CN-121996677-A - Travel planning method and system based on multi-modal interaction

CN121996677ACN 121996677 ACN121996677 ACN 121996677ACN-121996677-A

Abstract

The application relates to the technical field of artificial intelligence and discloses a travel planning method and system based on multi-mode interaction, wherein the method comprises the steps of setting a data transmission interface for connecting a product database of a travel product platform and an information database of a social media platform, and accessing a semantic analysis model to establish corresponding standardized data query service; the method comprises the steps of defining semantic information of standardized data query service in a semantic analysis model, packaging the standardized data query service into data query MCP service, constructing an information retrieval agent, an information screening agent and a trip planning agent based on an AI large model, further creating a trip planning agent comprising the information retrieval agent, the information screening agent and the trip planning agent, butting the trip planning agent against the data query MCP service and map platform MCP service, and setting an output rendering component for the trip planning agent.

Inventors

YUAN YUANLIN
LI YONGSEN
CHEN QUAN
LIN YUNAN
LI JUNFENG
LIU QI
LI JIAQI
LI BO

Assignees

广东省城乡规划设计研究院科技集团股份有限公司

Dates

Publication Date: 20260508
Application Date: 20260122

Claims (10)

1. A travel planning method based on multi-modal interaction is characterized by comprising the following steps: S1, setting a data transmission interface for connecting a product database of a travel product platform and an information database of a social media platform, and accessing a semantic analysis model to establish a corresponding standardized data query service; S2, defining semantic information of the standardized data query service in a semantic analysis model, and packaging the standardized data query service into a data query MCP service; S3, constructing an information retrieval agent, an information screening agent and a trip planning agent based on an AI large model, and further creating a trip planning agent comprising the information retrieval agent, the information screening agent and the trip planning agent; And S4, inquiring the trip planning agent docking data to an MCP service and a map platform MCP service, and setting an output rendering component for the trip planning agent.
2. The method for planning travel based on multi-modal interactions as set forth in claim 1, wherein the step of setting up a data transmission interface connecting a product database of a travel product platform and an information database of a social media platform, and accessing a semantic parsing model to establish a corresponding standardized data query service further includes: S11, extracting travel product data from a travel agency enterprise system based on a data acquisition period, extracting key information fields from the travel product data and storing the key information fields in a product database of the travel product platform; S12, acquiring travel information data from a social media platform based on a data acquisition period and a keyword group, extracting a keyword information field and feedback trend information from the travel information data, and storing the keyword information field and feedback trend information into an information database of the social media platform; s13, setting a data transmission interface and a semantic analysis model for accessing travel planning agents to the product database and the information database, and establishing standardized data query service; the key information fields comprise index information, travel places, consumption amounts, personnel constitution, time and weather, the key word group comprises a plurality of key words used for identifying travel information data and a plurality of key words used for identifying key information fields and feedback trend information from unstructured data, and the feedback trend information is information of preference and dislike trend feedback of a social media platform user on a travel item.
3. The method for traveling planning based on multi-modal interactions of claim 1, wherein the constructing an information retrieval agent, an information screening agent, and a trip planning agent based on the AI large model to create a traveling planning agent comprising the information retrieval agent, the information screening agent, and the trip planning agent comprises: S31, constructing an information retrieval agent with a natural language intelligent question-answer function based on an AI large model to extract first constraint information from dialogue contents of a user, call a data query MCP service, and retrieve and structuralized process first associated data; S32, constructing an information screening agent based on the multi-mode visual large model to perform content understanding on the first associated data, screening content patches meeting preset quality standards, and generating a labeling text; S33, constructing a trip planning agent based on the AI large model, extracting second constraint information from dialogue contents of the user to perform product screening and trip planning, and generating a trip planning form; The first constraint information comprises a travel place, the first association data refers to a plurality of key information fields and feedback trend information which are retrieved from a product database and an information database and accord with the first constraint information, and the original content paste is obtained again based on index information, the second constraint information comprises a rigid constraint and a flexible constraint, the rigid constraint comprises budget, number of people, time and weather, and the flexible constraint comprises travel preference.
4. The method for planning travel based on multi-modal interactions as set forth in claim 3, wherein the creating a travel plan agent comprising the information retrieval agent, the information screening agent, and the travel plan agent includes: s34, creating travel planning agents based on the AI large model and connecting each professional agent; S35, the trip planning agent receives dialogue information of the user, so as to carry out multi-round question and answer with the user through natural language, and circularly drive subsequent information retrieval and trip planning tasks based on thinking-action-observation; S36, generating subtasks by the trip planning agents based on the identified demand information of the users, forwarding the subtasks to corresponding professional agents, collecting and re-integrating the output content of each professional agent, and generating a final decision report; the professional agents include information retrieval agents, information screening agents and trip planning agents.
5. The method for planning travel based on multi-modal interactions as set forth in claim 1, wherein the interfacing the travel planning agent with a data query MCP service and a map platform MCP service, setting an output rendering component for the travel planning agent, comprises: S41, enabling the trip planning agent to be in butt joint with a map platform MCP service, wherein the map platform MCP service integrates a journey planning tool, a path planning tool and a weather inquiry tool; s42, the front end of the rendering component dynamically analyzes the content output of the travel planning intelligent body, extracts a multimedia link and an HTML structure in the output content, and pertinently defines different style rendering templates.
6. Travel planning system based on multimode interaction, characterized by comprising: The data query service establishment module is used for setting a data transmission interface for connecting a product database of the travel product platform and an information database of the social media platform, and accessing a semantic analysis model to establish a corresponding standardized data query service; The data query service packaging module is used for defining semantic information of the standardized data query service in a semantic analysis model and packaging the standardized data query service into a data query MCP service; The travel planning agent creation module is used for constructing an information retrieval agent, an information screening agent and a travel planning agent based on the AI large model, and further creating a travel planning agent comprising the information retrieval agent, the information screening agent and the travel planning agent; and the output rendering setting module is used for inquiring the trip planning agent docking data with the MCP service and setting the trip planning agent with the map platform MCP service and setting an output rendering component for the trip planning agent.
7. The multi-modal interaction based travel planning system of claim 6 wherein the data query service creation module comprises: The travel product data extraction sub-module is used for extracting travel product data from a travel agency enterprise system based on a data acquisition period, extracting key information fields from the travel product data and storing the key information fields in a product database of the travel product platform; The travel information data extraction sub-module is used for acquiring travel information data from the social media platform based on a data acquisition period and a keyword group, extracting a keyword information field and feedback trend information from the travel information data and storing the keyword information field and feedback trend information into an information database of the social media platform; The data query service sub-module is used for setting a data transmission interface and a semantic analysis model for accessing the trip planning agent to the product database and the information database and establishing standardized data query service; the key information fields comprise index information, travel places, consumption amounts, personnel constitution, time and weather, the key word group comprises a plurality of key words used for identifying travel information data and a plurality of key words used for identifying key information fields and feedback trend information from unstructured data, and the feedback trend information is information of preference and dislike trend feedback of a social media platform user on a travel item.
8. The multi-modal interaction based travel planning system of claim 6 wherein the travel planning agent creation module comprises: The first associated data acquisition sub-module is used for constructing an information retrieval agent with a natural language intelligent question-answering function based on the AI large model so as to extract first constraint information from dialogue contents of a user, call a data query (MCP) service and retrieve and structuralized process first associated data; The labeling text generation sub-module is used for constructing information screening agents based on the multi-mode visual large model, carrying out content understanding on the first associated data, screening content patches meeting preset quality standards and generating labeling texts; The trip plan form generation sub-module is used for constructing a trip plan agent based on the AI large model, extracting second constraint information from dialogue contents of a user to carry out product screening and trip planning, and generating a trip plan form; The first constraint information comprises a travel place, the first association data refers to a plurality of key information fields and feedback trend information which are retrieved from a product database and an information database and accord with the first constraint information, and the original content paste is obtained again based on index information, the second constraint information comprises a rigid constraint and a flexible constraint, the rigid constraint comprises budget, number of people, time and weather, and the flexible constraint comprises travel preference.
9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the steps of the multi-modal interaction based travel planning method according to any one of claims 1 to 5 when the computer program is executed.
10. A computer readable storage medium storing a computer program, wherein the computer program when executed by a processor implements the steps of the multimodal interaction based travel planning method of any of claims 1 to 5.

Description

Travel planning method and system based on multi-modal interaction Technical Field The application relates to the technical field of artificial intelligence, in particular to a travel planning method and system based on multi-modal interaction. Background At present, a traveler with travel and travel requirements mainly views travel consultation, travel attack and user evaluation on social media and travel service platforms, and then makes a travel plan, however, information which can be seen by the traveler on each platform is limited by platform camping services, multiple types of information are difficult to collect on one platform, and the personal information collecting capability of the traveler is limited, so that time and labor are wasted when the travel plan is made. Therefore, the above-mentioned related art has a problem in that the efficiency of making a travel plan by a passenger is low. Disclosure of Invention In order to improve the efficiency of travel planning, the application provides a travel planning method and system based on multi-modal interaction. The first technical scheme adopted by the application is as follows: a travel planning method based on multi-modal interaction comprises the following steps: S1, setting a data transmission interface for connecting a product database of a travel product platform and an information database of a social media platform, and accessing a semantic analysis model to establish a corresponding standardized data query service; S2, defining semantic information of the standardized data query service in a semantic analysis model, and packaging the standardized data query service into a data query MCP service; S3, constructing an information retrieval agent, an information screening agent and a trip planning agent based on an AI large model, and further creating a trip planning agent comprising the information retrieval agent, the information screening agent and the trip planning agent; And S4, inquiring the trip planning agent docking data to an MCP service and a map platform MCP service, and setting an output rendering component for the trip planning agent. By adopting the technical scheme, the data transmission interfaces of the product database connected with the travel product platform and the information database of the social media platform are arranged so as to obtain product and information data from the travel product platform and the social media platform in real time, the data are accessed into a semantic analysis model to carry out semantic analysis on the obtained data, the data support of subsequent interactive travel planning is used for establishing standardized data query service, semantic information of the standardized data query service is defined in the semantic analysis model so as to perfect a semantic analysis function, the standardized data query service is packaged into a data query MCP service, subsequent access of an agent constructed based on an AI large model is facilitated, an information retrieval agent, an information screening agent and a travel planning agent comprising the information retrieval agent, the information screening agent and the travel planning agent are further established, functional combination travel planning agents are realized, efficient planning is facilitated for passengers in the subsequent process, the travel planning agent is connected with the data query service and a map platform MCP service through the introduction of the MCP service, compared with a traditional text-based retrieval enhancement frame, the data query system is free from the fact that the data is set up in a new manner, the real-time knowledge base can be output and the data can be conveniently and synchronously output. In a preferred example, the method comprises the steps of setting a data transmission interface for connecting a product database of a travel product platform and an information database of a social media platform, and accessing a semantic analysis model to establish a corresponding standardized data query service, and further comprises the following steps: S11, extracting travel product data from a travel agency enterprise system based on a data acquisition period, extracting key information fields from the travel product data and storing the key information fields in a product database of the travel product platform; S12, acquiring travel information data from a social media platform based on a data acquisition period and a keyword group, extracting a keyword information field and feedback trend information from the travel information data, and storing the keyword information field and feedback trend information into an information database of the social media platform; s13, setting a data transmission interface and a semantic analysis model for accessing travel planning agents to the product database and the information database, and establishing standardized data query service; the key information fields co