An Improved Back-Propagation Neural Network for the Prediction of College Students’ English Performance

Wei Liu

Abstract


The global economic boom has greatly boosted the need for communication be-tween different cultures and difference countries. The effective communication requires good command of foreign languages, especially English. This paper highlights the necessity to predict the English performance of college students, and sums up the types and features of neural network (NN) models. On this ba-sis, the backpropagation (BP) NN was selected to predict the English perfor-mance of college students. The Spearman’s R correlation test was conducted to analyze how the English performance is affected by the following factors: the score in National College Entrance Examination (NCEE), gender, age and learn-ing attitude. Then, the improved BPNN was adopted to predict the English per-formance of college students. The results show that the NCEE score has the greatest impact on English performance, followed in descending order by learn-ing attitude and gender, while age does not greatly affect English scores; the im-proved BPNN achieved a desirable effect in predicting the English performance of college students. The research findings shed new lights on college English teachers and learners.

Keywords


neural network (NN); English performance; prediction; error

Full Text:

PDF


Copyright (c) 2019 Wei Liu


International Journal of Emerging Technologies in Learning (iJET) – eISSN: 1863-0383
Creative Commons License
Indexing:
Scopus logo Clarivate Analyatics ESCI logo EI Compendex logo IET Inspec logo DOAJ logo DBLP logo Learntechlib logo EBSCO logo Ulrich's logo Google Scholar logo MAS logo