Modeling Method and Implementation Mechanism for QoS Prediction Based on Multi-Source Data Fusion
TL;DR
This study integrates three types of heterogeneous information to construct a unified embedding space, achieves cross-source alignment of semantically heterogeneous data through a hierarchical architecture, introduces differentiated dynamic weight allocation among three data sources, and employs multi-source context-guided sparse compensation to address missing entries in the service invocation matrix.
Abstract
: Quality of Service (QoS) prediction in service-oriented computing supports candidate service identification and resource scheduling. However, the limited single-source data dimension fails to capture dynamic characteristics such as network fluctuations, spatiotemporal migration, and user behavior coupling. This study integrates three types of heterogeneous information to construct a unified embedding space, achieves cross-source alignment of semantically heterogeneous data through a hierarchical architecture, introduces differentiated dynamic weight allocation among three data sources, captures nonlinear evolution patterns through temporal joint representations, and employs multi-source context-guided sparse compensation to address missing entries in the service invocation matrix. Experimental results demonstrate that the proposed method significantly outperforms single-source models in both prediction accuracy and robustness under complex scenarios.