Workload-Aware Materialization of Junction Trees

Visa fullständig post



Permalänk

http://hdl.handle.net/10138/346638

Citation

Ciaperoni , M , Aslay , C , Gionis , A & Mathioudakis , M 2022 , Workload-Aware Materialization of Junction Trees . in EDBT: 25th International Conference on Extending Database Technology : EDBT 2022 . Advances in Database Technology , vol. 25 , OpenProceedings.org , pp. 65-77 , International Conference on Extending Database Technology , Edinburgh , United Kingdom , 29/03/2022 . https://doi.org/10.5441/002/edbt.2022.06

Titel: Workload-Aware Materialization of Junction Trees
Författare: Ciaperoni, Martino; Aslay, Cigdem; Gionis, Aristides; Mathioudakis, Michael
Upphovmannens organisation: Department of Computer Science
Algorithmic Data Science
Utgivare: OpenProceedings.org
Datum: 2022
Språk: eng
Sidantal: 13
Tillhör serie: EDBT: 25th International Conference on Extending Database Technology
Tillhör serie: Advances in Database Technology
ISBN: 978-3-89318-086-8
ISSN: 2367-2005
DOI: https://doi.org/10.5441/002/edbt.2022.06
Permanenta länken (URI): http://hdl.handle.net/10138/346638
Abstrakt: Bayesian networks are popular probabilistic models that capture the conditional dependencies among a set of variables. Inference in Bayesian networks is a fundamental task for answering probabilistic queries over a subset of variables in the data. However, exact inference in Bayesian networks is NP-hard, which has prompted the development of many practical inference methods. In this paper, we focus on improving the performance of the junction-tree algorithm, a well-known method for exact inference in Bayesian networks. In particular, we seek to leverage information in the workload of probabilistic queries to obtain an optimal workload-aware materialization of junction trees, with the aim to accelerate the processing of inference queries. We devise an optimal pseudo-polynomial algorithm to tackle this problem and discuss approximation schemes. Compared to state-of-the-art approaches for efficient processing of inference queries via junction trees, our methods are the first to exploit the information provided in query workloads. Our experimentation on several real-world Bayesian networks confirms the effectiveness of our techniques in speeding-up query processing.
Subject: 113 Computer and information sciences
Referentgranskad: Ja
Licens: unspecified
Användningsbegränsning: openAccess
Parallelpublicerad version: publishedVersion
Finansierad av:
Finansierings ID: 322046


Filer under denna titel

Totalt antal nerladdningar: Laddar...

Filer Storlek Format Granska
paper_22.pdf 1019.Kb PDF Granska/Öppna

Detta dokument registreras i samling:

Visa fullständig post