DTED: Evaluation of Machine Translation Structure Using Dependency Parsing and Tree Edit Distance
Year: 2016
Type: conference-paper
Abstract: We present DTED, a submission to the WMT 2016 Metrics Task using structural information generated by dependency parsing and evaluated using tree edit distances.In this paper we apply this system to translations produced during WMT 2015, and compare our scores with human rankings from that year.We find moderate correlations, despite the human judgements being based on all aspects of the sentences while our metric is based only on word order.
DOI: 10.18653/v1/w16-2340
Authors: Martin McCaffery, Mark-Jan Nederhof
Institution: University of St Andrews
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
OpenAlex ID: W2509854663