CNCC2019-陈华钧-可解释的知识图谱推理及应用.pdf

发布时间：2022-05-31 发布人：admin 分类：说明书资料大小：5.35M 资料格式：pdf 举报版权申诉

weixin_42716010-12722375-4744300845407084316.pdf-第1页.png

第1页 / 共26页

weixin_42716010-12722375-4744300845407084316.pdf-第2页.png

第2页 / 共26页

weixin_42716010-12722375-4744300845407084316.pdf-第3页.png

第3页 / 共26页

weixin_42716010-12722375-4744300845407084316.pdf-第4页.png

第4页 / 共26页

weixin_42716010-12722375-4744300845407084316.pdf-第5页.png

第5页 / 共26页

weixin_42716010-12722375-4744300845407084316.pdf-第6页.png

第6页 / 共26页

weixin_42716010-12722375-4744300845407084316.pdf-第7页.png

第7页 / 共26页

weixin_42716010-12722375-4744300845407084316.pdf-第8页.png

第8页 / 共26页

幻灯片编号 1

幻灯片编号 2

知识图谱：刻画事物关系、沉淀领域知识

知识 + 图谱：两个互补的技术维度

What is Reasoning

Reasoning with Knowledge Graph

Deductive Reasoning with Ontological Axioms

Inductive Reasoning with Graph Structure and Rule Learning

Inductive Reasoning with Embedding or Representation Learning

Inductive Reasoning with Graph Neural Network

可微分的知识图谱推理：可解释性和稀疏性Differentiable Reasoning with Knowledge Graph: Explanations and Sparsity

Iteratively Learning Embeddings and Rules for Knowledge Graph Reasoning

知识图谱的稀疏性问题：Meta Relational Learning

利用知识图谱增强电商应用的可解释性

知识图谱的未来发展：融合的知识表征学习

知识图谱的未来发展：知识图谱与低资源问题

知识图谱的未来发展：知识的信任链与众包确权

新书：《知识图谱：方法、实践与应用》by OpenKG

幻灯片编号 26

可解释的知识图谱推理及应用陈华钧浙江大学计算机科学与技术学院教授/博导 CNCC2019

目录知识图谱概述知识图谱推理近期相关工作知识图谱未来 CNCC2019

知识图谱：刻画事物关系、沉淀领域知识知识图谱旨在利用图结构建模、识别和推断事物之间的复杂关联关系和沉淀领域知识，是实现认知智能的重要基石，已经被广泛应用于搜索引擎、智能问答、语言语义理解、大数据决策分析、智能物联等众多领域。浅谈知识图谱推理技术前沿 3 CNCC2019

知识 + 图谱：两个互补的技术维度 Knowledge Graph is more expressive than pure Graph but less complex than formal logic. 浅谈知识图谱推理技术前沿 4 CNCC2019

What is Reasoning 浅谈知识图谱推理技术前沿 5 CNCC2019

Reasoning with Knowledge Graph •Machine reasoning is a difficult task, •Reasoning with KG simplifies the problem to fact prediction or relational reasoning. •Many real life problems (link prediction, causal reasoning, KG-based question answering, recommendation, etc.. ) can be formulated as knowledge graph reasoning or reasoning over a graph. 浅谈知识图谱推理技术前沿 6 关系推理/链接预测/知识图谱补全/质量校验 CNCC2019

Deductive Reasoning with Ontological Axioms Simple logic deduction with ontological axioms (rules) performed mostly over TBox or Schema of KGs. 浅谈知识图谱推理技术前沿 7 CNCC2019

Deductive Reasoning with Ontological Axioms Deductive reasoning with ontology can be very complex when considering more ontological axioms (can be reformulated as rules) or Datalog-like rules. 浅谈知识图谱推理技术前沿 8 CNCC2019

分享到：

赞收藏

资料库

CNCC2019-陈华钧-可解释的知识图谱推理及应用.pdf

相关推荐

人工智能

热门标签

最新资料