Machine Learning Models for Predicting Entrepreneurial Intentions of Higher Education Students

Angeleski, Marjan and Iliev, Dean and Chochkova, Natasha (2022) Machine Learning Models for Predicting Entrepreneurial Intentions of Higher Education Students. In: DisCo 2022: Empowering Digital and Entrepreneurial Competences through E-learning, June, 20-21, 2022, Prague, Czech Republic. (In Press)

Full text not available from this repository.


In addition to the innate characteristics, when creating an entrepreneur profile, the acquired knowledge, skills and competencies also play a major role. The student's entrepreneurial profile is formed as a result of the non-formal learning processes, formally acquired competencies and potentially influential factors such as place of residence, gender, age, scientific field to which the study program the student studies belong, etc.
The intention of this paper is to evaluate several different algorithms for classification and prediction of the entrepreneurial intentions of higher education students by applying machine learning.
The paper follows the IMRD methodology. It is structured in two parts: the theoretical part, dominated by pedagogical and didactic methodological aspects of students studying in higher education, and the second part, dedicated to the application of machine learning in predicting the entrepreneurial aspirations of students in higher education. Namely, using Python programming language, an analysis was performed and several classification algorithms were evaluated, such as: logistic regression, decision tree, random forest, support vector machine and k-nearest neighbours. Different model performance metrics have been used: Accuracy, ROC, AUC and F1-score. Also, based on the Feature importance scores, it is determined which of the initial features have the greatest impact on the models leading to an appropriate way of feature selection. The initial dataset is composed of data from a survey of students at two universities, one from the Republic of North Macedonia and the other from the Republic of Albania. By testing the sample of 795 students the general hypothesis in the research that machine learning, as part of artificial intelligence, can be used for predictions of entrepreneurial intentions of higher education students, was confirmed. This research model saves educational efforts and at the same time, provides rationalization in the processes of planning educational solutions by applying machine learning modelling.

Item Type: Conference or Workshop Item (Paper)
Subjects: Scientific Fields (Frascati) > Social Sciences > Economics and Business
Scientific Fields (Frascati) > Social Sciences > Educational sciences
Divisions: Faculty of Economics
Faculty of Education
Depositing User: Mr Dimitar Risteski
Date Deposited: 14 Sep 2023 04:57
Last Modified: 14 Sep 2023 04:57

Actions (login required)

View Item View Item