In the traditional matching methods, the correlation between the input noun and the translated noun is not the best match, which leads to the low recall and accuracy of the noun translation. This paper introduces a computer-aided method for the automatic translation of college English nouns. English proper nouns are preprocessed and English parallel corpus is constructed to provide corpus for machine assistance. The nouns input by users are divided into blocks, word meaning retrieval is carried out, word meaning is matched according to word meaning correlation coefficient and editing distance, word meaning matching evaluation system is established, and the best matching translation is selected. Taking English proper nouns as an example, the experimental results show that the method proposed in this paper not only improves the recall rate and accuracy of translation, but also improves the harmonic average of the two indicators, so that the translation of proper nouns has a higher accuracy.