Back to results
US 2026/0288761 A1Published application
System
AssigneeSoftBank Group Corp.
InventorYu SUNAGAWA
Life of this patent
Estimating the term…
Abstract
A system includes a processor that is configured to collect information regarding a user's income, fixed costs, and family composition, and calculate an optimal food budget based on the collected information, input a prompt sentence into a generative artificial intelligence model based on the calculated food budget to generate a home-cooked menu that is healthy and nutritionally balanced, and propose products or services based on at least one of a purchase history of the user, preferences of the user, and an emotional state of the user.
Claims20
A system comprising:
- circuitry configured to:
- acquire, via a communication interface coupled to a packet-switched network, first numerical parameter data and second numerical parameter data from a terminal device;
- compute a constraint boundary value by executing a numerical computation operation on the first numerical parameter data and the second numerical parameter data, the numerical computation operation comprising a subtraction operation and a proportional apportionment operation based on stored weighting coefficients;
- construct a constrained instruction sequence comprising the constraint boundary value and preference parameter data acquired from the terminal device, and input the constrained instruction sequence into a generative neural network model comprising a transformer-based architecture to generate candidate output data satisfying constraint conditions derived from the constraint boundary value;
- acquire external record data from an external information processing apparatus via the communication interface, extract category-classified records from the external record data using a structured data processing operation, compute aggregated numerical values from the category-classified records, and update the constraint boundary value based on a deviation between the aggregated numerical values and the constraint boundary value;
- reconstruct the constrained instruction sequence based on the updated constraint boundary value and input the reconstructed constrained instruction sequence into the generative neural network model to generate updated candidate output data;
- and
- transmit the updated candidate output data to the terminal device via the communication interface.
5 more elements
A system comprising:
- circuitry configured to:
- acquire first numerical parameter data and second numerical parameter data from a terminal device via a communication interface;
- compute a constraint boundary value by a numerical computation operation based on the first numerical parameter data and the second numerical parameter data;
- construct a constrained instruction sequence comprising the constraint boundary value and preference parameter data, and input the constrained instruction sequence into a generative neural network model to generate candidate output data satisfying constraint conditions;
- acquire external record data from an external information processing apparatus, compute aggregated numerical values, update the constraint boundary value based on the aggregated numerical values, and regenerate the candidate output data;
- and
- transmit the candidate output data to the terminal device.
4 more elements
A method comprising:
- acquiring, by circuitry via a communication interface coupled to a packet-switched network, first numerical parameter data and second numerical parameter data from a terminal device;
- computing a constraint boundary value by executing a numerical computation operation on the first numerical parameter data and the second numerical parameter data, the numerical computation operation comprising a subtraction operation and a proportional apportionment operation based on stored weighting coefficients;
- constructing a constrained instruction sequence comprising the constraint boundary value and preference parameter data acquired from the terminal device, and inputting the constrained instruction sequence into a generative neural network model comprising a transformer-based architecture to generate candidate output data satisfying constraint conditions derived from the constraint boundary value;
- acquiring external record data from an external information processing apparatus via the communication interface, extracting category-classified records from the external record data using a structured data processing operation, computing aggregated numerical values from the category-classified records, and updating the constraint boundary value based on a deviation between the aggregated numerical values and the constraint boundary value;
- reconstructing the constrained instruction sequence based on the updated constraint boundary value and inputting the reconstructed constrained instruction sequence into the generative neural network model to generate updated candidate output data;
- and
- transmitting the updated candidate output data to the terminal device via the communication interface.
4 more elements
Dependencies are inferred from the claim text.