Design And Implementation Of Student Personalized Evaluation System For Software Engineering Courses

Student learning evaluation is the use of certain technical means to interpret the quality or status of students’ learning through specific indicators.The new era of education evaluation reform points out the need to improve student evaluation in scientific,moral,educational,and academic aspects.At the same time,with the deep integration of information technology and curriculum teaching,blended learning has become a popular and effective teaching method in universities.In response to the problem of insufficient personalization in student evaluation under the current blended learning mode,this paper designs a personalized student evaluation system for the software engineering course,combining the Zone of proximal development and Outcome based education concepts to provide targeted evaluations of students from multiple dimensions.This helps to some extent to solve the problem of insufficient personalization in student learning evaluation and improve teaching quality.The main contributions of this paper include:(1)Generating data-driven personalized student evaluation reports.The data-driven personalized evaluation report generation algorithm designed in this paper integrates cuttingedge educational theories such as Outcome based education and Zone of proximal development,Personalized evaluation of students is carried out from three dimensions:knowledge mastery,course literacy,and learning style.Multi-dimensional feature construction is achieved based on learning data,and a fusion model using multiple machine learning algorithms such as KNN,Naive Bayes,Decision tree,SVM,Random forest,Xgboost,etc.is used to analyze and design student knowledge mastery.A rating rule algorithm based on student data is used to design course literacy and learning style evaluation.Generate highquality personalized student evaluation reports,which improved teaching quality and student classroom satisfaction.(2)Generating student performance prediction reports.In the design of score prediction algorithm,a variety of machine learning algorithm models are used to carry out model prediction contrast experiments based on multi-dimensional student characteristics.Based on this,a performance prediction integration fusion model based on Decision tree,Random forest and MLP feedforward neural network is designed and implemented,and the accuracy of student score prediction is high.(3)Completing system development and teaching application experiments.The latest technology frameworks are used to code and implement the personalized student evaluation system,which is applied to the "Software Engineering" course teaching project.The application results show that the personalized evaluation system interface for students in software engineering courses is simple and intuitive,easy to operate,the system runs normally,and the evaluation and prediction results for students are in line with the design expectations.After application,the average score of student chapters and classroom satisfaction have been improved,and students have high accuracy and satisfaction with the evaluation report.The system application effect is good,and it has promotion value.Through the above research,a personalized student evaluation system for the software engineering course is realized,which has a high degree of personalization in evaluation and can solve the problem of insufficient personalization in student evaluation to some extent.Its related applications can help teachers teach according to students’ aptitudes,improve teaching quality,and strengthen positive feedback effects on students,thereby enhancing students’ satisfaction with software engineering course teaching and stimulating their motivation for independent learning.It provides a solution for continuously improving outcome based personalized teaching design.