Machine Learning Methods for Predicting Entrepreneurial Intentions Based on Personality Traits

Hafiz Waheed ud din; Rizwan Ullah; Yahya khan

Authors

Hafiz Waheed ud din Department of Computer and software engineering, faculty of computing Gomal University
Rizwan Ullah Department of Computer and software engineering, faculty of computing Gomal University
Yahya khan Department of Computer and software engineering, faculty of computing Gomal University

Abstract

This research assesses the ability of different machine learning models in forecasting entrepreneurial intentions using selected personality constructs of university students inclusive of proactive personality, self-regulation and hope. Using a dataset of 1,500 entries, we applied advanced and traditional machine learning methods, including Linear Regression, Multilinear Regression, Gradient Boosting (optimized with hyperparameters: Other included models were Linear Regressor, Decision Tree Regressor and Random Forest Regressor with optimal values of {‘learning_rate’: 0.05, ‘max_depth’: 3, ‘n_estimators’: 500}, Neural Networks (with feature scaling both with and without regularisation), RandomForestRegressor with the hyperparameters {‘max_depth’: 20, ‘n_estimators’: 100}, Lasso Regression with the best value of The findings suggest that there is a dramatic improvement in performance and reliability for complex models such as Gradient Boosting and Neural Networks. The current study advances knowledge in the field of entrepreneurship by showing the role that machine learning can play into improving predictive analysis that relies on psychological and personality factors.

Keywords- Machine Learning Methods, Predicting Entrepreneurial, Intentions Based and Personality Traits