Data Scientist Job Change Prediction Using Machine Learning Classification Techniques

Abstract

In this ever-expanding world, each and every industry is highly competitive. Recently, the Artificial Intelligence [AI] technologies are influencing every facet of data science domain. This technology driven-competitive world creates a lot of difficulties for people working in the data science sector by developing a notion of abandoning the data scientist profession. On the other hand, the human expectations aren’t matching reality, followed by No Clear Benchmarking in Salary Pay-Outs, Mapping a Data Scientist’s Role to Business Goals, and finally, there is a lack of upskilling for Data Science Professionals. These are the causes for the job change in the individuals and as the competition increases, there is always a substitute, henceforth the job security also falls in this competitive market. Different machine learning methods are used to figure out or to forecast the job change for the employees in the firm. This can be observed by the HR and give additional rewards for employee since the employee is an important asset to the company and its development. The synergy of deep learning approaches, machine learning and ensemble methods are utilized. This research work attempts to perform a comparison on various classification approaches such as logistic regression, random forest, KNN and support vector machine in order to demonstrate which method performs the best to offer an insight to the general community.