Work

Combining Rule Learning and Nonmonotonic Reasoning for Link Prediction in Knowledge Graphs.

Year: 2017

Type: conference-paper

Abstract: Learning rules from knowledge graphs (KG) is a crucial task for KG completion, cleaning and curation.The majority of existing approaches are capable of learning only Horn rules from KGs, which are, however, insufficiently expressive for capturing exceptions and thus might make incorrect predictions on missing facts.In this paper we discuss our recent progress in addressing this limitation.More specifically, we report the challenges of learning rules with exceptions, briefly describe our recent approach which combines rule learning and nonmonotonic reasoning and outline the ongoing and future research directions.

Source: MPG.PuRe (Max Planck Society)

Authors: Francesca A. Lisi, Daria Stepanova

Institutions: University of Bari Aldo Moro, Max Planck Society, Max Planck Institute for Informatics

Language: English

Topic: Advanced Graph Neural Networks

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

OpenAlex ID: W2753693235

FWCI:

Cites: 5

Cited by:

Related to: 10