Predictive Analysis Tool for Predicting Student Performance and Placement Performance using ML algorithms.

Computer Science Engineering

Recently, Predictive Analysis using machine learning and data science has had greater growth in almost every field. Predictive analysis for predicting student performance has always been difficult than other predictions such as House price prediction, stock market prediction, sales prediction, etc., Due to the difficulty in finding and collecting appropriate datasets, especially in college-level education. Since the grading system differs from universities to universities, it is even more difficult to collect proper datasets that can be used as a global dataset, created based on the system that is common to almost every university. In this paper, we have taken this difficult task and created a dataset, which contains student details of over 185 students. The dataset was collected from SRM Valliammai engineering college, it includes the data of 2018 and 2019 passed out students. Other than academic data, the dataset also covers other attributes of the students. The dataset contains around 20 attributes which would improve the accuracy score. Using this dataset our aim is to develop a training model that predicts the final CGPA of a student and also the placement result of the particular student. We have successfully predicted the student’s final CGPA and their placement result. We have predicted the results using algorithms like Linear Regression (LR1), Decision tree (DT) algorithm, K-nearest neighbor algorithm (KNN), Logistical regression (LR2) and Lasso regression (LR3) algorithm. And we have obtained excellent results with accuracy over 94%. The predicted results are obtained from comparing the accuracy score of the above-given algorithms. The one with the highest accuracy was displayed as the final output.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.