powering graph convolution network for recommendation. In: Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval, pp. 639–648 (2020)
Kong, T., et al.: Linear, or non-linear, that is the question. In: Proceedings of the 15th ACM International Conference on Web Search and Data Mining, pp. 517–525 (2022)
Yang, L., Cao, X., Jin, D., Wang, X., Meng, D.: A unified semi-supervised community detection framework using latent space graph regularization. IEEE Trans. on Cybern. 45(11), 2585–2598 (2015)
Li, X., Zhu, Z., Li, Q., Liu, K.: A provable splitting approach for symmetric nonnegative matrix factorization. IEEE Trans. Knowl. Data Eng. 35(3), 2206–2219 (2023)
Luo, X., Zhou, M., Xia, Y., Zhu, Q.: An efficient non-negative matrix-factorization-based approach to collaborative filtering for recommender systems. IEEE Trans. Ind. Inform. 10(2), 1273–1284 (2014)
Luo, X. Wang, Z., Shang, M.: An instance-frequency-weighted regularization scheme for non-negative latent factor analysis on high-dimensional and sparse data. IEEE Trans. Syst. Man Cybern. Syst. 51(6), 3522–3532 (2021)
Xu, Y., Yin, W., Wen, Z., Zhang, Y.: An alternating direction algorithm for matrix completion with non-negative factors. Front. Math. China. 7(2), 365–384 (2012)
Lu, S., Hong, M., Wang, Z.: A nonconvex splitting method for symmetric nonnegative matrix factorization: convergence analysis and optimality. IEEE Trans. Sig. Process. 65(12), 3120–3135 (2017)
Szklarczyk, D., et al.: STRING v11: protein-protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets. Nucleic Acids Res. 47(D1), D607–D613 (2019)
Luo, X., Yuan, Y., Chen, S., Zeng, N., Wang, Z.: Position-transitional particle swarm optimization-incorporated latent factor analysis. IEEE Trans. Knowl. Data Eng. 8(6), 1524–1536 (2022)
Bergstra, J., Bardenet, R. Bengio, Y. et al.: Algorithms for hyper-parameter optimization. In: Proceedings of 2011 International Conference on Neural Information Processing Systems, pp. 2546–2554 (2011)
Yang, K., Liu, Y., Yu, Z., Philip, C.: Extracting and composing robust features with broad learning system. IEEE Trans. on Knowl. Data Eng. 35(4), 3885–3896 (2023)
Li, W., Wang, R., Luo, X., Zhou, M.: A second-order symmetric non-negative latent factor model for undirected weighted network representation. IEEE Trans. Netw. Sci. Eng. 10(2), 606–618 (2022)
Yu, Z., et al.: Semisupervised classification with novel graph construction for high-dimensional data. IEEE Trans. Neural Netw. Learn. Syst. 33(1), 75–88 (2022)
Balasubramaniam, T., Nayak, R., Yuan, C., Tian, Y.: Column-wise element selection for computationally efficient nonnegative coupled matrix tensor factorization. IEEE Trans. Knowl. Data Eng. 33(9), 3173–3186 (2021)
Hu, L., Guo, G.: An augmented lagrangian alternating direction method for overlapping community detection based on symmetric nonnegative matrix factorization. Int. J. Mach. Cybern. 11(2), 403–415 (2020)
Acknowledgments
This work was supported by the Guangdong Basic and Applied Basic Research Foundation under Grant 2022A1515110579 and 2021B1515140046.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Zhong, Y., Xie, Z., Li, W., Luo, X. (2023). Proximal Symmetric Non-negative Latent Factor Analysis: A Novel Approach to Highly-Accurate Representation of Undirected Weighted Networks. In: Huang, DS., Premaratne, P., Jin, B., Qu, B., Jo, KH., Hussain, A. (eds) Advanced Intelligent Computing Technology and Applications. ICIC 2023. Lecture Notes in Computer Science(), vol 14089. Springer, Singapore. https://doi.org/10.1007/978-981-99-4752-2_6
Download citation
DOI: https://doi.org/10.1007/978-981-99-4752-2_6
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-99-4751-5
Online ISBN: 978-981-99-4752-2
eBook Packages: Computer ScienceComputer Science (R0)