Query Path Generation via Bidirectional Reasoning for Multihop Question Answering From Knowledge Bases
Query Path Generation via Bidirectional Reasoning for Multihop Question Answering From Knowledge Bases
复制标题
DOI：
10.1109/tcds.2022.3198272
复制
发表时间：
2023-09
影响因子：
5
通讯作者：
Geng Zhang;Jin Liu;Guangyou Zhou;Zhiwen Xie;Xiao Yu;Xiaohui Cui
中科院分区：
文献类型：
--
作者：
Geng Zhang;Jin Liu;Guangyou Zhou;Zhiwen Xie;Xiao Yu;Xiaohui Cui
基于知识库的多跳问答是自然语言处理领域的一个研究热点。最近，基于图神经网络（GNN）的方法已经取得了可喜的成果，因为知识库可以组织为知识图（KG）。然而，他们往往遭受的稀疏性的KG，这是有害的结构编码和推理能力的GNN。具体来说，KG是一个稀疏的图形连接的有向关系和以前的研究很少注意的KG中的关系的方向特性，限制了模式的关系路径，基于GNN的方法可以解决。本研究提出一种双向递归GNN（BRGNN）来解决这些困难。为了对关系的双向信息进行建模，将实体的所有相邻关系按其方向进行分组，并分别在向外和向内的方向上聚合到实体表示中。在推理过程中，BRGNN同时考虑两个方向上的邻居关系，以覆盖更多的关系路径模式，提高答案的召回率。在三个基准测试：WebWebWebsSP，ComplexWebQuestions和MetaQA上进行了大量的实验，验证了BRGNN可以通过考虑方向信息来回答更多的问题，并且与所有最先进的方法相比具有竞争力。
Multihop question answering from knowledge bases (KBQA) is a hot research topic in natural language processing. Recently, the graph neural network-based (GNN-based) methods have achieved promising results as the KB can be organized as a knowledge graph (KG). However, they often suffered from the sparsity of the KG which was detrimental to the structure encoding and reasoning capabilities of GNN. Specifically, a KG is a sparse graph linked by directed relations and previous studies have paid scant attention to the directional characteristic of relations in the KG, limiting the patterns of relation path that GNN-based approaches could resolve. This study proposes a bidirectional recurrent GNN (BRGNN) to tackle these difficulties. To model the bidirectional information of relations, all adjacent relations of an entity are grouped by their directions, and they are separately aggregated into the entity representation in outward and inward directions. For the reasoning process, BRGNN simultaneously considers the neighbor relations in both directions to cover more patterns of relation paths and improve the recall of answers. Extensive experiments on three benchmarks: WebQuestionsSP, ComplexWebQuestions, and MetaQA, verify that BRGNN can answer more questions by taking into account the directional information, and it is competitive to all state-of-the-art approaches.