Back to results
US 2026/0289435 A1Published application
System
AssigneeSoftBank Group Corp.
InventorMiho AYABE
Life of this patent
Estimating the term…
Abstract
A system includes a processor that is configured to receive input information from a user, analyze the received input information and generate, based on a result of the analysis, a prompt for instructing generation of a travel plan, and input the generated prompt to a generative artificial intelligence model to cause the generative artificial intelligence model to generate the travel plan.
Claims20
A system comprising:
- circuitry configured to:
- receive, from a client terminal via a packet-switched network, input information comprising a plurality of constraint parameters specified by a user;
- extract attributes from the input information comprising a category type, resource constraint information, physical capacity information, priority items, temporal range information, and target region information, and convert the extracted attributes into standardized structured data;
- generate a prompt sentence based on the standardized structured data, the prompt sentence specifying an output format for a generative neural network model, the output format comprising a sequential action plan divided by temporal segments with associated activity descriptions and estimated resource allocations;
- transmit the prompt sentence to the generative neural network model via the packet-switched network and obtain plan data from the generative neural network model;
- perform verification on the plan data by determining whether at least a number of temporal segments, a division of time periods, and a total resource allocation are within constraint ranges derived from the standardized structured data;
- and
- convert the verified plan data into display data formatted for presentation on the client terminal, and transmit the display data to the client terminal via the packet-switched network.
5 more elements
A system comprising:
- circuitry configured to:
- receive input information comprising constraint parameters from a client terminal via a packet-switched network;
- convert the input information into standardized structured data;
- generate a prompt sentence based on the standardized structured data specifying an output format for a generative neural network model;
- transmit the prompt sentence to the generative neural network model and obtain plan data;
- verify the plan data against constraint ranges derived from the standardized structured data;
- when verification detects inconsistency, generate a re-prompt sentence and retransmit to the generative neural network model;
- estimate an emotional state of a user and adjust a presentation of the plan data based on the emotional state;
- and
- transmit display data formatted from the verified plan data to the client terminal via the packet-switched network.
7 more elements
A method performed by circuitry, the method comprising:
- receiving input information comprising constraint parameters from a client terminal via a packet-switched network;
- converting the input information into standardized structured data;
- generating a prompt sentence based on the standardized structured data specifying an output format for a generative neural network model;
- transmitting the prompt sentence to the generative neural network model and obtaining plan data;
- verifying the plan data against constraint ranges derived from the standardized structured data;
- converting the verified plan data into display data;
- and
- transmitting the display data to the client terminal via the packet-switched network.
5 more elements
Dependencies are inferred from the claim text.