Research on Segmented Translation Algorithm of Korean Complex Long Sentences Based on Feature Extraction

In the practice of Korean-Chinese translation, the translation of Korean long sentences is very common, and the problems in the translation process are particularly prominent. Traditional machine translation is difficult to obtain accurate translation results. However, at present, there is little research on translation skills and strategies of Korean long sentences. This paper uses some skills and strategies of English long sentence translation for reference, combined with translation difficulties encountered in translation practice, and takes sentence structure, grammar paradigm and key words as the breakthrough points. By analyzing the sentence components, sentence structures and key words of various long sentences, this paper studies the translation skills of Korean long sentences into Chinese. The algorithm uses the concept hierarchy network theory to preprocess the feature semantic blocks, divides clauses based on semantic features and logical concept definitions, uses the regular translation system to translate clauses, and obtains the translation results of the whole sentences through order adjustment and combination. The results show that the translation algorithm can effectively improve the translation effect of complex long sentences.