Design of English Translation Computer Intelligent Scoring System Based on Natural Language Processing

Abstract

Abstract With the development and maturity of automatic scoring technology of English composition, the computer-aided composition marking system based on automatic scoring technology has begun to enter colleges and universities to assist English writing teaching. However, there are still many problems in the current scoring system, such as long running time and large deviation of scoring results, which requires us to design an English translation computer intelligent scoring system based on natural language processing. Through the system, we can reduce the workload of manual scoring, which will improve the efficiency of scoring. Therefore, we need to construct the structure of English translation scoring system, including translation data collection module, information feature extraction module, analysis model construction module and result feedback scoring module. By building a language model, the system can translate the probability distribution of specific sentences or word sequences. As an adaptive learning model, BP neural network has more advantages in dealing with the relationship between complex variables. Through BP network model, the system can extract the feature information of translation translation and translation training set. Through fitting calculation, the system will realize the intelligent scoring of English translation.