Simulation of Improved GLR Algorithm in Abnormal Diagnosis Model of English Translation Robot

With the continuous development of artificial intelligence technology, machine translation has been widely used in cross-language communication, global communication and other fields. However, there are some abnormal problems in the practical application of English translation robot, such as inaccurate machine translation results and the model can’t identify uncommon words. Firstly, this paper proposes an improved generalized probabilistic grammar (GLR) algorithm for the abnormal diagnosis model of English translation robot, and verifies it with simulation experiments. Firstly, the shortcomings and improvements of the current anomaly diagnosis model of translation robot are introduced, including the problems of accuracy and efficiency. Then, the traditional GLR algorithm is improved, and the prediction method based on deep learning is adopted, and the introduction of language model is added to improve its accuracy. In the experiment, the real abnormal data set is used for simulation, and the performance of the traditional GLR algorithm and the improved algorithm is compared. The results show that the improved GLR algorithm has significantly improved the accuracy and efficiency. Compared with the traditional method, the average accuracy is improved by about 6%, and the average processing time is shortened by about 30%. Therefore, the improved GLR algorithm can achieve better results and performance in the abnormal diagnosis model of English translation robot.