A Machine Learning Approach for Future Career Planning

In this paper, we work on the modeling of peoples career paths. We first collect a large number of people’s profile and extract features from the descriptive information. Hand rules and clustering algorithm has been applied to help avoid the negative effect of natural language. We model people’s career developments with Markov Chain, and present our approach to estimate the transition probability matrix. Finally, we solve the problem that given a person’s current career path and his/her goal, what is the best best career development recommendation for him/her. As a conclusion, we will analyze the results and discuss possible improvements of our model.