EXPLORING LEARNING BEHAVIOUR TO PREDICT & IMPROVE INFORMATION LITERACY FOR STUDENTS
Keywords:
Machine learning, information literacy, learning behavior characteristics, learning effect, innovative talents.Abstract
This project delves into the crucial role of information literacy in college students' learning behaviors and
outcomes. Through an examination of diverse learning behaviors, with a focus on information literacy, predictive
models were developed using various supervised classification algorithms. Building upon the base paper's successful
utilization of Decision Trees, KNN, Naive Bayes, Neural Networks, and Random Forest, which achieved an
impressive 92.50% accuracy, this study extends the analysis by incorporating additional techniques. By integrating
XGBoost and a Voting Classifier into the ensemble method, the accuracy soared to a perfect 100%. This
enhancement signifies the potential for advanced methodologies to refine the predictive capabilities of models,
offering valuable insights into tailored interventions for optimizing information literacy education. The findings
underscore the significance of understanding and leveraging diverse learning behaviors to cultivate innovative
individuals equipped for lifelong learning and adaptation to evolving social needs. This research contributes to the
ongoing discourse on information literacy's pivotal role in higher education and its implications for fostering
adaptable, self-directed learners.
