Knowledge graphs (KGs) naturally capture the convergence of data and knowledge, making them expressive frameworks for describing and integrating heterogeneous data in a coherent and interconnected manner. However, based on the Open World Assumption (OWA), the absence of information within KGs does not indicate falsity or non-existence; it merely reflects incompleteness. Inductive learning over KGs involves predicting new relationships based on existing statements in the KG, using either numerical or symbolic learning models. The Partial Completeness Assumption (PCA) heuristic efficiently guides inductive learning methods for Link Prediction (LP) by refining predictions about absent KG relationships. Nevertheless, numeric techniques– like KG embedding models– alone may fall short in accurately predicting missing information, particularly when it comes to capturing implicit knowledge and complex relationships. We propose a hybrid method named SPaRKLE that seamlessly integrates symbolic and numerical techniques, leveraging the PCA heuristic to capture implicit knowledge and enrich KGs. We empirically compare SPaRKLE with state-of-the-art KG embedding and symbolic models, using established benchmarks. Our experimental outcomes underscore the efficacy of this hybrid approach, as it harnesses the strengths of both paradigms. SPaRKLE is publicly available on GitHub1.
%0 Conference Paper
%1 Purohit2023
%A Purohit, Disha
%A Chudasama, Yashrajsinh
%A Rivas, Ariam
%A Vidal, Maria-Esther
%B Proceedings of the 12th Knowledge Capture Conference 2023
%C New York, NY, USA
%D 2023
%I Association for Computing Machinery
%K myown from:gabydler
%P 44–52
%R 10.1145/3587259.3627547
%T SPaRKLE : Symbolic caPtuRing of knowledge for Knowledge graph enrichment with LEarning
%U https://doi.org/10.1145/3587259.3627547
%X Knowledge graphs (KGs) naturally capture the convergence of data and knowledge, making them expressive frameworks for describing and integrating heterogeneous data in a coherent and interconnected manner. However, based on the Open World Assumption (OWA), the absence of information within KGs does not indicate falsity or non-existence; it merely reflects incompleteness. Inductive learning over KGs involves predicting new relationships based on existing statements in the KG, using either numerical or symbolic learning models. The Partial Completeness Assumption (PCA) heuristic efficiently guides inductive learning methods for Link Prediction (LP) by refining predictions about absent KG relationships. Nevertheless, numeric techniques– like KG embedding models– alone may fall short in accurately predicting missing information, particularly when it comes to capturing implicit knowledge and complex relationships. We propose a hybrid method named SPaRKLE that seamlessly integrates symbolic and numerical techniques, leveraging the PCA heuristic to capture implicit knowledge and enrich KGs. We empirically compare SPaRKLE with state-of-the-art KG embedding and symbolic models, using established benchmarks. Our experimental outcomes underscore the efficacy of this hybrid approach, as it harnesses the strengths of both paradigms. SPaRKLE is publicly available on GitHub1.
%@ 9798400701412
@inproceedings{Purohit2023,
abstract = {Knowledge graphs (KGs) naturally capture the convergence of data and knowledge, making them expressive frameworks for describing and integrating heterogeneous data in a coherent and interconnected manner. However, based on the Open World Assumption (OWA), the absence of information within KGs does not indicate falsity or non-existence; it merely reflects incompleteness. Inductive learning over KGs involves predicting new relationships based on existing statements in the KG, using either numerical or symbolic learning models. The Partial Completeness Assumption (PCA) heuristic efficiently guides inductive learning methods for Link Prediction (LP) by refining predictions about absent KG relationships. Nevertheless, numeric techniques– like KG embedding models– alone may fall short in accurately predicting missing information, particularly when it comes to capturing implicit knowledge and complex relationships. We propose a hybrid method named SPaRKLE that seamlessly integrates symbolic and numerical techniques, leveraging the PCA heuristic to capture implicit knowledge and enrich KGs. We empirically compare SPaRKLE with state-of-the-art KG embedding and symbolic models, using established benchmarks. Our experimental outcomes underscore the efficacy of this hybrid approach, as it harnesses the strengths of both paradigms. SPaRKLE is publicly available on GitHub1.},
added-at = {2024-02-15T15:30:45.000+0100},
address = {New York, NY, USA},
author = {Purohit, Disha and Chudasama, Yashrajsinh and Rivas, Ariam and Vidal, Maria-Esther},
biburl = {https://www.bibsonomy.org/bibtex/20568ea32110bb41e50b2590f8b4ec46c/l3s},
booktitle = {Proceedings of the 12th Knowledge Capture Conference 2023},
day = 5,
doi = {10.1145/3587259.3627547},
interhash = {c04bcdd8ae80f0ab58b118b5db59a7ee},
intrahash = {0568ea32110bb41e50b2590f8b4ec46c},
isbn = {9798400701412},
keywords = {myown from:gabydler},
location = {<conf-loc>, <city>Pensacola</city>, <state>FL</state>, <country>USA</country>, </conf-loc>},
month = {12},
pages = {44–52},
publisher = {Association for Computing Machinery},
series = {K-CAP '23},
timestamp = {2024-02-15T15:30:45.000+0100},
title = {SPaRKLE : Symbolic caPtuRing of knowledge for Knowledge graph enrichment with LEarning},
url = {https://doi.org/10.1145/3587259.3627547},
year = 2023
}