Translation engine optimization system and method based on clustering evolution

Abstract

The invention provides a translation engine optimization system and method based on clustering evolution, a corresponding computer readable storage medium and an OpenKiwi evaluation system, and belongs to the technical field of translation. The engine optimization system comprises a document input subsystem, a clustering subsystem in output connection with the document input subsystem, and an automatic evaluation model system in feedback connection with the clustering subsystem. According to the method, the evaluation matrix is introduced, internal basic data can be iterated continuously basedon characteristic root judgment of the evaluation matrix, the quality and effectiveness of the basic data are guaranteed, and the best engine of a certain category or the best sequence of several engines is mapped through the engine mapping table; meanwhile, openkiwi and manual feedback intervention are introduced, so that the whole evaluation process is objective and accurate, priori rules are not needed for optimization and evolution, and therefore, an optimal translation engine is evolved / selected.

Description

technical field [0001] The invention belongs to the technical field of translation, and in particular relates to a translation engine optimization system and method based on clustering evolution, a corresponding computer-readable storage medium and an OpenKiwi evaluation system. Background technique [0002] In the era of hot artificial intelligence, there are many machine translation engines. How translators should choose the translation engine that suits them has become a technical problem that needs to be solved urgently. At the same time, how do existing translation engines continue to evolve and choose based on actual translation results and human feedback? , has become an important factor restricting translation efficiency. [0003] In recent years, in the field of natural language research, evaluation issues have received more and more attention. It can be said that evaluation is the core and key part of the entire natural language field. The evaluation of machine tr...

AI Technical Summary

Problems solved by technology

In the prior art, automatic evaluation of translation engines is usually carried out directly. Although the evaluation process is automated, the evaluation results largely depend on the standards and algorithms used in the automatic evaluation process, and most of these standards and algorithms They all pass the threshold of the prior rules, which in fact makes the evaluation results highly subjective, resulting in low accuracy of the selection results and questionable objectivity

For example, the traditional automatic evaluation method based on reference translations is like this, which is difficult to meet the needs of technological development and users

Images