Design And Implementation Of Automated Scoring System For English Essay Based On Expert Assistance

In the comprehensive test of the English application ability,the essay is an extremely important question,which is helpful to fully understand the students' grasp of the English.At present,even if paperless marking is introduced in the examination,it can't solve the problem of inconsistent scoring standards caused by subjective factors of manual marking,and the objective equity of the evaluation process is difficult to be fully satisfied.Automated essay scoring(AES)is helpful to establish a consistent scoring standard for marking work.This scoring standard is no longer implemented by subjective "people",but by objective machines to ensure the consistency of the whole marking process.But because of lots of training data and poor migratory,it is difficult to fit the actual needs.In this paper,based on the current paperless marking process,the automatic scoring of essay is introduced and a set of automatic scoring method based on expert aided English essay is put forward,which can suit for the actual needs well.In paperless marking,to solve the problem that the scoring standard of the obtained sample volume can not meet the actual needs,a filter method for the sample candidate set is given,which can be used to get the sample volume candidate set distributed in the same way as the paper database.So,the sample set that satisfies the actual demand of the score can be obtained.In this paper,analyzed existing AES methods,a suitable method is selected for the training of AES model.Based on this model,an AES system for English is designed and implemented.Through sufficient experimental,it proves that the proposed expert-based AES model is feasible,and the designed AES system is available.