Ensemble Classifier-Based Features Ranking on Employee Attrition

The departure of good employee incurs direct and indirect cost impacts for an organization. arises from hiring to training the relevant employee. replacement time lost productivity affect running business processes. This work presents use ensemble classifier identify important attributes that affects attrition significantly. data consists related job function, education level, satisfaction towards working relationship, compensation, frequency travel. Both bagging boosting classifiers were used testing. results show selected features (nine features) achieve same result as full features. are age, income, years, source employment, years since last promotion, salary hike, travelling frequency. These using classifiers. Satisfaction on relationship do not appear be significant in classifier’s results.