Answering Multiple Conjunctive Queries with the Lifted Dynamic Junction Tree Algorithm

Abstract

The lifted dynamic junction tree algorithm (LDJT) answers filtering and prediction queries efficiently for probabilistic relational temporal models by building and then reusing a first-order cluster representation of a knowledge base for multiple queries and time steps. We extend LDJT to answer conjunctive queries over multiple time steps by avoiding eliminations, while keeping the complexity to answer a conjunctive query low. The extended version of saves computations compared to an existing approach to answer multiple lifted conjunctive queries.
Original languageEnglish
Title of host publicationAI 2018: Advances in Artificial Intelligence
EditorsTanja Mitrovic, Bing Xue, Xiaodong Li
Number of pages13
Volume11320
Place of PublicationCham
PublisherSpringer International Publishing
Publication date10.11.2018
Pages543-555
ISBN (Print)978-3-030-03990-5
ISBN (Electronic)978-3-030-03991-2
DOIs
Publication statusPublished - 10.11.2018
Event31st Australasian Joint Conference on Artificial Intelligence
- Wellington, Niger
Duration: 11.12.201814.12.2018
https://ecs.victoria.ac.nz/Events/AI2018/WebHome#gallery

Research Areas and Centers

  • Centers: Center for Artificial Intelligence Luebeck (ZKIL)
  • Research Area: Intelligent Systems

DFG Research Classification Scheme

  • 4.43-01 Theoretical Computer Science

Fingerprint

Dive into the research topics of 'Answering Multiple Conjunctive Queries with the Lifted Dynamic Junction Tree Algorithm'. Together they form a unique fingerprint.

Cite this