Extracting behavioral patterns of people in crowdfunding projects using artificial intelligence and identify key investors

Crowdfunding as a new method has an important contribution to projects, especially for small and medium-sized businesses. Despite the many years of activity of crowdfunding platforms in Iran, the scale of projects and the volume of capital provided by these platforms are still far from the expected performance compared to other countries. With the aim of improving the current situation, first, the success factors of crowdfunding projects were categorized into four main groups: investors, project owner, project, and financing platform. In the next stage, it was determined that the investor group was less addressed, so the present study focused on factors related to this group. Then, using artificial intelligence tools, the information recorded in the database of one of the active platforms was analyzed and the dominant trends in the investment process, behavioral patterns, and interpersonal relationships were identified. Finally, using machine learning and the "Association Rule Learning" tool, meaningful relationships emerged between the participation of individuals and their degree of importance, and some of them were identified as key users. By discovering meaningful relationships and recurring patterns in individual participation, the results of this paper will be useful for new idea owners, project owners, crowdfunding policymakers, researchers, crowdfunding platform managers, and investors in this field.