Readability Classification and Prediction of Machine Translation Texts Based on Machine Learning Algorithms

Against the backdrop of the continuous acceleration of global communication and cross-border dissemination of digital content, machine translation has become a core tool to break down language barriers and is widely applied in various scenarios such as cross border business communication, academic literature sharing, and multilingual media dissemination. At present, most mainstream machine translation systems focus on ensuring semantic accuracy, but often neglect the key dimension of text readability, that is, whether the translation can conform to the reading habits, cognitive levels and specific scene requirements of the target language users. To make up for the deficiencies of the existing models, this paper proposes a BiLSTM classification algorithm based on the Red-billed blue magpie algorithm and the Attention mechanism optimization. The research first conducts violin graph analysis and correlation analysis, and then compares the performance of the proposed algorithm with that of various machine learning algorithms. The experimental results show that the RBMO-Attention-BiLSTM algorithm proposed in this paper exhibits the optimal performance in all evaluation indicators. Compr