ClearTK-TimeML: A minimalist approach to TempEval 2013

Joint Conference on Lexical and Computational SemanticsPublished 1 June 2013

Steven Bethard

Citations114

TL;DR

The ClearTK-TimeML submission to TempEval 2013 competed in all English tasks: identifying events, identifying times, and identifying temporal relations and ranked 1 st for temporal relation F1, time extent strict F1 and event tense accuracy.

Abstract

The ClearTK-TimeML submission to TempEval 2013 competed in all English tasks: identifying events, identifying times, and identifying temporal relations. The system is a pipeline of machine-learning models, each with a small set of features from a simple morpho-syntactic annotation pipeline, and where temporal relations are only predicted for a small set of syntactic constructions and relation types. ClearTKTimeML ranked 1 st for temporal relation F1, time extent strict F1 and event tense accuracy.

Keywords

Computer Science

SemEval-2013 Task 1: TempEval-3: Evaluating Time Expressions, Events, and Temporal Relations

360 Citations2013Naushad UzZaman, Héctor Llorens +4 more

The participants’ approaches, results, and the observations from the results, which may guide future research in this area are described.

SemEval-2007 task 15

291 Citations2007Marc Verhagen, Robert Gaizauskas +4 more

The TempEval task proposes a simple way to evaluate automatic extraction of temporal relations by defining three sub tasks that allow pairwise evaluation of temporal Relations.

View PDF

SemEval-2010 Task 13: TempEval-2

286 Citations2010Marc Verhagen, R. Supyan Sauri +2 more

Tempeval-2 comprises evaluation tasks for time expressions, events and temporal relations, the latter of which was split up in four sub tasks, motivated by the notion that smaller subtasks would make both data preparation and temporal relation extraction easier.

TIPSem (English and Spanish): Evaluating CRFs and Semantic Roles in TempEval-2

142 Citations2010Héctor Llorens, Estela Saquete +1 more

TIPSem, a system to extract temporal information from natural language texts for English and Spanish, learns CRF models from training data and achieves the best F1 score in all the tasks.

TRIPS and TRIOS System for TempEval-2: Extracting Temporal Information from Text

115 Citations2010Naushad UzZaman, James F. Allen

This paper describes two systems, TRIPS and TRIOS, that were submitted to the TempEval 2 challenge, for extracting temporal information from raw text using a combination of deep semantic parsing, Markov Logic Networks and Conditional Random Field classifiers.

CU-TMP

74 Citations2007Steven Bethard, James Martin

A variety of syntactically and semantically motivated features are introduced, including temporal-logic-based features derived from running the Task B system on the Task A and C data, to approach the temporal relation identification tasks of TempEval 2007 as pair-wise classification tasks.

View PDF

International Journal of Semantic ComputingFINDING TEMPORAL STRUCTURE IN TEXT: MACHINE LEARNING OF SYNTACTIC TEMPORAL RELATIONS

23 Citations2007Steven Bethard, James Martin +1 more

This research proposes and evaluates a linguistically motivated approach to extracting temporal structure from text by using a combination of event-level features like tense and aspect with syntax- level features like the paths through the syntactic tree to identify new temporal relations.