论文标题
图形神经网络如何帮助记录检索：关于cord19的案例研究，并产生概念图的生成
How Can Graph Neural Networks Help Document Retrieval: A Case Study on CORD19 with Concept Map Generation
论文作者
论文摘要
图形神经网络（GNN）是一组在不规则数据上表示学习的强大工具，在各种下游任务中都表现出了优越性。使用称为概念图的非结构化文本，可以为文档检索等任务利用GNN。 GNNS如何帮助记录检索，我们对大规模多学科数据集Cord-19进行了经验研究。结果表明，我们提出的面向语义的图形函数不是以结构为导向的GNN（例如杜松子酒和gats），基于BM25检索的候选者，取得了更好，更稳定的性能。我们在此案例研究中的见解可以作为未来工作的指南，以开发有效的GNN，并具有适当的语义为导向的归纳偏见，用于文本推理任务，例如文档检索和分类。该案例研究的所有代码均可在https://github.com/hennyjie/gnn-docretrieval上获得。
Graph neural networks (GNNs), as a group of powerful tools for representation learning on irregular data, have manifested superiority in various downstream tasks. With unstructured texts represented as concept maps, GNNs can be exploited for tasks like document retrieval. Intrigued by how can GNNs help document retrieval, we conduct an empirical study on a large-scale multi-discipline dataset CORD-19. Results show that instead of the complex structure-oriented GNNs such as GINs and GATs, our proposed semantics-oriented graph functions achieve better and more stable performance based on the BM25 retrieved candidates. Our insights in this case study can serve as a guideline for future work to develop effective GNNs with appropriate semantics-oriented inductive biases for textual reasoning tasks like document retrieval and classification. All code for this case study is available at https://github.com/HennyJie/GNN-DocRetrieval.
