Placement Analysis for Students using Machine Learning

Abstract

Every University/College want their students to get placed in a good company with a better package. They create the syllabus so, that students will gain the most knowledge out of the study period. But they are not sure whether the students are getting trained with proper guidance and instruction to be placed. There is a need for metric to find the progress of the student’s placement. So, the student can speed up his/her preparation to meet the demands of the minimum criteria of placement. This metric is called the placement analysis system, it takes attributes like internships completed, papers published, aptitude scores, etc to predict where the individual will get placed i.e., Dream company, Core company, Normal company or not get placed. For this the machine learning algorithms are used to predict the results using three basic algorithms Random Forest, Decision tree and K-means clustering the accuracy of the algorithms are determined to find the optimal algorithm. The past data on the placement results were fed as the training dataset and a part of it is used for testing the accuracy of the model. Then if the accuracy is good, this can be used to predict the possibility of a student getting placed. If the student is unhappy with the result, then the model can be used to find the area where the student needs to improve to get to his desired goal. If properly implemented and the students work consistently, this aids in providing solutions to meet every student's goal.