Knowledge Integration for Commonsense Reasoning with Default Logic

Priit Järv, Tanel Tammet, Martin Verrev, Dirk Draheim

2022

Abstract

Commonsense reasoning in artificial intelligence is the problem of inferring decisions and answers regarding mundane situations. Several research groups have built large knowledge graphs with the goal of capturing some aspects of commonsense knowledge. Using these knowledge graphs for problem solving and question answering is a subject of active research. Our contribution is encoding and integrating knowledge graphs like Quasimodo, ConceptNet, and Wordnet for symbolic reasoning. A major challenge in symbolic commonsense reasoning is coping with contradictory and uncertain knowledge, which we handle by extending first order logic with numeric confidences and default logic rules. To our knowledge this is the first large scale commonsense knowledge base seriously using default logic. We give several examples of the proposed representation and solving questions on the basis of the knowledge base built.

Paper Citation

in Harvard Style

Järv P., Tammet T., Verrev M. and Draheim D. (2022). Knowledge Integration for Commonsense Reasoning with Default Logic. In Proceedings of the 14th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management (IC3K 2022) - Volume 2: KEOD; ISBN 978-989-758-614-9, SciTePress, pages 148-155. DOI: 10.5220/0011532200003335

in Bibtex Style

@conference{keod22,
author={Priit Järv and Tanel Tammet and Martin Verrev and Dirk Draheim},
title={Knowledge Integration for Commonsense Reasoning with Default Logic},
booktitle={Proceedings of the 14th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management (IC3K 2022) - Volume 2: KEOD},
year={2022},
pages={148-155},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0011532200003335},
isbn={978-989-758-614-9},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 14th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management (IC3K 2022) - Volume 2: KEOD
TI - Knowledge Integration for Commonsense Reasoning with Default Logic
SN - 978-989-758-614-9
AU - Järv P.
AU - Tammet T.
AU - Verrev M.
AU - Draheim D.
PY - 2022
SP - 148
EP - 155
DO - 10.5220/0011532200003335
PB - SciTePress