Abstract: Data analysis and instrument upgrading technologies are described. In some examples, a system tracks usage data for an instrument (controlling access to an account) over time. The system processing the usage data using a trained machine learning model to generate a predictive simulation that identifies a predicted result of an upgrade of the instrument. The system dynamically updates the predictive simulation as the usage data continues to be received. The system detects that the update to the predictive simulation caused a change in the predicted result, and automatically upgrades the instrument from the first category of instrument to the second category of instrument based on the change in the predicted result. The system tracks further usage data for the instrument after the upgrade, and further trains the machine learning model based on a comparison between the further usage data and the predicted result to improve accuracy of further predictive simulations.

Abstract: Systems and methods for automatic and dynamic pre-qualification determinations associated with secured and unsecured payment instruments are provided. The systems and methods are used to automatically generate and provide pre-qualification determinations for different payment instruments based on various factors. The systems and methods further provide these pre-qualification determinations along with any applicable terms and conditions corresponding to the payment instruments a user is pre-qualified for.

Abstract: Systems and methods for resource disbursements associated with secured and unsecured payment instruments. In response to either a request to graduate or cancel a secured payment instrument, a deposit disbursement system can automatically disburse an amount corresponding to a deposit provided for issuance of the secured payment instrument. The disbursement of this amount can be performed automatically according to selection of one or more disbursement options from a set of disbursement options that are automatically, and dynamically, provided using a machine learning algorithm according to various factors corresponding to an account associated with the secured payment instrument.

Abstract: Systems and methods for graduating a secured dual-feature payment instrument to an unsecured dual-feature payment instrument are provided. A secured dual-feature payment instrument may be secured by a security deposit and associated with an account and a computational model with the account. The computational model may then be updated by continually applying a machine learning model to transactions of the account. The updated computational model may then be used to determine whether to offer graduation of the secured dual-feature payment instrument to an unsecured dual-feature payment instrument. If it is determined to offer graduation, the offer may be communicated and, if the offer is accepted, the secured dual-feature payment instrument may be graduated to an unsecured dual-feature payment instrument.