Abstract
Financial inclusion aims to provide financial services at an affordable cost to low-income groups in need. However, the lack of effective credit evaluation information for such groups has hindered the innovative development of financial inclusion in the banking industry. This study proposes a slack constrained matrix factorisation model to supplement missing credit information. The method fills in missing data with known data in groups of similar credit behaviours. We empirically analyse the performance of this method in a supplemented credit information matrix and a sparse credit information matrix. We use actual credit data of farmers and herdsmen in extremely poor areas and small, medium and micro enterprises in National Equities Exchange and Quotations, China. This study concludes that the proposed credit evaluation methods based on sparse credit information can effectively improve the performance of traditional credit classification algorithms.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call