Career Path Prediction Using Machine Learning Classification Techniques

In today's era, choosing the right career option is a challenging task [5]. Starting at the early stage of life students usually fail to grasp the idea of which career to pursue as they lack maturity and the experience related to that field. Furthermore, students suffer greatly in deciding which career would result the highest benefit. Students do not have sufficient knowledge to take the decision on their own which may lead to complications in future. In order to avoid future complications students should make a proper decision in selecting a highest benefit career for them. Selecting a wrong career which is not meant for them will end up with work in which they are not interested or they do not have that much knowledge in that field. As students lack in decision making, they reach fortune tellers hoping that they will guide them on the right path for a bright future [3]. Instead of relying on fortune tellers to make the best prediction for the future. By considering all these things in this work we will scientifically and systematically study the feasibility of career path prediction from the survey data. This model will recommend students a career choice according to their abilities and qualities with respect to their field. If students end up having good abilities and qualities in their respective field, they can select that field otherwise they have to drop that field and choose another one. This paper presents a career path prediction using machine learning which will help students to select the appropriate career for their bright future. As career recommendations are a unique approach, we feel it should be an interactive platform. So, while building the application we presented an interactive framework which will allow students interactively perform the task and get results. The present work has 15 different types of career options. Experiments have been done using machine learning supervised classification techniques like Logistic Regression, Decision Tree, KNN, Naïve Bayes, SVM, Random Forest, Stochastic Gradient Descent, AdaBoost, XgBoost, and some hybrid algorithms using stacking like SvmAda, RfAda and KnnSgd.