Abstract: Systems and methods are provided for facilitating and managing unsecured loan products that utilize borrower's existing credit accounts as a modified collateral. In some embodiments, a lender can provide lending products by using customer's existing credit account as a modified collateral. A first machine learning model trained on data sets not presently utilized by financial industry to determine interest rates without increasing the lender's financial risk. A second machine learning model may be used to optimize the distribution of the pre-authorization hold amounts between revolving credit accounts using a second machine learning model that would determine weighted decision that combines the categorization and the confidence score based on the output calculated by the first machine learning model.

Abstract: Systems and methods are provided for facilitating and managing unsecured loan products that utilize borrower's existing credit accounts as a modified collateral. In some embodiments, a lender can provide lending products by using customer's existing credit account as a collateral. In some embodiments, a lender may place preauthorization hold on one of customer's existing credit accounts.