Exposing wiktionary translations with performance in mind

Abstract

In this paper, we explore some of the challenges to extracting Wiktionary data so it can be used for machine translation. We provide a use case of considerations made in the development of a web service that transforms raw data in a wiktionary xml dump file into a web-service that provides language translation. Our case study focuses on the performance of this service in comparison to a service using remote calls to the Wiktionary API.