Forecasting the Effectiveness of Children's Learning Based on Data Mining Technique

Abstract

This study used data mining (DM) technologies to estimate how well toddlers would learn. Both the public and commercial sectors are rapidly expanding the number of academic institutions. Despite this, learners who are classified as medium- and comparatively low-risk still face unemployment. This work employed a novel fine-tuned seagull-optimized weighted k-nearest neighbor (FSOA-KNN) method to improve the efficacy of the children's learning. We included three hundred students in this study, collected their features, and analyzed them. We apply preprocessing techniques to the collected data samples to enhance prediction performance. We implement the suggested method and assess its effectiveness using measurements for accuracy, recall, f-measure, and precision. According to the study's findings, the suggested model can predict children's learning efficiency with accuracy (97%), precision (96.5%), recall (96%), and F-Measure (95.5%). This article also helps identify students who need additional advice or counseling from a high-quality teaching provider.