Machine Learning Approach for Personality Prediction from Resume using XGBoost Classifier and Comparing with Novel Random Forest Algorithm to Improve Accuracy

Abstract

The aim of this research is to analyze personality prediction from resumes using XG-boost algorithm and comparison with the novel Random forest model to increase the accuracy value in predicting the personality using resumes. Materials and Methods: For this analysis 80 samples were collected in two groups of 40 samples each.novel Random forest is used in group 1 while the XG-boost classifier algorithm is used in group 2, The dataset was imported using the Kaggle tool in this study's workflow, and Jupiter notebook was used to train the dataset using the Novel Random Forest algorithm. Using an online statistical tool with a pretest power of 95% and an alpha value of 0.039, the sample size is determined from the results of the previous studies. Result:From simulation results, the novel Random Forest algorithm has an accuracy of 90% and XG-Boost has an accuracy of 86% with a significance level of 0.846 (p>0.05) which shows that the hypothesis is insignificant and is carried out using an independent sample T-test. Conclusion: Random Forest and xgboost are also excellent machine learning algorithms that can be used to predict personality for the given dataset but novel random forest algorithm has more accuracy value when compared with xgboost algorithm.