Machine Translation Post-Editing Types

Machine Translation Post-Editing Types

Reading Time: 2 minutes

Post Editing is the next step after completing the machine translation (MT) process and evaluating its output. A human translator processes the document to verify that the source and target texts convey the same information and that the tone of the translation is consistent with the original document. The quality of machine translation varies and affects the subsequent effort required for post editing. There are contributory factors to the quality of the MT such as the clarity and quality of the source text; it is important to make sure that the source text is well-written and well-suited for machine translation beforehand. Other considerable factors that affect MT output quality include: the type of MT used, and the compatibility of the source and target languages.

There are two types or levels of post editing, light and full. The post editing process varies greatly in terms of turnaround time, and cost:

1- Light post-editing involves proofreading and minimal editing of the MT output to make sure the translation is grammatically correct and understandable. A light post-edit process focuses on lexical errors and syntax errors inherent in the MT. The realm of machine translation light post-editing tasks include correcting typos, errors in grammar and word usage errors. The process may involve rewriting sentences or changing phrases to make the text more understandable and the meaning clear. Thus light post editing will improve the quality of the MT by correcting machine errors, and deleting the unnecessary additional translation options from the document. The main goals of the post-editing process are to make sure that the meaning of the source document is properly conveyed and that the translation is free of critical errors. For conscientious, detail oriented linguists, not striving for perfection, can be the biggest challenge in doing light post-editing work. Again, light post-editing of machine translation output with extremely poor quality cannot be expected to bring the target text up to publishable quality standards.

2- Full post-editing is a more in-depth process and the goal is to achieve a high quality translation that closely mimics a document originally written in the target language. While light post-editing focuses on lexical errors and syntax errors inherent in the MT, machine translation full post-editing expands to correct less obvious errors, insure appropriate style, and achieve comparable fluency. The full post-editing process includes all the features of light post-editing plus additional tasks to perfect the translated document. These additional tasks include verifying that terminology is translated correctly and consistently throughout the document and making sure that technical terms are properly translated and standardised when compared to other translations in the same subject or field. The linguist is also responsible to make sure that the source and target document contain all the same information by carefully comparing the documents. The translation must use appropriate idioms, appropriate cultural references, and the syntax and style must all rival documents written by native speakers of the target language. The goal is to produce a translation that is the same quality as a document that was translated from start to finish by a human. It is the hope that the post-editing process following MT produces a translated document faster than the turnaround time required for human translation.

The level of post-editing adopted for an MT project depends on two factors: 1) the evaluation of the machine translation output to be post-edited, and 2) the instructions received from the client. If no post-editing approach is specifically requested by the client, then MT full post-editing should be applied.



Print Friendly, PDF & Email
Spread Knowledge
  • 6
Comments are closed.