Analysis and Prediction of Student Performance by Using A Hybrid Optimized BFO-ALO Based Approach

Abstract

Data mining offers effective solutions for a variety of industries, including education. Research in the subject of education is expanding rapidly because of thebigquantityof student data that can be utilized to uncover valuable learning behavior patterns. This research presents a method for forecasting the academic presentation of students in Portuguese as well as math subjects, and it is describing with the help of 33 attributes. Forecasting the educationalattainment of students is the most popular field of study in the modern period. Previous research has employed a variety of categorization algorithms to forecast student performance. Educational data mining is a topic that needs a lot of research to improve the precision of the classification technique and predict how well students will do in school. In this study, we made a method to predict how well a student will do that uses a mix of optimization techniques. BFO and ALO-based popular optimization techniques were applied to the data set. Python was used to process all the files and conduct a performance comparison analysis. In this study, we compared our model's performance with various existing baseline models and examined the accuracy with which the hybrid algorithm predicted the student data set. To verify the expected classification accuracy, a calculation was performed. The experiment's findings indicate that the BFO-ALO Based hybrid model, which, out of all the methods, with a 94.5 percent success rate, is the preferred choice.