WEIGHED QUANTUM PARTICLE SWARM OPTIMIZATION TECHNIQUE TO MEASURE THE STUDENT PERFORMANCE
Forecasting each student's unique achievement could provide useful information about which students are most likely to fail or drop out, as well as which qualities hunt the student's educational career. Data mining (DM) gives the tools required to deal with this educational data in the hunt for knowledge and patterns. Inability to forecast students' marks, this research employs an educational dataset retrieved from the UCI Machine Learning (ML) repository connected to students' grades in India as well as offers methods based on absolute learning machine networks, ensemble learning, and particle swarm optimization. As a result, a storage facility strategy is required, in which we might efficiently organize information that will then be processed in such a manner that we'll have perfect knowledge and which various other operations like mining and artificial intelligence can be performed. Furthermore, two generated data sets were employed to check the reliability of the results produced using the suggested regression models. Hypothesis tests were used to verify the accuracy of the results after collecting the error value for each suggested method. The results show that the method that integrates ensemble techniques, Weighed Quantum Particle Swarm Optimization (WQPSO) performs better.
Data mining; Machine Learning; PSO; Regression Models; Education field