Back to results
US 11,080,624 B2Granted
Application programming interface for a learning concierge system and method
AssigneeVisa International Service Association
InventorJerry Wald
Life of this patent
Estimating the term…
Abstract
The described system and method reviews past receipts from purchases and past purchase patterns and, in response to a request, returns a recommendation about future scheduling or future purchases.
Claims18
A computer based system for providing a schedule recommendation to a user comprising:
- a transaction processor and filter server that receives purchase data associated with purchases made by the user from a payment system via an application programming interface (API), the transaction processor and filter being configured to review and extract useful data from the purchase data for making schedule recommendations, the useful data including transaction amounts, merchant identifications, merchant category codes, merchant locations, product categories, and transaction intents;
- a relational database server that receives the useful data from the transaction processor and filter server via an API, the relational database server comprising:
- a calendar server that stores calendar data for the user;
- a merchant category code database server that stores merchant category code (mcc) data including the transaction amounts, the merchant identifications, the merchant category codes, the merchant locations, the product categories, and the transaction intents for the user;
- and
- a user preference server that stores user preference data for the user;
- and
- a concierge processor that receives the calendar data, the mcc data, and the user preference data from the relational database server via an API, the concierge processor being configured according to computer-executable instructions for
- executing a machine learning algorithm on the calendar data, the mcc data, and the user preference data to make a schedule recommendation for the user, wherein the machine learning algorithm is a clustering analysis algorithm, and wherein the clustering analysis algorithm clusters similar individuals with similar purchase patterns and uses the cluster of similar individuals in making the schedule recommendation for the user;
- and
- communicating the schedule recommendation to a personal computing device of the user via an application programming interface.
8 more elements
A computer-implemented method for providing a schedule recommendation to a user comprising:
- receiving, at a transaction processor and filter server, purchase data associated with purchases made by the user from a payment system via an application programming interface (API);
- at the transaction processor and filter server, reviewing and extracting useful data from the purchase data for making schedule recommendations, the useful data including transaction amounts, merchant identifications, merchant category codes, merchant locations, product categories, and transaction intents;
- receiving, at a relational database server, the useful data from the transaction processor and filter server via an API, calendar data, and user preference data, the relational database server including a calendar server that stores the calendar data, a merchant category code (mcc) database server that stores mcc data including the useful data, and a user preference server that stores the user preference data;
- receiving, at a concierge processor, the calendar data, the mcc data, and the user preference data from the relational database server via an API;
- at the concierge processor, executing a machine learning algorithm on the calendar data, the mcc data, and the user preference data to make a schedule recommendation for the user, wherein the machine learning algorithm is a clustering analysis algorithm, and wherein the clustering analysis algorithm clusters similar individuals with similar purchase patterns and uses the cluster of similar individuals in making the schedule recommendation for the user;
- and
- communicating the schedule recommendation to a personal computing device of the user via an application programming interface.
4 more elements
Dependencies are inferred from the claim text.