Determining Event Durations: Models and Error Analysis

Vempala, Alakananda; Blanco, Eduardo; Palmer, Alexis

doi:10.18653/v1/n18-2026

Cited by 16 publications

(18 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To the best of our knowledge, there are no other available systems for the "stationarity", "typical time", and "frequency" phenomena studied here. As for "duration" and "temporal order", there are existing systems (e.g., Vempala et al (2018); Ning et al (2018b)), but they cannot be directly applied to the setting in MCTACO where the inputs are natural languages.…”

Section: Methodsmentioning

confidence: 99%

“Going on a vacation” takes longer than “Going for a walk”: A Study of Temporal Commonsense Understanding

Zhou¹,

Khashabi²,

Ning³

et al. 2019

Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conferen

107

105

View full text Add to dashboard Cite

Understanding time is crucial for understanding events expressed in natural language. Because people rarely say the obvious, it is often necessary to have commonsense knowledge about various temporal aspects of events, such as duration, frequency, and temporal order. However, this important problem has so far received limited attention. This paper systematically studies this temporal commonsense problem. Specifically, we define five classes of temporal commonsense, and use crowdsourcing to develop a new dataset, MCTACO , that serves as a test set for this task. We find that the best current methods used on MCTACO are still far behind human performance, by about 20%, and discuss several directions for improvement. We hope that the new dataset and our study here can foster more future research on this topic. 1 1 The dataset, annotation interfaces, guidelines, and qualification tests are available at:

show abstract

Section: Methodsmentioning

confidence: 99%

“Going on a vacation” takes longer than “Going for a walk”: A Study of Temporal Commonsense Understanding

Zhou¹,

Khashabi²,

Ning³

et al. 2019

Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conferen

107

105

View full text Add to dashboard Cite

show abstract

“…This makes the current work the first to allow analysis of symmetry of temporal uncertainty, and the first to annotate such complete durations. Various methods have been proposed to predict coarse-grained event durations in the TimeBank corpus [10], [11], [39], [40], for which the state of the art is a Long Short-Term Memory (LSTM) network ensemble [41], which we retrain on our data and adopt as a baseline.…”

Section: B Event Durationmentioning

confidence: 99%

“…To predict event durations, we use a simple model, taking as input the event, and its local left and right context (size: 1), as this has shown to be effective features for estimating event duration [41]. We encode the event and its context using either an LSTM [47] or CNN 12 [48].…”

Section: B Event Durationsmentioning

confidence: 99%

“…1) Event Duration Baseline (D-LSTM): As this is the first clinical corpus to annotate full event durations, as a baseline used to predict event duration we implement the current state-of-the-art model for news texts by [41]. Their model is an LSTM-ensemble built on top of GloVe embeddings.…”

Section: B Baselinesmentioning

confidence: 99%

“…To adapt their model to the clinical domain, we retrain the GloVe embeddings on 100M words of clinical reports from MIMIC III [17]. As [41] only classify events into two duration categories: ≤ a day, and > a day, instead of a binary softmax output on top of its event encoder, we use three regression layers, explained in section VI-D, to predict the duration mode, and its left and right deviations.…”

Section: B Baselinesmentioning

confidence: 99%

See 2 more Smart Citations

Towards Extracting Absolute Event Timelines From English Clinical Reports

Leeuwenberg

Moens

2020

IEEE/ACM Trans. Audio Speech Lang. Process.

View full text Add to dashboard Cite

Temporal information extraction is a challenging but important area of automatic natural language understanding. Existing approaches annotate and extract various parts of the temporal information conveyed in language like relative event order, temporal expressions, or event durations. Most schemes focus primarily on annotation of temporally certain (often explicit) information, resulting in partial annotation, and underrepresentation of implicit information. In this paper, we propose an approach towards extraction of more complete (implicit and explicit) temporal information for all events, and obtain probabilistic absolute event timelines by modeling temporal uncertainty with information bounds. As a case study, we use our scheme to annotate a set of English clinical reports, and propose and evaluate a multi-regression model for predicting probabilistic absolute timelines, obtaining promising results.

show abstract