Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

An Adaptive Latent Factorization of Tensors Model for Embedding Dynamic Communication Network

Xin Liao, Qicong Hu, Peng Tang

TL;DR

This work tackles embedding and forecasting in Dynamic Communication Networks by representing DCNs as High-Dimensional Sparse tensors and applying a CPD-based, low-rank factorization enhanced with a Temporal-Weight Matrix to model temporal dependencies. The Adaptive Temporal-dependent Tensor Low-Rank Representation (ATT) combines nonnegativity constraints, biases, and a DEA-driven hyperparameter adaptation to achieve robust, scalable learning. Empirical results on four real-world DCNs show ATT attains lower RMSE and MAE and requires fewer training iterations than state-of-the-art baselines, demonstrating improved accuracy and efficiency for large-scale, time-evolving networks. The approach offers a practical pathway for dynamic network analysis with strong generalization and convergence characteristics in real data settings.

Abstract

The Dynamic Communication Network (DCN) describes the interactions over time among various communication nodes, and it is widely used in Big-data applications as a data source. As the number of communication nodes increases and temporal slots accumulate, each node interacts in with only a few nodes in a given temporal slot, the DCN can be represented by an High-Dimensional Sparse (HDS) tensor. In order to extract rich behavioral patterns from an HDS tensor in DCN, this paper proposes an Adaptive Temporal-dependent Tensor low-rank representation (ATT) model. It adopts a three-fold approach: a) designing a temporal-dependent method to reconstruct temporal feature matrix, thereby precisely represent the data by capturing the temporal patterns; b) achieving hyper-parameters adaptation of the model via the Differential Evolutionary Algorithms (DEA) to avoid tedious hyper-parameters tuning; c) employing nonnegative learning schemes for the model parameters to effectively handle an the nonnegativity inherent in HDS data. The experimental results on four real-world DCNs demonstrate that the proposed ATT model significantly outperforms several state-of-the-art models in both prediction errors and convergence rounds.

An Adaptive Latent Factorization of Tensors Model for Embedding Dynamic Communication Network

TL;DR

This work tackles embedding and forecasting in Dynamic Communication Networks by representing DCNs as High-Dimensional Sparse tensors and applying a CPD-based, low-rank factorization enhanced with a Temporal-Weight Matrix to model temporal dependencies. The Adaptive Temporal-dependent Tensor Low-Rank Representation (ATT) combines nonnegativity constraints, biases, and a DEA-driven hyperparameter adaptation to achieve robust, scalable learning. Empirical results on four real-world DCNs show ATT attains lower RMSE and MAE and requires fewer training iterations than state-of-the-art baselines, demonstrating improved accuracy and efficiency for large-scale, time-evolving networks. The approach offers a practical pathway for dynamic network analysis with strong generalization and convergence characteristics in real data settings.

Abstract

The Dynamic Communication Network (DCN) describes the interactions over time among various communication nodes, and it is widely used in Big-data applications as a data source. As the number of communication nodes increases and temporal slots accumulate, each node interacts in with only a few nodes in a given temporal slot, the DCN can be represented by an High-Dimensional Sparse (HDS) tensor. In order to extract rich behavioral patterns from an HDS tensor in DCN, this paper proposes an Adaptive Temporal-dependent Tensor low-rank representation (ATT) model. It adopts a three-fold approach: a) designing a temporal-dependent method to reconstruct temporal feature matrix, thereby precisely represent the data by capturing the temporal patterns; b) achieving hyper-parameters adaptation of the model via the Differential Evolutionary Algorithms (DEA) to avoid tedious hyper-parameters tuning; c) employing nonnegative learning schemes for the model parameters to effectively handle an the nonnegativity inherent in HDS data. The experimental results on four real-world DCNs demonstrate that the proposed ATT model significantly outperforms several state-of-the-art models in both prediction errors and convergence rounds.
Paper Structure (14 sections, 19 equations, 2 figures, 4 tables, 1 algorithm)

This paper contains 14 sections, 19 equations, 2 figures, 4 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Background
  3. Symbol Definitions
  4. Problem Formulation
  5. ATT model
  6. Objective Function
  7. Parameters Learning
  8. The DEA-based Hyper-parameters Adaptation
  9. Empirical Studies
  10. Datasets
  11. Evaluation Metrics
  12. Training settings
  13. Comparison Results
  14. Conclusions and future works

Figures (2)

  • Figure 1: A low-rank representation of HDS tensor constructed by a DCN.
  • Figure 2: An illustration of the temporal-dependent.