Research And Implementation Of Academic Performance Prediction System For English Majors'

With the rapid development of the Internet,modern education is no longer limited to time and space constraints.The way of education has changed,and also produced a large number of education-related data.How to use computer technology to make education more reasonable,more efficient,more targeted,it draws more and more domestic and foreign educational researchers' and related educational institutions' attention.The development of data mining technology makes a large number of accumulated educational data obtained in-depth analysis and research.The use of data mining technology in the field of education and teaching scene is called educational data mining(Educational Data Mining,referred to as EDM).Education Data mining is an interdisciplinary subject involving computer,Internet,pedagogy,statistics,psychology and other research fields.Predicting students' academic achievement is a typical research direction in educational data mining,which has high educational significance and academic research significance.With the deepening of reform and opening up and the impact of globalization,our country is paying more and more attention to English education.Different with other disciplines in higher education,accordance to the current instructions of the Ministry of Education,Chinese students began to have English courses from the third year of primary school,until the university graduate,and each student basically have to accept twelve years of English education at least.Students with different regions,different backgrounds and different characteristics access different English educational resources before entering the university.It results in college students' learning strategies and professional ability is different after entering the university.However the lack of understanding of the characteristics of students lead to the teacher is difficult to carry out targeted teaching and guidance,and also difficult to identify the potential students and students with high risk.At present,the study of English education in the field of educational data mining mainly focuses on the influence of language test on teaching,the relationship between some characteristics of students and English skills,such as reading and dictation.Research results can only be used to analyze specific data,and the universality is weak.In addition,it still no form an intelligent analytical teaching platform.Based on this,this paper analyzes the family background,socioeconomic status of English majors,learning-related data,achievement motivation,English learning diary,network log,and consumption data to predict the English majors' professional ability and undergraduate academic performance.Whether it predicts that whether English majors' may fail in TEM-4(Test for English Majors-Band 4),whether they can get excellent scores,and whether they may at high risk in their undergraduate studies.In order to improve the accuracy of the prediction,the author has processed the original data.Including data cleansing,data transformation and data reduction,and feature extraction of unstructured data.The author adds new features to the model in a certain order,and combines the six different types of dataset of students.By taking the optimal combination of features,the precision and recall rate reach over 75%.This study shows that English majors' academic performance can be predicted by English educational background,family background,school performance,network access log.The research results have certain practical application value.The dataset collected in this is diversified,we takes into account the characteristics of English learning and the factors that may affect students' academic performance.The theoretical approach is portable,and forecasting system developed according to this research has a high practicality,reusability,scalability.