Temporal and numerical expression understanding is of great importance in
many downstream Natural Language Processing (NLP) and Information Retrieval
(IR) tasks. However, much previous work covers only a few sub-types and focuses
only on entity extraction, which severely limits the usability of identified
mentions. In order for such entities to be useful in downstream scenarios,
coverage and granularity of sub-types are important; and, even more so,
providing resolution into concrete values that can be manipulated. Furthermore,
most previous work addresses only a handful of languages. Here we describe a
multi-lingual evaluation dataset - NTX - covering diverse temporal and
numerical expressions across 14 languages and covering extraction,
normalization, and resolution. Along with the dataset we provide a robust
rule-based system as a strong baseline for comparisons against other models to
be evaluated in this dataset. Data and code are available at
https://aka.ms/NTX.

本文描述了一种多语言评估数据集 - NTX，该数据集涵盖了 14 种语言中的不同时间和数字表达式，覆盖了提取、规范化和解析，并提供了一种强大的基线对比其他模型在此数据集中的评估。

跨语言时间与数字表达式抽取与规范化的数据集与基线系统

Dataset and Baseline System for Multi-lingual Extraction and  Normalization of Temporal and Numerical Expressions

This paper describes a temporal expression identification and normalization
system, ManTIME, developed for the TempEval-3 challenge. The identification
phase combines the use of conditional random fields along with a
post-processing identification pipeline, whereas the normalization phase is
carried out using NorMA, an open-source rule-based temporal normalizer. We
investigate the performance variation with respect to different feature types.
Specifically, we show that the use of WordNet-based features in the
identification task negatively affects the overall performance, and that there
is no statistically significant difference in using gazetteers, shallow parsing
and propositional noun phrases labels on top of the morphological features. On
the test data, the best run achieved 0.95 (P), 0.85 (R) and 0.90 (F1) in the
identification phase. Normalization accuracies are 0.84 (type attribute) and
0.77 (value attribute). Surprisingly, the use of the silver data (alone or in
addition to the gold annotated ones) does not improve the performance.

本文介绍了一个用于 TempEval-3 挑战的时间表达式识别和归一化系统 ManTIME，其中识别阶段结合了条件随机场和后处理识别管道，而规范化阶段则使用 NorMA，探讨了不同特征类型对性能的影响。最好的运行在标识阶段达到 0.95（P），0.85（R）和 0.90（F1），规一化精度分别为 0.84（类型属性）和 0.77（值属性）