Investigating the Validity and Reliability of a Comprehensive Essay Evaluation Model of Integrating Manual Feedback and Intelligent Assistance

Abstract

How to respond effectively and efficiently to students’ writing and to maximize the potential of feedback to promote student writing skills deserves careful consideration. The use of intelligent algorithms to assist teachers in the manual evaluation of students’ essays is of practical value and importance in the context of "artificial intelligence + education". The existing intelligent evaluation techniques are subject to the interference of many factors such as openness of questions and students' language expression abilities. For this reason, this study conducts a study on comprehensive essay evaluation method with intelligent assistance and manual feedback and on reliability and validity tests. Before the intelligent evaluation, the semantic integrity of students' essays is analyzed, and a semantic integrity analysis model of students' essays based on BERT model is constructed. A fusion similarity algorithm for essay answer key points is proposed by extracting these characteristics that have an impact on the evaluation results, such as technique preferences, paragraph content and paragraph topic of the essays. The Siamese and ESIM networks are combined to propose an intelligent evaluation model for students’ essays, and the model framework and working principle are described in detail. The experimental results verify the effectiveness of the constructed model.