Design of English Verb Translation Model Based on Improved GLR Algorithm

Verbs play a great role in both English and Chinese, but verbs in English are much more complicated and varied than those in Chinese. The circuitous changes in English grammar are manifested in tense, voice, mood, participle, verb and other forms. Then, the improved GLR (generalized likelihood ratio) algorithm is used to translate the verbs in English into Chinese, and the Chinese translation is verified Then, the central point of the verb is used to construct the structure of the verb, so as to get the type of the word, and to correct the ambiguity of English and Chinese grammar, so as to get the semantics to be recognized. In this method, BPNN (BP neural network) is used to replace the shift and reduction operations in the GLR analysis table, and the input and output are converted into binary representations. The experimental results show that the improved GLR algorithm uses BPNN to replace the analysis table of GLR algorithm, simulates its move-in and reduction action, and generates a correct syntax tree, which makes it more flexible in the process of sentence analysis, and has more obvious advantages for sentences with complex structure and large syntax tree.