Influence Functions for interpretable link prediction in Knowledge Graphs for Intelligent Environments

Vista sencilla de ítem
dc.contributor.authorZulaika Zurimendi, Unai
dc.contributor.authorAlmeida, Aitor
dc.contributor.authorLópez de Ipiña González de Artaza, Diego
dc.date.accessioned2025-11-03T09:14:33Z
dc.date.available2025-11-03T09:14:33Z
dc.date.issued2022
dc.date.updated2025-11-03T09:14:33Z
dc.descriptionPonencia presentada en la 2022 7th International Conference on Smart and Sustainable Technologies (SpliTech), celebrada en Split y Bol, entre el 5 y el 8 de julio de 2022es
dc.description.abstractKnowledge graphs are large, graph-structured databases used in many use-case scenarios such as Intelligent Environments. Many Artificial Intelligent latent feature models are used to infer new facts in Knowledge Graphs. Despite their success, the lack of interpretability remains a challenge to overcome. This paper applies influence functions to obtain the most significant facts when predicting new knowledge and allows users to understand these models. However, Influence Functions do not scale well. We present an efficient method to scale up influence functions to large Knowledge Graphs to overcome such an issue. It drastically reduces the number of training samples when computing influences and uses fast curvature matrix-vector products to linearize the computation steps required for the inverse Hessian. We conduct experiments on different sized Knowledge Graphs demonstrating the scalability of our approach and its effectiveness in measuring the most influential facts. Our method provides an intuitive understanding of link prediction behaviour in Knowledge Graphs and Intelligent Environments.en
dc.description.sponsorshipWe gratefully acknowledge the support of the Basque Government's Department of Education for the predoctoral funding; the Ministry of Economy, Industry and Competitiveness of Spain under Grant No.: RTI2018-101045-A-C22 (FuturAAL), PEACEOFMIND project (ref. PID2019-105470RB-C31) and the INCEPTION projecten
dc.identifier.citationZulaika, U., Almeida, A., & Lopez-De-Ipina, D. (2022). Influence Functions for interpretable link prediction in Knowledge Graphs for Intelligent Environments. 2022 7th International Conference on Smart and Sustainable Technologies, SpliTech 2022. https://doi.org/10.23919/SPLITECH55088.2022.9854264
dc.identifier.doi10.23919/SPLITECH55088.2022.9854264
dc.identifier.isbn9789532901160
dc.identifier.urihttps://hdl.handle.net/20.500.14454/4208
dc.language.isoeng
dc.publisherInstitute of Electrical and Electronics Engineers Inc.
dc.subject.otherInfluence Functions
dc.subject.otherIntepretability
dc.subject.otherKnowledge Graphs
dc.titleInfluence Functions for interpretable link prediction in Knowledge Graphs for Intelligent Environmentsen
dc.typeconference paper
dcterms.accessRightsmetadata only access
oaire.citation.title2022 7th International Conference on Smart and Sustainable Technologies, SpliTech 2022
Archivos
Colecciones
Ponencias