CauseKG: A Framework Enhancing Causal Inference With Implicit Knowledge Deduced From Knowledge Graphs

Publikation: Beitrag in FachzeitschriftArtikelForschungPeer-Review

Autoren

  • Hao Huang
  • Maria Esther Vidal

Organisationseinheiten

Externe Organisationen

  • Technische Informationsbibliothek (TIB) Leibniz-Informationszentrum Technik und Naturwissenschaften und Universitätsbibliothek
Forschungs-netzwerk anzeigen

Details

OriginalspracheEnglisch
Seiten (von - bis)61810-61827
Seitenumfang18
FachzeitschriftIEEE ACCESS
Jahrgang12
Frühes Online-Datum29 Apr. 2024
PublikationsstatusVeröffentlicht - 7 Mai 2024

Abstract

Causal inference is a critical technique for inferring causal relationships from data and distinguishing causation from correlation. Causal inference frameworks rely on structured data, typically represented in flat tables or relational models. These frameworks estimate causal effects based only on explicit facts, overlooking implicit information in the data, which can lead to inaccurate causal estimates. Knowledge graphs (KGs) inherently capture implicit information through logical rules applied to explicit facts, providing a unique opportunity to leverage implicit knowledge. However, existing frameworks are not applicable to KGs due to their semi-structured nature. CauseKG is a causal inference framework designed to address the intricacies of KGs and seamlessly integrate implicit information using KG-specific entailment techniques, providing a more accurate causal inference process. We empirically evaluate the effectiveness of CauseKG against benchmarks constructed from synthetic and real-world datasets. The results suggest that CauseKG can produce a lower mean absolute error in causal inference compared to state-of-the-art methods. The empirical results demonstrate CauseKG's ability to address causal questions in a variety of domains. This research highlights the importance of extending causal inference techniques to KGs, emphasising the improved accuracy that can be achieved by integrating implicit and explicit information.

Zitieren

CauseKG: A Framework Enhancing Causal Inference With Implicit Knowledge Deduced From Knowledge Graphs. / Huang, Hao; Vidal, Maria Esther.
in: IEEE ACCESS, Jahrgang 12, 07.05.2024, S. 61810-61827.

Publikation: Beitrag in FachzeitschriftArtikelForschungPeer-Review

Huang H, Vidal ME. CauseKG: A Framework Enhancing Causal Inference With Implicit Knowledge Deduced From Knowledge Graphs. IEEE ACCESS. 2024 Mai 7;12:61810-61827. Epub 2024 Apr 29. doi: 10.1109/ACCESS.2024.3395134
Huang, Hao ; Vidal, Maria Esther. / CauseKG : A Framework Enhancing Causal Inference With Implicit Knowledge Deduced From Knowledge Graphs. in: IEEE ACCESS. 2024 ; Jahrgang 12. S. 61810-61827.
Download
@article{a8966c39c04b472bb73febde143fc0c0,
title = "CauseKG: A Framework Enhancing Causal Inference With Implicit Knowledge Deduced From Knowledge Graphs",
abstract = "Causal inference is a critical technique for inferring causal relationships from data and distinguishing causation from correlation. Causal inference frameworks rely on structured data, typically represented in flat tables or relational models. These frameworks estimate causal effects based only on explicit facts, overlooking implicit information in the data, which can lead to inaccurate causal estimates. Knowledge graphs (KGs) inherently capture implicit information through logical rules applied to explicit facts, providing a unique opportunity to leverage implicit knowledge. However, existing frameworks are not applicable to KGs due to their semi-structured nature. CauseKG is a causal inference framework designed to address the intricacies of KGs and seamlessly integrate implicit information using KG-specific entailment techniques, providing a more accurate causal inference process. We empirically evaluate the effectiveness of CauseKG against benchmarks constructed from synthetic and real-world datasets. The results suggest that CauseKG can produce a lower mean absolute error in causal inference compared to state-of-the-art methods. The empirical results demonstrate CauseKG's ability to address causal questions in a variety of domains. This research highlights the importance of extending causal inference techniques to KGs, emphasising the improved accuracy that can be achieved by integrating implicit and explicit information.",
keywords = "Causal inference, knowledge graphs, knowledge reasoning, semantics",
author = "Hao Huang and Vidal, {Maria Esther}",
note = "Publisher Copyright: {\textcopyright} 2013 IEEE.",
year = "2024",
month = may,
day = "7",
doi = "10.1109/ACCESS.2024.3395134",
language = "English",
volume = "12",
pages = "61810--61827",
journal = "IEEE ACCESS",
issn = "2169-3536",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

Download

TY - JOUR

T1 - CauseKG

T2 - A Framework Enhancing Causal Inference With Implicit Knowledge Deduced From Knowledge Graphs

AU - Huang, Hao

AU - Vidal, Maria Esther

N1 - Publisher Copyright: © 2013 IEEE.

PY - 2024/5/7

Y1 - 2024/5/7

N2 - Causal inference is a critical technique for inferring causal relationships from data and distinguishing causation from correlation. Causal inference frameworks rely on structured data, typically represented in flat tables or relational models. These frameworks estimate causal effects based only on explicit facts, overlooking implicit information in the data, which can lead to inaccurate causal estimates. Knowledge graphs (KGs) inherently capture implicit information through logical rules applied to explicit facts, providing a unique opportunity to leverage implicit knowledge. However, existing frameworks are not applicable to KGs due to their semi-structured nature. CauseKG is a causal inference framework designed to address the intricacies of KGs and seamlessly integrate implicit information using KG-specific entailment techniques, providing a more accurate causal inference process. We empirically evaluate the effectiveness of CauseKG against benchmarks constructed from synthetic and real-world datasets. The results suggest that CauseKG can produce a lower mean absolute error in causal inference compared to state-of-the-art methods. The empirical results demonstrate CauseKG's ability to address causal questions in a variety of domains. This research highlights the importance of extending causal inference techniques to KGs, emphasising the improved accuracy that can be achieved by integrating implicit and explicit information.

AB - Causal inference is a critical technique for inferring causal relationships from data and distinguishing causation from correlation. Causal inference frameworks rely on structured data, typically represented in flat tables or relational models. These frameworks estimate causal effects based only on explicit facts, overlooking implicit information in the data, which can lead to inaccurate causal estimates. Knowledge graphs (KGs) inherently capture implicit information through logical rules applied to explicit facts, providing a unique opportunity to leverage implicit knowledge. However, existing frameworks are not applicable to KGs due to their semi-structured nature. CauseKG is a causal inference framework designed to address the intricacies of KGs and seamlessly integrate implicit information using KG-specific entailment techniques, providing a more accurate causal inference process. We empirically evaluate the effectiveness of CauseKG against benchmarks constructed from synthetic and real-world datasets. The results suggest that CauseKG can produce a lower mean absolute error in causal inference compared to state-of-the-art methods. The empirical results demonstrate CauseKG's ability to address causal questions in a variety of domains. This research highlights the importance of extending causal inference techniques to KGs, emphasising the improved accuracy that can be achieved by integrating implicit and explicit information.

KW - Causal inference

KW - knowledge graphs

KW - knowledge reasoning

KW - semantics

UR - http://www.scopus.com/inward/record.url?scp=85192191045&partnerID=8YFLogxK

U2 - 10.1109/ACCESS.2024.3395134

DO - 10.1109/ACCESS.2024.3395134

M3 - Article

AN - SCOPUS:85192191045

VL - 12

SP - 61810

EP - 61827

JO - IEEE ACCESS

JF - IEEE ACCESS

SN - 2169-3536

ER -