Integer-Programming Ensemble of Temporal-Relations Classifiers

Files in This Item:
File Description SizeFormat 
IP_Ensemble_TemporalRelationClassifiers_DMKD_Accepted_Revision.pdf602.83 kBAdobe PDFDownload
Title: Integer-Programming Ensemble of Temporal-Relations Classifiers
Authors: Kerr, CatherineHoare, TerriCarroll, PaulaMareček, Jakub
Permanent link: http://hdl.handle.net/10197/11517
Date: 2-Jan-2020
Online since: 2020-08-28T15:51:01Z
Abstract: The extraction of temporal events from text and the classification of temporal relations among both temporal events and time expressions are major challenges for the interface of data mining and natural language processing. We present an ensemble method, which reconciles the outputs of multiple heterogenous classifiers of temporal expressions. We use integer programming, a constrained optimisation technique, to improve on the best result of any individual classifier by choosing consistent temporal relations from among those recommended by multiple classifiers. Our ensemble method is conceptually simple and empirically powerful. It allows us to encode knowledge about the structure of valid temporal expressions as a set of constraints. It obtains new state-of-the-art results on two recent natural language processing challenges, SemEval-2013 TempEval-3 (Temporal Annotation) and SemEval-2016 Task 12 (Clinical TempEval), with F1 scores of 0.3915 and 0.595 respectively.
Funding Details: European Commission Horizon 2020
Type of material: Journal Article
Publisher: Springer
Journal: Data Mining and Knowledge Discovery
Volume: 34
Start page: 533
End page: 562
Copyright (published version): 2020 the Authors
Keywords: Natural language processingTemporal reasoningEnsemble methodsInteger programming
DOI: 10.1007/s10618-019-00671-x
Language: en
Status of Item: Peer reviewed
Appears in Collections:Business Research Collection

Show full item record

Page view(s)

193
Last Week
36
Last month
checked on Sep 22, 2020

Download(s)

48
checked on Sep 22, 2020

Google ScholarTM

Check

Altmetric


This item is available under the Attribution-NonCommercial-NoDerivs 3.0 Ireland. No item may be reproduced for commercial purposes. For other possible restrictions on use please refer to the publisher's URL where this is made available, or to notes contained in the item itself. Other terms may apply.