Research on Computer–aided Translation of Vector Mixed Semantic Network in Information Age

Abstract

This study aimed at the shortcomings of traditional rule-based machine translation models such as inaccurate English translation and difficulty in accurately describing the relationship between words, designed and improved machine translation models used by semantic networks. This model adopts the semantic algorithm of vector mixing. In the conversion model, the algorithm is adopted for getting the correlation between vectors. After the calculation of weighted vector addition, it becomes easier to identify the character of sentences, obtain English translation results, and implement weight training on the sentences with the main phrases which make up the sentences. The translation results summarize the central idea of the sentence. The semantic network machine translation model is improved to introduce big data according to the needs of users and let linguists participate in the procedure. Thus, English translation results not only explain sentences independently but accurately describe the similarity of vectors. The experimental data illustrate that the framework can translate English accurately and efficiently.