Determining the Factors Influencing the Academic Accomplishment of Students and Predicting their Success Using Machine Learning Techniques

Educational data mining provides a suitable approach for finding the connections hidden in educational data and guessing students' academic achievement. Using the marks obtained by the student in school final exams and the cumulative grade point obtained in the undergraduate course, this research work recommends a novel model built on machine learning techniques to forecast the possibility of getting a job through campus placement. Student placement can be estimated by analyzing the performances of several machine learning approach namely Random Forest (RF), Support Vector Machines (SVM), Naive Bayes (NB), Neural Network (NN), and K-nearest Neighbour (KNN) were evaluated and assessed. The dataset included the academic performance grades of 1105 undergraduate final-year students. The findings indicate that the suggested models have a classification accuracy of between 75 and 94 percent. Predictions were established using student information like gender, Secondary School Leaving Certificate (SSLC) and Higher Secondary School Certificate (HSC) percentage of marks, medium of instruction, board of study in school, undergraduate admission quota and type, undergraduate course specialization, Cumulative Grade Point Average (CGPA), and history of arrears. A data-driven investigation is the key to obtaining a model for learning analytics which can be applied in higher education. Such models can influence decision-making in such environments. Lastly, this work identifies the most effective machine learning techniques which aid the early prediction of learners who are at higher threat of being unsuccessful in getting campus placements.