Employee-Future-Prediction

The ability to solve real life situation by creating highly efficient and accurate model is the beauty of every machine learning project, the paper is a report of a research to create a model that predict whether an employee would leave work or not within two years, this is achieved by modelling past dataset gotten from Kaggle, the dataset was tuned to get the best result, major classification algorithms were performed, after which the dataset was oversampled to cater for biasness and the algorithms were performed again and compared, although all models performed well generally, the decision tree classifier and the adaptive boosting classifier performed best and showed to be the most efficient and accurate. The most important feature of a model is unbiasedness and this was majorly tackled in this research.