Abstract:GLR algorithm model translation recognition results have data points that overlap, and accuracy cannot be effectively guaranteed. In order to accurately identify phrases, an optimized GLR algorithm based on intelligent recognition is designed. This algorithm constructs a corpus of phrases with a scale of about 740,000 English-Chinese words, makes the phrases searchable, and constructs the phrase structure through the center of the phrase to obtain part-of-speech recognition As a result, the ambiguity between English and Chinese structures in the part-of-speech recognition results was corrected according to the syntactic function of the analytical linear table, and finally the content of recognition was obtained. The actual evaluation results show that the algorithm overcomes the disadvantages of GLR. Compared with statistical algorithms and dynamic memory algorithms, it improves the operation speed and processing performance, is more suitable for machine translation tasks, and provides new ideas in the field of intelligent machine translation.
