Prediction of Online Learning Outcomes for Vocational College Students using Back Propagation-Artificial Neural Network in Big Data

Vocational college education plays a significant role in promoting the development of the national manufacturing industry. Recently, universities, vocational colleges, and information societies have generated a wealth of big data on research, education learning, student management, and personnel training. With the digital technology advancement, learning analytics and big data have become prevalent in higher education. However, vocational colleges face difficulties in predicting the student’s online learning outcomes due to the different range of student backgrounds, engagement levels, and learning paces with variability in online instructional quality and resources. This research proposes a Backpropagation-Artificial Neural Network (BP-ANN) to accurately predict the vocational college student online learning outcome in big data. Initially, the data is obtained from Vocational education colleges to evaluate the model performance. The data cleaning and transformation process is utilized in pre-processing stage. Then, Pearson’s correlation coefficient selects the appropriate features from the data. At last, BP-ANN is performed to predict the student outcome accurately. The BP-ANN achieves a better prediction accuracy of 98.20% compared to existing techniques like Random Forest (RF), Decision Tree (DT), and ANN.