I am exploring the employee dataset to solve following research questions:
- Why employee are prone to leave?
- What type of employee are prone to leave the company?
- Predict the future employee who would tend to leave the company.
Histogram of Current Employee
Histogram of ex-employee
A few observations can be made based on the information and histograms for numerical features. Many histograms are tail-heavy; indeed several distributions are right-skewed (e.g.time_spend_company, promotion_last_5years). Data transformation methods may be required to approach a normal distribution prior to fitting a model to the data.
I use a range of baseline algorithms before we move on to more sophisticated solutions. The algorithms considered in this section are: Logistic Regression, Random Forest, and, SVM.
Since this is my real project during my internship, I don't mention the exact result here, but I outlined some inportant features to predict the employee turnover. Based on ROC graph, the logistic regression showed higher AUC score compared to random forest.
As a concluded remark, as the company generates more data on its employees (on New Joiners and recent Leavers) the algorithm can be re-trained using the additional data and theoritically generate more accurate predictions to identify high-risk employees of leaving based on the probabilistic label assigned to each feature variable (i.e. employee) by the algorithm.
Employees can be assigning a "Risk Category" based on the predicted label such that:
- Low-risk for employees with label < 0.6
- Medium-risk for employees with label between 0.6 and 0.8
- High-risk for employees with label > 0.8
A strategic "Retention Plan" should be drawn for each Risk Category group. The face-to-face meetings between a HR representative and employees can be initiated for medium- and high-risk employees to discuss work conditions. Also, a meeting with those employee's Line Manager would allow to discuss the work environment within the team and whether steps can be taken to improve it.