System

A system comprising:

• circuitry configured to:
• receive, from a terminal device via a packet-switched network, input data expressed in natural language including a query, and analyze the input data by using a natural language processing function including tokenization, part-of-speech tagging, and semantic analysis to extract an intention and domain-related elements with associated confidence scores;
• access a storage device storing structured attribute information associated with a user identifier, and acquire the attribute information;
• generate a prompt sentence by combining the extracted intention, the domain-related elements, the attribute information, and constraint conditions, and configure the prompt sentence for input to a generative neural network model comprising a transformer architecture with self-attention layers;
• transmit the prompt sentence to the generative neural network model to generate answer data, add supplemental text to the answer data, and format the answer data into output data;
• refer to stored rule information representing domain knowledge to adjust contents included in the prompt sentence;
• and
• transmit the formatted output data to the terminal device via the packet-switched network.

A method comprising:

• receiving, by circuitry from a terminal device via a packet-switched network, input data expressed in natural language and including a query from a user, and analyzing the input data by using a natural language processing function to extract an intention of the user and domain-related elements;
• accessing a storage device that stores structured attribute information associated with an identifier of the user, and acquiring, based on the identifier, the attribute information;
• generating a prompt sentence by combining the extracted intention and domain-related elements, the acquired attribute information, and constraint conditions regarding at least one of an answer format, a level of explanation, or a description of limitations, and preparing the prompt sentence as input data for a generative neural network model;
• transmitting the prompt sentence to the generative neural network model, causing the generative neural network model to perform inference processing based on the prompt sentence to generate answer data in a text format, and acquiring the generated answer data as response data;
• adding, to the acquired answer data, predetermined supplemental text and performing formatting of the answer data into output data suitable for presentation to the user;
• and
• transmitting the formatted output data to the terminal device via the packet-switched network.