Towards Log-Linear Logics with Concrete Domains

TL;DR

This paper introduces $\

Contribution

It extends log-linear description logics with concrete domains using Markov logic networks and develops a novel inference method for datatypes.

Findings

Effective integration of concrete domains into MLNs

New CPI algorithm for datatype reasoning

Enhanced ontology learning from knowledge bases

Abstract

We present $\mathcal{MEL}^{++}$ (M denotes Markov logic networks) an extension of the log-linear description logics $\mathcal{EL}^{++}$-LL with concrete domains, nominals, and instances. We use Markov logic networks (MLNs) in order to find the most probable, classified and coherent $\mathcal{EL}^{++}$ ontology from an $\mathcal{MEL}^{++}$ knowledge base. In particular, we develop a novel way to deal with concrete domains (also known as datatypes) by extending MLN's cutting plane inference (CPI) algorithm.