This experiment predicts the employed/left status of an employee based on the given features.
The idea is to use data to predict whether an employee is likely to leave. Currently, the employee retention process is very retroactive. Once an employee leaves, he or she takes an "exit interview" and shares reasons for leaving. HR then tries to learn insights from that interview and make changes around the company accordingly. The problem with this approach is that it's too haphazard. The quality of insight gained from an interview depends heavily on the skill of the interviewer. They've asked their business intelligence analysts to provide a dataset of past employees and their status (still employed or already left). My task is to build a classification model using that dataset. **Problem Specifics** Deliverable: Predictive Machine Learning API Machine learning task: Classification Target variable: Status (Employed/Left) **Data Dictionary** For this project: The dataset has 14249 observations for past/present employees. The observations span 12 different departments. Each observation includes the employee’s current employment status. We have the following features: **Target variable** 'status' – Current employment status (Employed / Left) **Administrative information** 'department' – Department employees belong(ed) to 'salary' – Salary level relative to the rest of their department 'tenure' – Number of years at the company 'recently_promoted' – Was the employee promoted in the last 3 years? **Workload information** 'n_projects' – Number of projects employee is staffed on 'avg_monthly_hrs' – Average number of hours worked per month **Mutual evaluation information** 'satisfaction' – Score for employee’s satisfaction with the company (higher is better) 'last_evaluation' – Score for most recent evaluation of employee (higher is better) 'filed_complaint' – Has the employee filed a formal complaint in the last 3 years?