基于改进GLR算法的智能识别英语翻译模型设计

Design of intelligent recognition English translation model based on improved GLR algorithm

摘要:

GLR算法模型翻译识别结果存在数据点重合的情况，精确度无法得到有效保障。为了准确的识别短语，设计了基于改进GLR算法的短语智能识别算法，该算法构建标记规模约74万个英汉单词的短语语料库，使短语具备可搜索功能，通过短语中心点构建短语结构，可获得词性识别结果，依据解析线性表的句法功能校正词性识别结果中的英汉结构歧义，最终获得识别的内容。实际测评结果显示，该算法克服了GLR的弊端，相对统计算法和动态记忆算法提高了运算速度和处理性能，更加适合机器翻译任务，为在智能机器翻译领域提供了新的思路。

Abstract:

GLR algorithm model translation recognition results have data points that overlap, and accuracy cannot be effectively guaranteed. In order to accurately identify phrases, an optimized GLR algorithm based on intelligent recognition is designed. This algorithm constructs a corpus of phrases with a scale of about 740,000 English-Chinese words, makes the phrases searchable, and constructs the phrase structure through the center of the phrase to obtain part-of-speech recognition As a result, the ambiguity between English and Chinese structures in the part-of-speech recognition results was corrected according to the syntactic function of the analytical linear table, and finally the content of recognition was obtained. The actual evaluation results show that the algorithm overcomes the disadvantages of GLR. Compared with statistical algorithms and dynamic memory algorithms, it improves the operation speed and processing performance, is more suitable for machine translation tasks, and provides new ideas in the field of intelligent machine translation.