| Summary: | Employees leaving an organization has been a hassle problem for every organization. Several factors contribute to the left or churn of an employee, including receiving a better offer, dissatisfaction with the salary and working environment, and other variety of reasons. This research creates an employee future prediction model to predict the leave or stay of an employee based on features like education, city, joining year, age, gender, ever benched, payment tier, and experience in the current domain. The prediction model considers four different machine learning algorithms: Decision Forest (DF), Linear Regression (LR), Neural Network (NN), and Boosted Decision Tree (BDT). The prediction is conducted based on the regression approach, and the experiment includes five tests based on the data split of 5-fold for training and testing the model. The evaluation metrics used are Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), Mean Squared Error (MSE), and Coefficient of Determination or R-squared. The experiment results show that the BDT outperformed the other algorithms. The best average R-Squared scores are 0.7918 for the BDT algorithm and 0.7597 for the DF algorisms. The outcome of this work is hoped to bring useful insights to organizations and human resource analysts to plan and improve employee retention programs, reducing an organization's losses. © 2024 AIP Publishing LLC.
|
|---|