System

A system comprising:

• circuitry configured to:
• acquire position data from a positioning device via a communication interface coupled to a packet-switched network, and execute a spatial query against a storage device to identify location-associated records within a computed region based on the position data;
• generate structured information for each identified record comprising a name, a type, an outline, coordinate data, and distance data computed from the position data;
• retrieve user attribute data and combine the user attribute data with the structured information, priority data derived from the distance data, and display style instruction data to construct a prompt data structure;
• provide the prompt data structure to a generative neural network model and obtain a natural language description;
• format the natural language description and the structured information into guidance data, transmit the guidance data to a terminal device, and record a generation history and a transmission history in the storage device;
• determine, based on a comparison of updated position data against the generation history, whether a regeneration condition is satisfied, and upon satisfaction, selectively regenerate the prompt data structure and the natural language description while reusing cached guidance data for unchanged records;
• and
• acquire operation state data from a detection device, determine control conditions specifying timing and verbosity of information presentation based on the operation state data, and apply the control conditions to at least one of the prompt data structure and the guidance data.

A method comprising:

• acquiring, by circuitry via a communication interface coupled to a packet-switched network, position data from a positioning device associated with a mobile apparatus;
• executing, by the circuitry, a spatial query against a storage device to extract location-associated records within a computed surrounding region based on the position data, and generating structured information comprising at least a name field, a type field, and an outline field for each extracted record together with distance information;
• constructing, by the circuitry, a prompt data structure by combining the structured information, user attribute data, and priority information computed from the distance information, and providing the prompt data structure as input to a generative neural network model;
• obtaining, from the generative neural network model, a natural language description responsive to the prompt data structure;
• formatting guidance data comprising the natural language description and the structured information into a transmission payload and transmitting the transmission payload to a terminal device;
• and
• recording a generation history and a transmission history associated with the guidance data, and selectively updating the prompt data structure and the natural language description in response to updated position data by referencing the generation history to suppress redundant invocation of the generative neural network model.