Statistical machine translation method and system based on dependency tree

Abstract

The invention provides a statistical machine translation method based on a dependency tree. According to transformation rules extracted from a bilingual corpus, each dependency side of the dependency tree of source language sentences is transformed into corresponding target language phase dependency sides, and the obtained target language phase dependency sides are spliced to generate a target language translation. The method combines the advantages of a dependency syntax model and adopts a mode of analysis-transformation-generation to divide a translation process into three stages, the three processes can be respectively and independently modeled and the more accurate control of the generation process of target language sentences becomes possible. The transformation based on dependency sides reserves more knowledge, can tolerate a higher syntax non-isomorphism phenomenon and can obtain performance better than that of the current mainstream translation method based on phase models.

Description

technical field [0001] The invention belongs to the technical field of natural language processing, and in particular relates to a statistical machine translation method based on a dependency tree. Background technique [0002] Statistical machine translation is a hot topic in recent years. Along with its development process, it can be roughly divided into three categories: word-based translation, phrase-based translation and syntax-based translation. Although most of the current mainstream translation systems still use phrase-based translation models, syntax-based translation models have received more and more attention in recent years. Compared with the translation model based on words or phrases, the translation model based on syntax has both grammatical and semantic information, shows a better long-distance sequencing ability, and can better generalize the hierarchical structure of the language. modeling. But most syntax-based translation models (e.g., dependency pars...

AI Technical Summary

Problems solved by technology

This type of method directly maps the source language dependency tree to the target language string, and cannot reorder the fragments of the generated target language sentence

Compared with the phrase-based translation model, its translation accuracy and flexibility still need to be improved

Images