Automatic Grammar Correction Method for English Translation Based on Multi-Feature Fusion

In order to improve the automation and intelligence of Chinese-English translation in machine translation, an automatic grammar correction method for English machine translation based on multi-feature fusion and semantic feature extraction is proposed. Multi-dimensional feature fusion corpus construction method is adopted to construct the context mapping model of automatic grammar error correction in English machine translation, and continuous text corpus analysis method is adopted to establish the semantic tree of automatic grammar error correction in English machine translation. The feature quantity of Chinese-English translation grammar standard sentence-level coding and decoding is extracted, and the relevant information of English translation sentence is analyzed according to different combinations of grammar standard sentence-level feature distribution. Multidimensional feature fusion method is adopted to establish the grammar error correction text library of single sentence in Chinese-English translation. According to the semantic modification target in the grammar error correction text library of single sentence, automatic coding and adjustment of grammar structure are carried out to realize automatic grammar error correction and the adjustment of grammatical subject-predicate consistency in English machine translation. The feature quantity of each clause’s grammar error correction modeling parameter is calculated, and the positive sample fusion algorithm is used for automatic optimization to realize automatic grammar error correction in English machine translation. The simulation results show that the accuracy of automatic grammar correction in English machine translation is high, and the correlation degree of translation calibration is strong.