Automated Postediting of Documents

TL;DR

This paper presents an automated postediting system for improving low- to medium-quality English texts, focusing on a portable module for article selection in machine translation, with promising results compared to human performance.

Contribution

It introduces a portable, rule-based postediting module for English article correction in MT outputs, derived from large online text resources.

Findings

Over 200,000 rules automatically derived for article correction

System achieves high accuracy in article selection tasks

Performance compares favorably with human posteditors

Abstract

Large amounts of low- to medium-quality English texts are now being produced by machine translation (MT) systems, optical character readers (OCR), and non-native speakers of English. Most of this text must be postedited by hand before it sees the light of day. Improving text quality is tedious work, but its automation has not received much research attention. Anyone who has postedited a technical report or thesis written by a non-native speaker of English knows the potential of an automated postediting system. For the case of MT-generated text, we argue for the construction of postediting modules that are portable across MT systems, as an alternative to hardcoding improvements inside any one system. As an example, we have built a complete self-contained postediting module for the task of article selection (a, an, the) for English noun phrases. This is a notoriously difficult problem for Japanese-English MT. Our system contains over 200,000 rules derived automatically from online text resources. We report on learning algorithms, accuracy, and comparisons with human performance.