来源: COLING 2018 Long Paper
原文

Movation

Modeling Semantics with Gated Graph Neural Networks for Knowledge Base Question Answering笔记
以往的模型在复杂问题(问题实体和答案实体之间相隔较远)上的表现很差。
作者认为:

We claim that one needs to explicitly model the semantic structure to be able to find the correct semantic parse for complex questions.

这里所说的“semantic parse(semantic graph)”如图1所示,之后可以转化为查询从KB中得到答案:
Modeling Semantics with Gated Graph Neural Networks for Knowledge Base Question Answering笔记

Semantic parsing

Semantic graphs

用图的方式对问题进行结构化表示,

Our semantic graphs consist of a question variable node (q), Wikidata entities ( Taylor Swift ), relation types from Wikidata ( PERFORMER ) and constraints (see Figure 3 for an example graph with a constraint).

Modeling Semantics with Gated Graph Neural Networks for Knowledge Base Question Answering笔记

Semantic graph construction

可能借鉴了强化学习的思路。
state: S=(E,F). E表示当前的图,F={e|eE} 是在问题中出现但还没有加入图的实体。显然,S1=({},F)
action: A={ae,ac,am}.

Representation learning

Deep Convolutional Networks: 使用DCNN得到向量vq 作为问题的表达。

结论与思考

旧问题,旧思路,新方法,扩展了查询图的定义,并且first to use GGNNs for semantic parsing and KB QA。

相关文章:

  • 2021-11-23
  • 2021-08-26
  • 2021-07-17
  • 2021-07-22
  • 2021-07-28
  • 2021-08-21
  • 2021-04-20
  • 2021-08-17
猜你喜欢
  • 2021-07-09
  • 2021-12-23
  • 2022-01-16
  • 2022-01-11
  • 2021-05-02
  • 2021-10-07
  • 2021-05-13
相关资源
相似解决方案