Methods and apparatus for constructing machine learning models to process user data and provide advance access to payments

A method, comprising:

• receiving sets of raw data for a current pay period of an entity, each set of raw data being associated with a different time period of a plurality of time periods in the current pay period and including (1) time data indicating time worked by a set of users affiliated with the entity during that time period and (2) earnings data indicating earnings made by the set of users during that time period;
• in response to receiving each set of raw data:
• identifying anomalous data in that set of raw data, the anomalous data including at least one of time data errors or earnings data errors;
• processing that set of raw data by removing the anomalous data identified in that set of raw data to produce processed data without at least one of the time data errors or the earning data errors;
• determining, for the time period associated with that set of raw data, a gross earnings amount, a net earnings amount, and a net-to-gross ratio for each user from the set of users based on the processed data;
• updating an available balance of each user from the set of users based on the gross earnings amount, the net earnings amount, and the net-to-gross ratio;
• sending information indicative of the available balance for each user from the set of users to a set of user devices such that the set of user devices displays the available balance for each user from the set of users;
• receiving, from a user device from the set of user devices, a request to transfer at least a portion of the available balance;
• and
• causing, in response to receiving the request to transfer, a transfer of at least the portion of the available balance to a predetermined user account associated with the user.

An apparatus, comprising:

• a memory;
• and
• a processor operatively coupled to the memory, the processor configured to:
• receive sets of raw data for a current pay period of an entity, each set of raw data being associated with a different time period of a plurality of time periods in the current pay period and including (1) time data indicating time worked by a set of users affiliated with the entity during that time period and (2) earnings data indicating earnings made by the set of users during that time period;
• in response to receiving each set of raw data:
• identify anomalous data in that set of raw data, the anomalous data including at least one of time data errors or earnings data errors;
• process that set of raw data by removing the anomalous data identified in that set of raw data to produce processed data without at least one of the time data errors or the earning data errors based on historical net earnings data, historical gross earnings data, and historical net-to-gross ratio data;
• determine, for the time period associated with that set of raw data a net earnings amount for each user from the set of users based on the processed data;
• update an available balance of each user from the set of users based on the net earnings amount;
• and
• transfer, at periodic intervals before the end of the current pay period, payment amounts to predetermined user accounts associated with the set of users, the payment amounts based on payment amounts previously transferred for each user from the set of users and the available balance of each user from the set of users.