2026-07-24T06:45:31Z https://drops.dagstuhl.de/oai

oai:drops-oai.dagstuhl.de:19094 2024-03-12T12:01:38Z ddc:004 open_access

Discovering Predictive Dependencies on Multi-Temporal Relations Amico, Beatrice Combi, Carlo Rizzi, Romeo Sala, Pietro temporal databases temporal data mining functional dependencies In this paper, we propose a methodology for deriving a new kind of approximate temporal functional dependencies, called Approximate Predictive Functional Dependencies (APFDs), based on a three-window framework and on a multi-temporal relational model. Different features are proposed for the Observation Window (OW), where we observe predictive data, for the Waiting Window (WW), and for the Prediction Window (PW), where the predicted event occurs. We then discuss the concept of approximation for such APFDs, introduce two new error measures. We prove that the problem of deriving APFDs is intractable. Moreover, we discuss some preliminary results in deriving APFDs from real clinical data using MIMIC III dataset, related to patients from Intensive Care Units. Schloss Dagstuhl – Leibniz-Zentrum für Informatik Beatrice Amico and Carlo Combi and Romeo Rizzi and Pietro Sala 2023 Is Part Of LIPIcs, Volume 278, 30th International Symposium on Temporal Representation and Reasoning (TIME 2023) InProceedings Text doc-type:ResearchArticle publishedVersion application/pdf doi:10.4230/LIPIcs.TIME.2023.4 urn:nbn:de:0030-drops-190945 https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.TIME.2023.4 eng https://creativecommons.org/licenses/by/4.0/legalcode