Multi-lingual Geoparsing based on Machine Translation

TL;DR

This paper presents a multi-lingual geoparsing approach that leverages machine translation and monolingual tools to identify location names across various languages, reducing development costs and enabling a unified interface.

Contribution

The paper introduces a novel multi-lingual geoparsing method that combines machine translation with monolingual tools, achieving comparable accuracy across languages without developing separate geoparsers.

Findings

Geoparsing accuracy for Chinese and Arabic is comparable to English.

Machine translation-based geoparsing matches manual translation accuracy.

Method reduces development time and cost for multi-language geoparsing.

Abstract

Our method for multi-lingual geoparsing uses monolingual tools and resources along with machine translation and alignment to return location words in many languages. Not only does our method save the time and cost of developing geoparsers for each language separately, but also it allows the possibility of a wide range of language capabilities within a single interface. We evaluated our method in our LanguageBridge prototype on location named entities using newswire, broadcast news and telephone conversations in English, Arabic and Chinese data from the Linguistic Data Consortium (LDC). Our results for geoparsing Chinese and Arabic text using our multi-lingual geoparsing method are comparable to our results for geoparsing English text with our English tools. Furthermore, experiments using our machine translation approach results in accuracy comparable to results from the same data that was translated manually.