Comparison between Rough Set and Naive Bayesian in Determining the Feasibility of Potential Entrepreneurial Types

Entrepreneurial activity is a combination of creativity, innovation, and courage to face risks in forming and maintaining a business. In Indonesia, the largest entrepreneurs are in the Micro, Small, and Medium Enterprises (MSME) sector, contributing 57.8% of the Gross Domestic Product (GDP) in 2018. A country can be said to be prosperous if it has at least 2% of its total population as entrepreneurs. Indonesia is currently estimated to only have around 0.2% of entrepreneurs in all fields, this number is of course inversely proportional to Singapore which has 7% of entrepreneurs from its total population. The 2019 Global Entrepreneurship Index survey showed that Indonesia was ranked 94th out of 137 countries surveyed and in Southeast Asia, Indonesia was in the second lowest position along with the Philippines. The problem with the small interest in entrepreneurship in Indonesia is due to the failure rate which is still quite high. It is recorded that 50-60% of MSME stop their business in the first three years. The biggest factors that influence failure include capital, human resources, innovation, and the ability to adapt the business to technology. Therefore, predictions are needed to determine the type of business that is feasible and has the potential to continue or whether the business needs to be modified. From the problem above, a comparison will be carried out between Rough Set and Naive Bayesian to obtain the highest level of accuracy in making predictions. The data in this research was obtained from 108 MSME in Semarang. The research results show that Naive Bayesian has the highest level of accuracy, reaching 95%, so it is suitable to be used as a method for predicting the feasibility of potential types of entrepreneurship. It is hoped that the results of this research will reduce the failure percentage of MSME, increase economic growth, and increase employment opportunities.