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

TL;DR

This paper introduces NTX, a multilingual dataset covering diverse temporal and numerical expressions across 14 languages, along with a rule-based baseline system for extraction, normalization, and resolution tasks in NLP.

Contribution

It provides the first comprehensive multilingual dataset and a baseline system for temporal and numerical expression extraction and normalization in 14 languages.

Findings

Dataset covers 14 languages with diverse expressions

Baseline system offers strong comparison point for future models

Enables improved multilingual temporal and numerical understanding

Abstract

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 \url{https://aka.ms/NTX}.