Neurocognitive Learning Design and Multiclass Algorithms: Academic and Business User Experience in Web 3.0

Abstract

The study aims to display the application of multiclass algorithms in machine learning for educational settings, particularly for skill assessment, learning style prediction, and peer recommendation. The method of skills assessment uses the RandomForestClassifier model to forecast learners' skill proficiency by using a mock dataset that represents the learners having different skill levels. The module of learning style prediction employs the model to categorize all students into their cognitive preference spectrum according to the dataset which contains their neurocognitive attributes and learning preferences. Through Peer recommendation, the system suggests to the study partners that students typically will be matched together with similar standards, interests, and learning styles. The study stresses the utility of personalized learning freed by multi-class algorithms for students to be assessed on basic skills, their learning patterns to be predicted as well as promotion of cooperative learning which eventually results in improvement of the learning experience and academic success.