ANALYSIS OF EMPLOYEE PERFORMANCE AND PREDICTION OF POTENTIAL ATTRITION- A SURVEY

Abstract

Data mining applications are becoming quite important tool in considerate and cracking problems related to education and administration in higher education and job profession. One of the common problems in a job sector or higher education is the evaluation of instructors’ performances in a course or employee’s performance in a job sector. The most widely applied method to evaluate the performance is through feedback of performance and details kept by the company or institution. In general, a classifier model can be built easily from many different methods and algorithms. The classifier system has been applied for predicting the employee performance using different machine learning techniques like SVM, random forest and decision tree system etc. The optimal performances of each of the algorithms are considered and prediction is made accordingly. Different features and attributes of employees that are essential in governing employees’ performance are extracted using feature extraction method. The datasets in the survey are found to be gathered from the HR analysis of specific company through HR datasets, questionnaires, feedback forms and the employee turnover probability is predicted. The effectiveness of these various models are calculated on the test data. After study of existing work, the outcomes and challenges are identified and conclusions are drawn accordingly