Prediction Employee Performance Using Weighted Average Ensemble Method

Employees are the most crucial resources in a firm, but they can leave their jobs unexpectedly, resulting in significant costs. Generally, attitude and effort of employee are driven by personality qualities, although job satisfaction that may be the consequence of particular observations from an organization with respect to environmental factors. Meanwhile, acquiring a new staff may be time-consuming and costly. Similarly, a newly employees hired may required for considering some efforts to make the particular organization productive. Job satisfaction is an important component in leaving the organization. Employee performance prediction and reasons to leave the organization have undertaken by Human Resource Management (HRM). This type of prediction must advance from HRM to examine the most talented and most experienced employees’ reasons for leaving the organization through different data mining techniques, but a precise prediction is not accomplished. This can be determined by observing some of the most experienced and talented employees departing their firm. As a result, this paper aimed to construct an ensemble model that can accurately predict employee performance and job satisfaction using an HR analytics dataset. The present research paper focused on examining the employee performance and job satisfaction mentioned by employees in the Employee Attrition dataset by prediction by Weighed Average (WA) as an ensemble approach with Logistic Regression (LR). Furthermore, the suggested ensemble approach performance evaluation has been contrasted to the traditional method for analyzing better employee performance and job satisfaction predictions.