Track: Entrepreneurship

Abstract

In recent years there has been a tendency to incorporate entrepreneurship related contents in engineering academic programs to help students identify opportunities, create, and innovate. It is interesting to analyze the interrelation between factors such as subject, working experience, university infrastructure, and social background in the entrepreneurial interest. This research compare models based on artificial neural networks (ANN) to predict the interest in entrepreneurial activities among engineering students. The research is carried out following a classic 4-stage methodology (analysis, design, development, and validation). During the analysis, the local results of the University Entrepreneurial Spirit Students’ Survey 2018 (GUESSS) are preprocessed. In the design, GUESSS questionnaire is divided to create different predictive models. Both phases, construction and validation, are carried out entirely using the software WEKA. The complete dataset is split up to create 2 subsets, one is used for training and test (80%) and the other for validation (20%). The results reveal that the models can predict correctly between 34% and 78% of the validation dataset’s cases. In conclusion, the investigation’s results show that some of the proposed models based on ANNs can help predict the interest in entrepreneurial activities among engineering students with a reasonable degree of certainty.In recent years there has been a tendency to incorporate entrepreneurship related contents in engineering academic programs to help students identify opportunities, create, and innovate. It is interesting to analyze the interrelation between factors such as subject, working experience, university infrastructure, and social background in the entrepreneurial interest. This research compare models based on artificial neural networks (ANN) to predict the interest in entrepreneurial activities among engineering students. The research is carried out following a classic 4-stage methodology (analysis, design, development, and validation). During the analysis, the local results of the University Entrepreneurial Spirit Students’ Survey 2018 (GUESSS) are preprocessed. In the design, GUESSS questionnaire is divided to create different predictive models. Both phases, construction and validation, are carried out entirely using the software WEKA. The complete dataset is split up to create 2 subsets, one is used for training and test (80%) and the other for validation (20%). The results reveal that the models can predict correctly between 34% and 78% of the validation dataset’s cases. In conclusion, the investigation’s results show that some of the proposed models based on ANNs can help predict the interest in entrepreneurial activities among engineering students with a reasonable degree of certainty.