Exploratory Data Analysis and Classification of Employee Retention based on Logistic Regression Model

Nowadays, a huge number of companies are using employee data records to understand the employee churn rate in the company. Companies use predictive analytics to predict the chances of an employee leaving the company. Organizations are effectively using Human Resource (HR) predictive analytics to increase revenue and turnover. This paper is about building a predictive model on the employee dataset to predict if an employee is likely to leave or continue service in the employed company. The steps carried out for visualization and Exploratory Data Analysis (EDA) has been highlighted in this paper by bringing out valuable insights from the employee dataset. Observations were made to the features that determine the chances of an employee leaving the company. For the prediction and classification of employee retention, a Logistic Regression (LR) model is applied to the dataset. The article focuses on the analysis of data and understanding the working of the LR model for predicting categorical features. The performance of the applied model was evaluated using a confusion matrix. The accuracy of the model was calculated from values in the confusion matrix.