A Comparative Analysis of Machine Learning Models on Firm Performance Predictions Using Human Resources Management Practices

Human resource management practices play a crucial role in the effects of firm performance. In this project, eight machine learning algorithms have been used to find the best-performing model that gives the best prediction of the practices and provides the highest firm performance. The followings models have produced the given accuracies: Decision Tree with 90.40%, K-Nearest Neighbors with 90.40%, GaussianNB with 91.20%, Logistic Regression with 92.80%, AdaBoost Classifier with 93.60%, XGB Classifier with 93.60%, Support Vector Classifier with 93.60%, and Random Forest with 95.20%. The random forest classifier provides the best result among all the eight models. The project's accuracy indicates a promising capability to distinguish the data, thus accomplishing the objective of finding the best machine learning model to predict the firm performance. These findings show that in order to meet user expectations and increase the system's practical applicability in real-world situations.