Abstract:The challenge for today’s learning systems is to provide effective access to knowledge and contents that are well-relevant to learners’ background and interest levels. Majority of personalized educational platforms lack methods to effectively support the needs of learners who are generally heterogeneous in terms of intellectual abilities, learning pace, preferences, academic background, etc. Hence, there is a need to provide powerful mechanisms to organize such learning and educational activities and to adapt best pedagogical decisions to the needs of each learner. In this work, we addressed major challenges of adaptive and personalized learning that have been neglected in prior studies. To this aim, we leverage effective Supervised Machine Learning (ML) techniques to adaptively schedule assignments and educational activities based on the students’ needs, preferences, and background. The proposed intelligent system is trained based upon different academic factors from student learners’ characteristics such as proficiency level, interest level, remote/in-person preference, and assignment type preference and prescribes a proper learning plan to maximize the students’ overall grade and satisfaction rate at the end of course. In addition, we conduct analysis of demographic parameters such as gender and race and their effects on students’ success and academic satisfaction. For a comprehensive analysis, five different ML models including Logistic Regression (LR), K-Nearest Neighbours (KNN), Support Vector Machine (SVM), Decision Tree (DT), and Random Forest (RF) are examined. The experimental results demonstrate the superior effectiveness of the Random Forest classifier in comparison to other ML algorithms. The proposed intelligent system based on RF model achieves a 94% F1-score and accuracy rates, enabling accurate assignment of the most effective learning mode out of the four available options for learners.

An optimized random forest for multi class classification to classify the students using machine learning approaches

Comparing Different Oversampling Methods in Predicting Multi-Class Educational Datasets Using Machine Learning Techniques

iBRF: Improved Balanced Random Forest Classifier

Optimizing Neural Networks for Academic Performance Classification Using Feature Selection and Resampling Approach

Advancing Personalized and Adaptive Learning Experience in Education with Artificial Intelligence

Class Weights Random Forest Algorithm for Processing Class Imbalanced Medical Data

Using random forest for reliable classification and cost-sensitive learning for medical diagnosis

An Effective Chronic Disease Prediction using Multi-Objective Firefly Optimisation Random Forest Algorithm

Multi-split Optimized Bagging Ensemble Model Selection for Multi-class Educational Data Mining

Optimizing the feature selection methods using a novel approach inspired by the TLBO algorithm for student performance prediction

A novel approach to mitigate academic underachievement in higher education: Feature selection, classifier performance, and interpretability in predicting student performance

Classification and prediction of student performance data using various machine learning algorithms

Optimizing the Isolation Forest Algorithm for Identifying Abnormal Behaviors of Students in Education Management Big Data

Hybrid Multi-Label Classification Model for Medical Applications Based on Adaptive Synthetic Data and Ensemble Learning

An Adaptive Firefly Optimization (AFO) with Multi-Kernel SVM (MKSVM) Classification for Big Data Dimensionality Reduction

Random Oversampling-Based Diabetes Classification via Machine Learning Algorithms

CURE-SMOTE algorithm and hybrid algorithm for feature selection and parameter optimization based on random forests

A Comparative Analysis of classification data mining techniques : Deriving key factors useful for predicting students performance

Extending Machine Learning to Predict Unbalanced Physics Course Outcomes

An evaluation of nature-inspired optimization algorithms and machine learning classifiers for electricity fraud prediction

Towards the effectiveness of Deep Convolutional Neural Network based Fast Random Forest Classifier