Design of Machine Translation Information Processing System Based on AI Technology

Artificial intelligence (AI) has greatly improved the quality of machine translation. In order to promote translation teaching, this article proposed machine translation teaching strategies for teachers and machine translation self-learning strategies for students. Considering machine translation as a way to improve translation skills, it is also necessary to use it wisely to enhance basic English skills such as reading, writing, and vocabulary. In response to the issue of recognition accuracy in English reading recognition ability grading, this article combined CNN (Convolutional Neural Networks) algorithm and LSTM (Long Short-Term Memory) algorithm to construct a CNN-LSTM network model to learn the characteristics of historical reading articles in groups, analyze students' English reading ability, and assist students in machine translation. The experimental results showed that the accuracy of reading classification based on CNN-LSTM algorithm was between 88% and 96%. The system can learn a large amount of language data and continuously optimize the translation algorithm to improve the accuracy and fluency of translation.