Computer-implemented method for data processing in a filtered spend program

A computer-implemented method, comprising:

• receiving, by one or more processors, a request from a first sub-system, wherein the request includes a plurality of data associated with an access device;
• determining, by the one or more processors, the first sub-system, the access device, or a combination thereof has authorization to access a service;
• determining, by the one or more processors, contextual data of one or more purses associated with the access device, status of the first sub-system, or a combination thereof, wherein the contextual data includes an initial expiration date of the one or more purses, one or more rules associated with the one or more purses, or a combination thereof, wherein each of the one or more purses includes a time-limited store value balance maintained for the access device, wherein determining the contextual data of the one or more purses, further comprises:
• monitoring, by a first machine learning model, the one or more purses to detect amounts credited to the one or more purses near or after the initial expiration date;
• and
• automatically setting, by the first machine learning model, without user intervention, an alternate expiration date to access the credited amount, wherein the alternate expiration date is based on a pre-determined duration, and the alternate expiration date is after the initial expiration date;
• transmitting, by the one or more processors, the request to a second sub-system based, at least in part, on the contextual data, the status of the first sub-system, or a combination thereof for categorizing one or more items associated with the request into one or more categories;
• and
• executing, by the one or more processors, the request based on the one or more categories.

A non-transitory computer readable medium, the non-transitory computer readable medium storing instructions which, when executed by one or more processors, cause the one or more processors to perform operations, comprising:

• receiving a request from a first sub-system, wherein the request includes a plurality of data associated with an access device;
• determining the first sub-system, the access device, or a combination thereof has authorization to access a service;
• determining contextual data of one or more purses associated with the access device, status of the first sub-system, or a combination thereof, wherein the contextual data includes an initial expiration date of the one or more purses, one or more rules associated with the one or more purses, or a combination thereof, wherein each of the one or more purses includes a time-limited store value balance maintained for the access device, wherein determining the contextual data of the one or more purses, further comprises:
• monitoring, by a first machine learning model, the one or more purses to detect amounts credited to the one or more purses near or after the initial expiration date;
• and
• automatically setting, by the first machine learning model, without user intervention, an alternate expiration date to access the credited amount, wherein the alternate expiration date is based on a pre-determined duration, and the alternate expiration date is after the initial expiration date;
• transmitting the request to a second sub-system based, at least in part, on the contextual data, the status of the first sub-system, or a combination thereof for categorizing one or more items associated with the request into one or more categories;
• and
• executing the request based on the one or more categories.