Knowledge Graph-driven Knowledge Recommendation based on Generative Adversarial Networks
Hanqing Wu, Zhenyong Wu, Lina He, Xianyu Zhang · 2025
With the increasing complexity of product design, effectively managing and recommending knowledge has become essential for improving efficiency and innovation. Knowledge Graphs (KGs) provide a structured approach to organizing and representing knowledge, enabling intelligent reasoning and retrieval. However, challenges such as knowledge heterogeneity, dynamic updates, and contextual dependencies hinder effective knowledge utilization. To address these issues, this paper proposes a KG-driven knowledge recommendation framework that integrates generative adversarial networks (GANs) for enhanced reasoning and recommendation accuracy. By leveraging KGs, the system captures complex relationships between design elements, facilitating precise knowledge retrieval and reuse. Experimental results demonstrate the effectiveness of this approach in optimizing knowledge dissemination, reducing redundancy, and supporting intelligent decisionmaking in complex product design.