A Two-Phase Ensemble-Based Method for Predicting Learners’ Grade in MOOCs

Abstract

MOOCs are online learning environments which many students use, but the success rate of online learning is low. Machine learning can be used to predict learning success based on how people learn in MOOCs. Predicting the learning performance can promote learning through various methods, such as identifying low-performance students or by grouping students together. Recent machine learning has enabled the development of predictive models, and the ensemble method can assist in reducing the variance and bias errors associated with single-machine learning. This study uses a two-phase classification model with an ensemble technique to predict the learners’ grades. In the first phase, binary classification is used, and the non-majority class is then sent to the second phase, which is multi-class classification. The new features are computed based on the distance from the class’s center. The distance between the data and the center of an overlapping cluster is calculated using silhouette score-based feature selection. Lastly, Bayesian optimization boosts the performance by fine tuning the optimal parameter set. Using data from the HMPC- and the CNPC datasets, the experiment results demonstrate that the proposed design, the two-phase ensemble-based method, outperforms the state-of-the-art machine learning algorithms.

Design and Implementation of Machine Learning Algorithms in Automatic Grading of Students' Assignments

“…By developing prediction models for each cluster, it enhances accuracy in predicting student performance, surpassing traditional semester-grade-based methods. In [14], two-phase ensemble classification model to predict learning success in MOOCs, leveraging machine learning to identify lowperformance students and optimize learning outcomes. By integrating silhouette score-based feature selection and Bayesian optimization, this method surpasses current algorithms in predicting learner grades, thereby improving the effectiveness of online learning.…”

Deep FM-Based Predictive Model for Student Dropout in Online Classes

“…Recent advancements in machine learning have made deep understanding the most advanced method for predicting dropouts [21]. Feature representations that are not linear are automatically utilized by Jiao's deep and fully linked feed-forward neural network [10]. Similar reasoning may be used to [11], who employed a recurrent neural network model with LSTM cells that stored attributes in adjacent states.…”

“…However, the majority of current deep learning models for dropout prediction ignore the potential advantages of factorization methods in favor of recurrent neural networks (RNNs) or long short-term memory (LSTM) networks. An innovative two-phase ensemble-based strategy for forecasting students' grades in MOOCs was put out by [10] in their study. Their method, which combined a Random Forest algorithm with specific distance characteristics, had a high accuracy of 97%.…”

“…DeepFM offers the potential to increase the precision and interpretability of dropout prediction models in educational situations by merging factorization techniques and deep learning. The work by [10] concentrated on utilizing a Random Forest algorithm to analyze and forecast MOOC learners' dropout behavior. Their method identified students who were at danger of dropping out with a 91% accuracy rate.…”

“…The key to reducing dropout rates is to use these indicators to identify students who are likely to drop out and to put focused retention initiatives in place before students start showing drop-out signs. In this study, a ML approach is employed to create prediction models that are then trained with fresh data using data from the information systems of online education institutions [10] [11]. After training, the produced prediction model samples might be used to forecast future dropout behavior.…”