Predicting Career Transitions Through Insights from Academic Background and Workforce Dynamics Using Machine Learning

The increased dynamism in modern employment landscapes has emanated to the frequent career changes and, therefore, the importance of accustoming to the factors explaining occupational change is growing. The paper discusses the various machine learning models in the prediction of career changes with the help of the data set that contains record number of 38444 and has 22 features including the academic background of the person, job satisfaction, skills gap, and the growth of the industry. Therefore, the investigations conducted regarding the connections between these features and career mobility offer practical recommendations for various applications of HR analytics, career guidance, and workforce planning. Logistic regression, decision tree and random forests algorithms were used from the family of machine learning algorithms. Of the models presented, the random forest model was found to be most effective with an accuracy of 85%. In the feature correlation analysis we determined other factors that deem to influence this factors for example field of study, levels of job satisfaction and growth in the industry. Demographic factors are also identified as critical drivers of career choices in the study. The research is useful to identify opportunities for the support of predictive analytics in context to current issues pertaining to talent management, talent development, and the overall talent retention. This research extends prior literature in the field of HR analytics by providing a quantitative approach for analyzing career mobility.