A recommender system for investing in early-stage enterprises
Johannes Luef, Christian Ohrfandl, Dimitris Sacharidis, Hannes Werthner · 2020
Selecting the appropriate enterprise to invest in can become a difficult task for prospective investors, particularly in the case of startups where limited information is available. In this work, we design, implement, and evaluate a system that makes recommendations for investing in early-stage enterprises. We first perform a qualitative and quantitative study involving prominent investors to explore their decision-making process and set out the requirements for a recommender system. Then, we consider various recommendation approaches that best convey the results of our requirements analysis. To evaluate the different methods, we simulate an offline experiment, placing real investors in a hypothetical investment decision scenario.