BLEU has frequently been reported as correlating well with human judgement,[1][2][3] and remains a benchmark for the assessment of any new evaluation metric. There are however a number of criticisms that have been voiced. It has been noted that although … Read More
The Word error rate (WER) is a metric based on the Levenshtein distance, where the Levenshtein distance works at the character level, WER works at the word level. It was originally used for measuring the performance of speech recognition systems, … Read More
Various methods for the evaluation for machine translation have been employed. This article will focus on the evaluation of the output of machine translation, rather than on performance or usability evaluation. Before covering the large scale studies, a brief comment … Read More
The United Nations has published a report on “the world economic situation and perspectives for 2008.” According to the study’s conclusions a slowing is to be expected. It provides an overview of recent global economic performance and short-term prospects for … Read More
