An English Translation Quality Evaluation based on Generative Adversarial Network and Naive Bayes Classification

In this era, English translation technologies have gained extensive popularity in today’s globalized, transforming daily communication and the translation industry by breaking text and enabling efficient cross-lingual information exchange. However, English translation technologies often struggle to understand the context, leading to potential misinterpretation of visual content. To overcome these issues proposed a classification model of Naive Bayes classifier. The method commences with the gathering input text data from the Cycle Generative Adversarial Network (GAN) dataset, then the relevant features are extracted through Asymmetric GAN it discriminately identifies the English language related text. Then the classification is done through the Naive Bayes classifier. The proposed method effectively identifying high-quality translations is subsequently trained on these extracted features, allowing it to identify English into distinct quality categories. From the results, the proposed Naive Bayes classifier model showed improved results in contrast to existing impact learning by providing metrics values of Peak Signal-To-Noise Ratio (PNSR) with 78.98 decibels (dB) and Structural Similarity Index Measure (SSIM) with value 0.95 respectively compared with Ghosting and star GAN.