KBQA
关注
分享
扫一扫
无知书童
这个作者很懒，什么都没留下…
展开
专栏收录文章
- 默认排序
- 最新发布
- 最早发布
- 最多阅读
- 最少阅读
-
KBQA论文阅读
Semantic Parsing on Freebase from Question-Answer Pairs2013 EMNLP 语义解析方法:通过对自然语言进行语义上的分析，转化成为一种能够让知识库看懂的语义表示步骤：1.词汇映射：即构造底层的语法树节点。将单个自然语言短语或单词映射到知识库实体或知识库实体关系所对应的逻辑形式。我们可以通过构造一个词汇表（Lexicon）来完成这样的映...原创 2019-02-22 11:32:12 · 1863 阅读 · 0 评论 -
QUINT 知识库上的解释性问答
QUINT:Interpretable Question Answering over Knowledge Bases2017 EMNLP 模板+信息提取QUINT分为两个阶段步骤1.已知问题语句utterance和答案a2.在知识库当中查询到问题的主题词和答案a的最小子图做为backbone(骨干查询)3.把问题进行依存树分析4.把问题的依存树和骨干查询的节点进行ILP(整型线性...原创 2019-02-24 20:36:41 · 690 阅读 · 0 评论 -
Question Answering over Freebase via Attentive RNN with Similarity Matrix based CNN
随着近年来知识库的快速发展，基于知识库的问答系统（KBQA ）吸引了业界的广泛关注。该类问答系统秉承先编码再比较的设计思路，即先将问题和知识库中的三元组联合编码至统一的向量空间，然后在该向量空间内做问题和候选答案间的相似度计算。该类方法简单有效，可操作性比较强，然而忽视了很多自然语言词面的原始信息。因此，本文提出了一种 Attentive RNN with Similarity Matrix b...原创 2019-03-01 20:30:47 · 823 阅读 · 1 评论 -
Knowledge Base Question Answering via Encodin of Complex Query Graphs
Knowledge Base Question Answering via Encodin of Complex Query GraphsACL 2018 CompQ 42.84 WebQ 52.66这篇论文好像有问题常规操作：What is the second longest river in United States我们需要推理一些语意线索the answer is co...原创 2019-05-14 11:51:30 · 1724 阅读 · 2 评论 -
Improved Neural Relation Detection for Knowledge Base Question Answering
Improved Neural Relation Detection for Knowledge Base Question Answering2017 ACL SQ78.7 WebQSP63.9KBQA Relation Prediction VS general relation dectection1）一般的关系检测任务，目标分类通常限制在不超过100，二KBQA大于60002）K...原创 2019-05-13 08:33:30 · 1264 阅读 · 0 评论 -
An End-to-End Model for Question Answering over Knowledge Base with Cross-Attention Combining Global
An End-to-End Model for Question Answering over Knowledge Base with Cross-Attention Combining Global Knowledge2017 ACL WQ42.9Candidate GenerationFreebase API method is able to resolve as many as 86...原创 2019-05-13 08:54:46 · 1266 阅读 · 0 评论 -
The APVA-TURBO Approach To Question Answering in Knowledge Base
The APVA-TURBO Approach To Question Answering in Knowledge BaseACL 2018原先APA模型求 p(s,r,o|q)APVA模型多一步 p(y|s,r,q)判断实体和关系的准确度原创 2019-03-11 09:02:24 · 524 阅读 · 2 评论 -
A State-transition Framework to Answer Complex Questions over Knowledge Base
A State-transition Framework to Answer Complex Questions over Knowledge Base原创 2019-06-01 08:52:05 · 942 阅读 · 0 评论