Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

GACL: Graph Attention Collaborative Learning for Temporal QoS Prediction

Shengxiang Hu, Guobing Zou, Bofeng Zhang, Shaogang Wu, Shiyi Lin, Yanglan Gan, Yixin Chen

TL;DR

A novel Graph Attention Collaborative Learning (GACL) framework for temporal QoS prediction is proposed, which significantly outperforms state-of-the-art methods for temporal QoS prediction across multiple evaluation metrics, achieving the improvements of up to 38.80%.

Abstract

Accurate prediction of temporal QoS is crucial for maintaining service reliability and enhancing user satisfaction in dynamic service-oriented environments. However, current methods often neglect high-order latent collaborative relationships and fail to dynamically adjust feature learning for specific user-service invocations, which are critical for precise feature extraction within each time slice. Moreover, the prevalent use of RNNs for modeling temporal feature evolution patterns is constrained by their inherent difficulty in managing long-range dependencies, thereby limiting the detection of long-term QoS trends across multiple time slices. These shortcomings dramatically degrade the performance of temporal QoS prediction. To address the two issues, we propose a novel Graph Attention Collaborative Learning (GACL) framework for temporal QoS prediction. Building on a dynamic user-service invocation graph to comprehensively model historical interactions, it designs a target-prompt graph attention network to extract deep latent features of users and services at each time slice, considering implicit target-neighboring collaborative relationships and historical QoS values. Additionally, a multi-layer Transformer encoder is introduced to uncover temporal feature evolution patterns, enhancing temporal QoS prediction. Extensive experiments on the WS-DREAM dataset demonstrate that GACL significantly outperforms state-of-the-art methods for temporal QoS prediction across multiple evaluation metrics, achieving the improvements of up to 38.80%.

GACL: Graph Attention Collaborative Learning for Temporal QoS Prediction

TL;DR

A novel Graph Attention Collaborative Learning (GACL) framework for temporal QoS prediction is proposed, which significantly outperforms state-of-the-art methods for temporal QoS prediction across multiple evaluation metrics, achieving the improvements of up to 38.80%.

Abstract

Accurate prediction of temporal QoS is crucial for maintaining service reliability and enhancing user satisfaction in dynamic service-oriented environments. However, current methods often neglect high-order latent collaborative relationships and fail to dynamically adjust feature learning for specific user-service invocations, which are critical for precise feature extraction within each time slice. Moreover, the prevalent use of RNNs for modeling temporal feature evolution patterns is constrained by their inherent difficulty in managing long-range dependencies, thereby limiting the detection of long-term QoS trends across multiple time slices. These shortcomings dramatically degrade the performance of temporal QoS prediction. To address the two issues, we propose a novel Graph Attention Collaborative Learning (GACL) framework for temporal QoS prediction. Building on a dynamic user-service invocation graph to comprehensively model historical interactions, it designs a target-prompt graph attention network to extract deep latent features of users and services at each time slice, considering implicit target-neighboring collaborative relationships and historical QoS values. Additionally, a multi-layer Transformer encoder is introduced to uncover temporal feature evolution patterns, enhancing temporal QoS prediction. Extensive experiments on the WS-DREAM dataset demonstrate that GACL significantly outperforms state-of-the-art methods for temporal QoS prediction across multiple evaluation metrics, achieving the improvements of up to 38.80%.
Paper Structure (21 sections, 19 equations, 5 figures, 4 tables)

This paper contains 21 sections, 19 equations, 5 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Problem Definition
  3. Approach
  4. Dynamic User-Service Invocation Graph Modeling
  5. Target-prompt Deep Latent Feature Extraction of Users and Services
  6. Temporal Feature Evolution Mining of Users and Services
  7. Temporal QoS Prediction and Model Training
  8. Experiments
  9. Experimental Setup and Dataset
  10. Competing Methods
  11. Evaluation Metrics
  12. Competing Results and Analyses
  13. Ablation Study
  14. Performance Impact of Parameters
  15. Performance Impact Analysis of $l_g$
  16. ...and 6 more sections

Figures (5)

  • Figure 1: Overview of the GACL framework for temporal QoS prediction, which comprises four key components: a) Dynamic User-Service Invocation Graph Modeling, b) Target-prompt User/Service Deep Feature Extraction, c) Temporal Feature Evolution Pattern Mining, and d) Temporal QoS Prediction.
  • Figure 2: Message passing and aggregation in the target-prompt graph attention network. It adjusts the attention mechanism and refines the attention scores by fusing the implicit collaborative relationships between target service and the neighboring services, as well as the historical QoS values of user-service interactions.
  • Figure 3: Ablation experiment results of NMAE and RMSE over RT and TP.
  • Figure 4: Parameter impact of layers of target-prompt graph attention network $l_g$ and dimension of user/service feature $d$.
  • Figure 5: Parameter impact of various combination of window size $ws$ and layers of the transformer encoder $l_{tf}$.

Theorems & Definitions (4)

  • Definition 1: Temporal Service Ecosystem
  • Definition 2: User-service QoS Invocation
  • Definition 3: Temporal QoS Prediction
  • Definition 4: Dynamic User-Service Invocation Graph