Researchers have developed AEGIS, a novel framework designed to improve link prediction in sparse bipartite knowledge graphs. This edge-only augmentation method resamples existing training edges, preserving the original node set to avoid fabricated endpoints. Experiments on datasets like Amazon, MovieLens, and a game design pattern network demonstrated that AEGIS, particularly with semantic augmentation, can enhance prediction accuracy and calibration, especially when descriptive node information is available. AI
影响 Introduces a new method for improving link prediction in sparse knowledge graphs, potentially aiding recommendation systems and data analysis.
排序理由 This is a research paper detailing a new framework for link prediction in knowledge graphs.
AI 生成摘要 · Google Gemini · 来自 1 个来源。 我们如何撰写摘要 →
