AI-based trip planning methods, electronic devices, and storage media

By setting up personalized itinerary planning assistants for different travel scenarios in travel service applications and automatically loading target parameters, the problem of users repeatedly describing their needs in different travel scenarios is solved, and an efficient and accurate itinerary planning and booking process is achieved.

CN123117399APending Publication Date: 2026-11-17浙江飞猪网络技术有限公司
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611073918.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-07-17
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

In existing travel service applications, users need to repeatedly describe their travel needs in different travel scenarios, resulting in high interaction input costs and low matching degree of itinerary plans, leading to a cumbersome and inefficient planning process.

Method used

By presenting multiple trip planning assistants in a graphical user interface, each assistant is associated with personalized parameters for a specific travel scenario, and target parameters are automatically loaded, reducing the number of interaction rounds and improving the matching degree of the plan.

Benefits of technology

It reduces the burden of repetitive operations when switching between different scenarios, improves the interactive efficiency and matching degree of itinerary planning, and shortens the decision-making and operation chain.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123117399A_ABST
    Figure CN123117399A_ABST
Patent Text Reader

Abstract

This application provides an AI-based itinerary planning method, electronic device, and storage medium, relating to the field of computer technology. The method includes: pre-binding personalized travel parameters of the same user in different travel scenarios with corresponding itinerary planning assistants; automatically loading corresponding parameters when the user selects a target AI assistant, thereby achieving rapid access to personalized travel parameters and accurate scenario-based matching. This reduces the number of human-computer interaction rounds required for users to supplement constraints through multiple rounds of dialogue, lowers the interactive input cost of itinerary planning, and enables the itinerary plan generated by the system to highly match the user's true intention in the current travel scenario. It avoids the mixed features of the plan caused by generalized recommendations. By directly providing the booking entry corresponding to the itinerary plan in the AI ​​dialogue window, it shortens the decision-making and operation link from itinerary planning to product booking for users, and improves the overall interactive efficiency of itinerary planning and the user's travel decision-making efficiency.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to an AI-based trip planning method, an electronic device, and a computer-readable storage medium. Background Technology

[0002] In the travel services sector, a complete travel experience typically involves multiple stages, including pre-trip information gathering and planning, booking comparison and decision-making, order management and personalized services after booking, and post-trip after-sales processing and receipt collection. With the increasing sophistication of online travel agency (OTA) services, users have become accustomed to independently completing these stages within the platform using a graphical user interface (GUI).

[0003] In relevant travel service applications, a unified, global AI dialogue assistant or search portal is typically provided to receive users' itinerary planning requests. In this process, users either filter their requests through the AI ​​dialogue assistant and complete the itinerary planning through multiple rounds of dialogue; or they prepare travel guides in advance and then search for matching travel service products based on those guides. However, in practical applications, the same user often has different travel scenarios (e.g., business travel, family travel, road trips, trips with friends), and these different scenarios often have significant differences in the requirements for travel service products such as flight times, hotel locations, room types and facilities, and attraction arrangements. For example, business travel prioritizes efficient completion of business activities, placing high demands on transportation timeliness, hotel location, and reimbursement compliance; while family travel focuses on the comfort and experience of family members, giving more consideration to family-friendly accommodations, age-appropriate attractions, and itinerary intensity.

[0004] Therefore, this trip planning method has at least the following shortcomings when facing different travel needs of the same user: First, the interaction input cost is high, and the planning process is cumbersome. On the one hand, since the relevant global AI assistant often needs to collect the user's travel needs through multiple rounds of dialogue, if the user's input information is too brief, the AI ​​assistant cannot understand the preferences of the current scenario, leading to an increase in the number of human-computer interaction rounds, increasing the user's input burden, and reducing the efficiency of trip planning. On the other hand, users need to re-describe their travel needs for different travel scenarios, and the trip planning process for each travel scenario itself involves multiple operations such as destination confirmation, time selection, transportation comparison, and accommodation selection. Each time a scenario is switched, the entire planning process needs to be experienced, further adding the burden of repetitive operations across scenarios, resulting in a significant increase in the overall interaction cost and a cumbersome and inefficient planning process.

[0005] Second, the matching degree of itinerary options is low, resulting in low efficiency in filtering and decision-making. Without collecting sufficient information on users' travel constraints, AI assistants typically only make generalized recommendations based on users' historical behavior data or popular preferences. This leads to the returned itinerary plans often mixing features from different scenarios, resulting in low relevance to the user's current true intentions. Consequently, users need to manually filter and compare numerous returned options, greatly increasing decision-making time and reducing the overall interactive efficiency of itinerary planning. Summary of the Invention

[0006] This application provides an AI-based trip planning method, device, electronic device, and computer-readable storage medium to address the problems of high human-computer interaction input costs, low trip plan matching, and low user decision-making efficiency encountered by some users during the trip planning process.

[0007] This application discloses an AI-based itinerary planning method. It provides a graphical user interface (GUI) via an electronic device. The GUI presents a travel service AI assistant provided by a travel service application. The travel service AI assistant includes at least two itinerary planning assistants, each associated with personalized travel parameters for different travel scenarios of the same user. The method includes: In response to a triggering operation that selects a target AI assistant from at least one of the trip planning assistants, an AI dialogue window is displayed, and target personalized travel parameters associated with the target AI assistant are determined; In response to an input operation to the AI ​​dialogue window, the AI ​​dialogue window displays the itinerary planning requirement information corresponding to the input operation, and displays at least one itinerary plan that matches the itinerary planning requirement information and the target personalized travel parameters, as well as the booking entry corresponding to the itinerary plan.

[0008] This application also discloses an electronic device, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; The memory is used to store computer programs; When the processor executes a program stored in the memory, it implements the method described in the embodiments of this application.

[0009] This application also discloses a computer-readable storage medium storing instructions that, when executed by one or more processors, cause the processors to perform the methods described in this application.

[0010] The embodiments of this application have the following advantages: In this embodiment, the electronic device presents a travel service AI assistant containing at least two trip planning assistants in a graphical user interface. These different trip planning assistants are associated with personalized travel parameters for the same user in different travel scenarios. Then, in response to a trigger operation that selects a target AI assistant from at least one trip planning assistant, an AI dialogue window is displayed, and the target personalized travel parameters associated with that target AI assistant are automatically determined. Next, in response to an input operation to the AI ​​dialogue window, the trip planning requirement information corresponding to the input operation is displayed in the AI ​​dialogue window, along with at least one trip plan matching the trip planning requirement information and the target personalized travel parameters, and a booking entry corresponding to the trip plan. This achieves the goal of pre-binding the personalized travel parameters of the same user in different travel scenarios with different trip planning assistants and automatically loading the matching mechanism of the corresponding scenario parameters when the user triggers a selection. The system enables rapid access to personalized travel parameters and precise scenario-based matching. Furthermore, by associating personalized travel parameters for different travel scenarios with corresponding itinerary planning assistants, users can select appropriate personalized travel parameters based on their actual needs. This reduces the burden of repetitive operations when switching between scenarios. It also significantly reduces the number of human-computer interaction rounds required for users to supplement constraints through multiple dialogues, lowering the interactive input cost of itinerary planning. Moreover, the system-generated itinerary plans highly match the user's true intentions in the current travel scenario, avoiding the mixed features of generalized recommendations. This enhances the scenario awareness and service consistency of travel service applications when facing differentiated travel needs of the same user. Additionally, by directly providing the booking entry point corresponding to the itinerary plan in the AI ​​dialogue window, the system shortens the decision-making and operation chain from itinerary planning to product booking, improving the overall interactive efficiency of itinerary planning and the efficiency of user travel decisions. Attached Figure Description

[0011] Figure 1This is a flowchart illustrating the steps of an AI-based itinerary planning method provided in this application embodiment; Figure 2 This is a schematic diagram of the AI ​​operation interface provided in the embodiments of this application; Figure 3 This is a schematic diagram of the assistant page provided in the embodiments of this application; Figure 4 This is a schematic diagram of the assistant creation page provided in the embodiments of this application; Figure 5 This is a schematic diagram of the AI ​​dialogue window provided in the embodiments of this application; Figure 6 This is a schematic diagram of the AI ​​dialogue window provided in the embodiments of this application; Figure 7 This is a schematic diagram of the AI ​​dialogue window provided in the embodiments of this application; Figure 8 This is a schematic diagram of the AI ​​dialogue window provided in the embodiments of this application; Figure 9 This is a schematic diagram of the AI ​​dialogue window provided in the embodiments of this application; Figure 10 This is a schematic diagram of the AI ​​dialogue window provided in the embodiments of this application; Figure 11 This is a schematic diagram of the application page provided in the embodiments of this application; Figure 12 This is a schematic diagram of the order details page provided in the embodiments of this application; Figure 13 This is a schematic diagram of the AI ​​dialogue window provided in the embodiments of this application. Detailed Implementation

[0012] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0013] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation entry points are provided for users to choose to authorize or refuse.

[0014] As examples, travel service apps typically offer a unified, global AI dialogue assistant or search portal. Users interact with the AI ​​dialogue assistant through multiple rounds to provide information about their travel needs, or search for matching travel services based on travel guides to complete pre-trip planning. However, in practice, the same user often has multiple different travel scenarios (e.g., business trips, family trips, road trips, trips with friends), and the needs for travel services such as flight times, hotel locations, room types and facilities, and attraction arrangements vary significantly across different scenarios. Under this unified AI dialogue assistant architecture, if the user's input is brief, the AI ​​assistant cannot understand the user's specific preferences in the current travel scenario and needs to gradually collect constraints through multiple rounds of follow-up questions, leading to an increase in human-computer interaction rounds and a heavier input burden on the user. At the same time, when switching between different travel scenarios, users need to redescribe their travel needs for the current scenario each time and go through the entire process of destination confirmation, time selection, transportation comparison, and accommodation selection, further adding to the burden of repetitive operations across scenarios, resulting in a significant increase in overall interaction costs and a cumbersome and inefficient trip planning process. Meanwhile, since the unified AI assistant cannot distinguish users' personalized preferences according to travel scenarios, it can usually only make generalized recommendations based on users' historical behavior data or popular preferences when planning trips. The generated trip plans often mix the characteristics of different travel scenarios and have a low degree of consistency with the user's true intention in the current travel scenario. Users need to manually screen and compare the numerous returned plans, which further increases decision-making time and reduces the overall interactive efficiency of trip planning.

[0015] In response, this application pre-binds personalized travel parameters for the same user in different travel scenarios with different trip planning assistants, and automatically loads the matching mechanism of the corresponding scenario parameters when the user triggers a selection. This achieves rapid retrieval and accurate scenario-based matching of target personalized travel parameters. Furthermore, by associating personalized travel parameters for different travel scenarios with corresponding trip planning assistants, users can select appropriate personalized travel parameters according to their actual needs, reducing the burden of repetitive operations when switching between scenarios. This significantly reduces the number of human-computer interaction rounds required for users to supplement constraints through multiple rounds of dialogue, lowering the interactive input cost of trip planning. Moreover, it ensures that the trip plans generated by the system highly match the user's true intentions in the current travel scenario, avoiding the feature mixing caused by generalized recommendations. This enhances the scenario perception capability and service consistency of travel service applications when facing differentiated travel needs of the same user. Additionally, by directly providing the booking entry corresponding to the trip plan in the AI ​​dialogue window, the decision-making and operation chain from trip planning to product booking is shortened, improving the overall interactive efficiency of trip planning and the user's travel decision-making efficiency.

[0016] To enable those skilled in the art to better understand the technical solutions in the embodiments of this application, some technical features involved in the embodiments of this application are explained and described below: In this embodiment, the graphical user interface (GUI) is a visual interface environment presented on the display screen of an electronic device, where the user interacts with the travel service application. When the electronic device is running the travel service application, the GUI can display the currently viewed traditional function pages of the travel service application, such as the homepage, list page, search page, product details page, order details page, personal center page, message page, and itinerary planning page. These traditional function pages are the main carriers for the travel service application to display travel service product information, transaction entry points, and user operation paths to the user, constituting the main scenario for the user to perform operations such as travel planning, product browsing, and booking decisions.

[0017] In this application embodiment, AI UI refers to the collective term for visual interface elements and interaction logic displayed in conjunction with traditional function pages in a graphical user interface, used to provide users with opportunities to interact with the AI ​​system. The methods of this associated display include, but are not limited to: AI UI embedded in the view hierarchy of traditional function pages, floating above the view hierarchy of traditional function pages, and parallel display with traditional function pages in a split-screen format, etc., which this application does not limit.

[0018] In addition, the AI ​​UI includes at least AI components and AI dialogue windows. Among them, AI components serve as the entry point for the AI ​​UI, representing the AI ​​functions in traditional functional pages and enabling user access; the AI ​​dialogue window serves as the interactive platform for the AI ​​UI, displaying the dialogue content in the AI ​​conversation and receiving user input.

[0019] Furthermore, AI components and AI dialogue windows are interconnected at the session level. That is, the same AI dialogue session can be created and invoked in AI components of different component forms and continued in the AI ​​dialogue window. The dialogue content in the AI ​​dialogue window can be presented in summary form in the associated area of ​​the AI ​​component or in the session storage control after the AI ​​dialogue window is closed. Users can quickly restore the AI ​​dialogue window and continue the same AI dialogue session by triggering the AI ​​component or the session storage control.

[0020] It should be noted that the AI ​​component has multiple switchable component forms in the graphical user interface, with different component forms having different size parameters and information carrying capacity. In the embodiments of this application, the AI ​​UI can adaptively determine the target component form based on the page context information of the current traditional function page, and present it in the current traditional function page in the target component form, so that the AI ​​UI can be presented in different traditional function pages with an information carrying capacity that adapts to the needs of the current page scenario. At the same time, the AI ​​UI can switch the AI ​​component from the current component form to another component form in response to the user's command to change the form of the AI ​​component, or in response to a page switching event, and maintain the continuity of the AI ​​dialogue session before and after the switch. That is, the AI ​​component after the switch shares the same AI dialogue session as the AI ​​component before the switch, and the dialogue content displayed in the AI ​​dialogue window remains consistent before and after the switch, so the user does not need to re-initiate AI interaction due to changes in the form of the AI ​​component or page switching.

[0021] Reference Figure 1 This document illustrates a flowchart of the steps involved in an AI-based itinerary planning method provided in this embodiment. A graphical user interface (GUI) is provided via an electronic device. The GUI displays content including a travel service AI assistant provided by a travel service application. This AI assistant includes at least two itinerary planning assistants, each associated with personalized travel parameters for different travel scenarios of the same user. Specifically, the method may include the following steps: Step 101: In response to a triggering operation that selects a target AI assistant from at least one of the trip planning assistants, display an AI dialogue window and determine the target personalized travel parameters associated with the target AI assistant; In this embodiment, the method described above can be applied to an electronic device that may have a travel service application provided by a travel service platform (e.g., an OTA platform, hereinafter referred to as the application) installed on it. The electronic device may include a smartphone, tablet, wearable device, personal computer, in-vehicle terminal, or other terminal device with display capabilities and network communication capabilities. The electronic device provides a graphical user interface (GUI) through its display screen, and the content presented in the GUI includes a travel service AI assistant provided by the travel service application.

[0022] The travel service application can communicate with the server of the travel service platform via the network to obtain travel service product data (such as flight information, hotel information, attraction ticket information, etc.) required for itinerary planning. Correspondingly, the user of the travel service application can be a natural person who has registered and has an account on the travel service platform. This user can use the travel service application for itinerary planning in different travel scenarios, such as business trips, road trips, family trips, couples' vacations, and trips with friends. It is understood that the above examples are only used to illustrate the application scenarios of this application and do not constitute a limitation on the application scenarios of this application.

[0023] Optionally, the travel service AI assistant can be an AI dialogue function module integrated into a travel service application. It can be presented in the graphical user interface as a standalone page within the application, a floating window, a bottom toolbar entry, or an embedded component; this application does not impose specific limitations on this. The travel service AI assistant may include at least two itinerary planning assistants, each corresponding to a different travel scenario. For example, a business travel assistant for business trips, a self-driving assistant for self-driving travel, and a family travel assistant for family trips.

[0024] It should be noted that the types of itinerary planning assistants mentioned above can be predefined and configured within the travel service application, or users can customize and add them to the travel service AI assistant according to their own needs. In a specific example, the travel service AI assistant is presented in the graphical user interface as a horizontally sliding card list. Each card corresponds to a type of itinerary planning assistant, and the card displays the name and brief description of the itinerary planning assistant (such as "Travel Assistant - Intelligent planning of business trips, automatically matching business trip standards and membership benefits", "Driving Assistant - Intelligent route planning, matching driving preferences and supply points"). Users can trigger the selection of the corresponding target AI assistant by clicking on any card, and this application does not impose any restrictions on this.

[0025] In practice, different trip planning assistants can be associated with personalized travel parameters for the same user in different travel scenarios. These personalized travel parameters can be a set of preferences and constraints pre-set by the user for a specific travel scenario, matching that scenario. These can include, but are not limited to, transportation budget, accommodation budget, hotel brand preference, frequent flyer membership information, travel compliance policies, vehicle type preference, driving intensity preference, route style preference, children's age information, activity intensity preference, and facility requirements preference. Optionally, personalized travel parameters are stored in the server-side database of the travel service application at the user account level, bound to the user's identity identifier (e.g., user ID). Multiple sets of different personalized travel parameters can be stored under each user account, each corresponding to a different trip planning assistant.

[0026] Each set of personalized travel parameters can be configured with a corresponding assistant type identifier (e.g., "Business Trip Assistant," "Road Trip Assistant," "Family Travel Assistant," etc.). Once a trip planning assistant is associated with a set of personalized travel parameters, this association is persistently stored in the server-side database, forming a data mapping relationship of "User ID + Trip Planning Assistant Type + Personalized Travel Parameter Set." When a user logs into the same account on a different terminal device, the electronic device synchronizes all of the user's trip planning assistants and their associated personalized travel parameters from the server, ensuring consistency across devices.

[0027] In some feasible implementations, the server-side database of the travel service application can maintain a "user itinerary preference table" for each user. This table can be structured with the following fields: user ID, scenario type (e.g., "business trip," "self-driving tour," "family trip"), budget limit (e.g., "transportation budget limit," "accommodation budget limit"), brand preference (e.g., "preferred hotel group," "preferred airline"), membership level (e.g., "hotel group gold card," "airline silver card"), special preferences (e.g., "non-smoking room preference," "high-floor preference," "child's age"), creation time, and update time. When a user uses a trip planning assistant for the first time and completes the personalized travel parameter settings, the electronic device can upload this set of parameters to the server via an API interface. The server then associates and stores this set of parameters with the user ID and the corresponding scenario type. When the user triggers the same trip planning assistant again, the electronic device retrieves this set of parameters from the server and displays it in the AI ​​dialogue window.

[0028] In other feasible implementations, personalized travel parameters can be pre-set by users through third-party applications or web pages, and then synchronized to the server of the travel service application after setting. For example, users can pre-enter travel standards (such as upper limits for air class and accommodation price) in a corporate travel management system. The travel service application obtains this information through API integration with the corporate travel management system and uses it as part of the personalized travel parameters associated with the travel assistant.

[0029] In this embodiment, the trip planning assistant includes business travel assistants, self-driving assistants, and family travel assistants. During the selection process of the trip planning assistant, an AI dialogue window can be displayed in response to a trigger operation targeting the AI ​​assistant. The selection operation can be a user's click, touch, or voice selection of the trip planning assistant card or chart. The electronic device, responding to this selection operation, can designate the user-selected trip planning assistant as the target AI assistant and then display the corresponding AI dialogue window. This AI dialogue window serves as the interface area for natural language interaction between the user and the target AI assistant. It can include a dialogue content area (for displaying system responses and user input) and a user input area (for allowing the user to input text, voice, or selection guidance prompts). Simultaneously, the electronic device can also, based on the type of the selected target AI assistant, search and determine the target personalized travel parameters corresponding to that target AI assistant from a pre-stored set of personalized travel parameters associated with the user. If a corresponding personalized travel parameter is found, it will be loaded into the current session context for use in subsequent itinerary matching. If no corresponding personalized travel parameter is found, itinerary constraint guidance information will be displayed in the AI ​​dialogue window to prompt the user to set personalized travel parameters for the first time in the current scenario. The specific implementation method will be further explained below.

[0030] Specifically, when a target AI assistant's associated personalized travel parameters are detected, if the target AI assistant is a business travel assistant, the associated personalized travel parameters and their modification entry will be displayed in the AI ​​dialogue window; if the target AI assistant is a self-driving assistant, the associated personalized self-driving parameters and their modification entry will be displayed in the AI ​​dialogue window; and if the target AI assistant is a family travel assistant, the associated family travel parameters and their modification entry will be displayed in the AI ​​dialogue window. Furthermore, when a user needs to adjust personalized travel parameters, the user can adjust any one or more of the associated personalized travel parameters through this parameter modification entry. The electronic device will respond to the user's adjustment operation, update and save the corresponding personalized travel parameters.

[0031] Accordingly, if no target personalized travel parameters are detected, i.e., when a user first uses a target AI assistant and has not yet set corresponding personalized travel parameters, the electronic device displays corresponding travel constraint guidance information and preference controls in the AI ​​dialogue window, depending on the type of the target AI assistant. Specifically, if the target AI assistant is a travel assistant, travel itinerary constraint guidance information and different types of standard travel controls can be displayed in the AI ​​dialogue window. These standard travel controls include, but are not limited to, at least one of the following: transportation budget, accommodation budget, hotel brand, frequent flyer membership, travel compliance policy, and special accommodation needs information. The system includes several key features: a transportation budget control to set the maximum cost for a single trip; a hotel brand preference control to select one or more preferred hotel groups (e.g., Marriott, InterContinental, Hilton); a frequent flyer membership control to input or select the membership level of an airline or hotel group (e.g., a silver card for an airline, a gold card for a hotel group) so that the system prioritizes options that offer accumulated benefits or membership privileges when recommending travel plans; a travel compliance policy control to set compliance options such as allowing users to pay for excess costs (i.e., users bear the cost of expenses exceeding the standard travel allowance) and allowing shared accommodation; and a special accommodation needs control to set preferences such as non-smoking rooms, high-floor rooms, and quiet rooms. Users set personalized travel parameters for business trips by manipulating the aforementioned controls. The electronic device responds to these settings by associating and storing the user-defined parameters with the travel assistant. For example, if a user sets a transportation budget of under 1000 yuan, an accommodation budget of under 500 yuan, a preference for Marriott hotels, Marriott Gold status as a frequent flyer, and special accommodation requirements such as non-smoking rooms and high floors, the electronic device packages these parameters into a personalized travel parameter set and stores it in the server-side database using the user ID and "travel assistant" as the key. Subsequent times the user triggers the travel assistant, the electronic device automatically uses this parameter set for itinerary matching.

[0032] If the target AI assistant is a self-driving assistant, the AI ​​dialogue window will display self-driving trip constraint guidance information and provide different types of self-driving preference controls. These self-driving preference controls include at least one of the following: vehicle type, driving intensity preference, route style preference, companion type, waypoint preference, and supply preference. Specifically, the vehicle type preference control may include options such as SUV preference, sedan preference, off-road vehicle preference, MPV preference, and convertible preference. Different vehicle types affect route planning strategies (e.g., off-road vehicles can plan routes including unpaved roads, while sedans are not suitable). The driving intensity preference control may include options such as daily driving mileage limit (e.g., no more than 300 km / day), single continuous driving time limit (e.g., no more than 3 hours / trip), and whether night driving is accepted (e.g., "do not accept driving after 22:00"), used to control daily trip node arrangements and rest stop planning. The route style preference control can... Options such as highway priority, national highway priority, scenic route priority, shortest time priority, and shortest distance priority are used to influence route preferences and stop recommendations along the way. The companion type control can include options such as solo travel, couples traveling, families traveling, and friends traveling together. The waypoint preference control is used to set whether to allow detours through specific locations (such as a preference for passing through a particular city or scenic spot) and whether to prefer the type of service area along the way. The supply preference control is used to set preferences for charging station coverage, gas stations, and restaurant stops; for example, under the preference for new energy vehicles, routes with higher charging station coverage are prioritized. Users complete the personalized travel parameters settings for self-driving scenarios by operating the above controls, and the electronic device associates and stores the set parameters with the self-driving assistant.

[0033] If the target AI assistant is a parent-child travel assistant, then parent-child travel constraint guidance information will be displayed in the AI ​​dialogue window, along with an entry point for adding child information and / or parent-child travel preference controls. These parent-child travel preference controls include at least one of the following: theme preference, mode of transportation preference, activity intensity preference, facility requirement preference, and safety requirement preference. The entry point for adding child information allows the user to input the child's age or age range. The electronic device will determine the appropriate recommendation strategy for different age groups based on the child's age information (e.g., preschool children are suitable for zoos, aquariums, and other observation- and interaction-based activities, while school-aged children are suitable for science museums, museums, and other educational activities). The theme preference controls can include options such as animal theme preferences, nature exploration theme preferences, theme park preferences, museum / science museum preferences, and beach preferences; the travel mode preference controls can include options such as driving preferences, public transportation preferences, and pedestrian-friendly preferences; the activity intensity preference controls can include options such as relaxed and leisurely preferences, moderate activity preferences, and fulfilling and compact preferences, to consider children's physical strength and avoid overexertion; the facility need preference controls can include options such as whether there is a need for a mother and baby room, whether there is a need for a family restroom, whether there is a need for stroller rental, and whether there is a need for children's meals; the safety need preference controls can include options such as crowd flow preferences (e.g., avoiding overly crowded areas), whether there are safety care services, and whether there are emergency medical points. Users enter their child's age information through the child information addition entry and set personalized travel parameters for parent-child travel scenarios through the above preference controls. The electronic device will associate and store the set parameters with the parent-child travel assistant.

[0034] In addition to the aforementioned trip planning assistant, the application can also provide an AI assistant customization function, allowing users to create a personalized trip planning assistant based on their unique travel scenarios. Optionally, the electronic device can also display a customization window in response to the AI ​​assistant customization operation. The customization window includes at least a role description control and a personal preference control. Then, the user can customize the AI ​​assistant settings through the role description control and the personal preference control. Based on the user's customization operation, the electronic device can obtain role description information in response to input operations on the role description control, and obtain personal preference information in response to input operations on the personal preference control. Finally, in response to the assistant creation operation, the corresponding target AI assistant is created based on the role description information and personal preference information.

[0035] In its implementation, the electronic device can provide a custom entry point for the AI ​​assistant in its graphical user interface (e.g., displaying a "Create Custom Assistant" button at the end of the trip planning assistant list). In response to user actions triggering this custom entry point, a custom window is displayed. This custom window includes at least a role description control and a personal preference control. The role description control allows the user to input a description of the target AI assistant, such as "I am a traveler who enjoys outdoor activities and prefers natural scenery and hiking routes." The personal preference control allows the user to input specific preferences, such as "Daily hiking not exceeding 10 kilometers" or "Preferring guesthouses to hotels." Users can input information into these controls via text input. The electronic device retrieves the role description information in response to input to the role description control and the personal preference information in response to input to the personal preference control. Subsequently, in response to user actions triggering the assistant creation (e.g., clicking the "Create" button in the custom window), the electronic device creates the corresponding target AI assistant based on the role description and personal preference information and adds the created target AI assistant to the trip planning assistant list of the travel service AI assistant. During the creation process, electronic devices can also use Large Language Models (LLM) to refine the role description information entered by the user, generate structured system instructions, and submit them to the user for secondary confirmation. Only after the user confirms will the final creation and saving operations be performed. In addition, to protect user privacy and platform content security, electronic devices can also conduct risk control audits on the role description information and personal preference information involved in the creation process, such as sensitive word verification. Only after the audit is passed can the information be saved and take effect.

[0036] Step 102: In response to an input operation to the AI ​​dialogue window, display the itinerary planning requirement information corresponding to the input operation in the AI ​​dialogue window, and display at least one itinerary plan that matches the itinerary planning requirement information and the target personalized travel parameters, as well as the booking entry corresponding to the itinerary plan.

[0037] After a user selects a target AI assistant, they can interact with it to complete the corresponding itinerary planning. Users can perform input operations in the user input area of ​​the AI ​​dialogue window. These input operations can include text information entered by the user (such as manually typed text on a keyboard), voice information (such as voice commands entered via voice input), and selection of guidance prompts displayed in the AI ​​dialogue window.

[0038] The AI ​​dialogue window can pre-display at least one guiding prompt. This prompt can be a shortcut automatically generated based on the target's personalized travel parameters, or it can be a shortcut generated based on the content displayed on the corresponding traditional function page. For example, when the target AI assistant is a business travel assistant, the prompt could include "Help me plan my business trip" or "Book flights and hotels to Shanghai"; when the target AI assistant is a self-driving assistant, the prompt could include "Plan a self-driving route from Hangzhou to Chengdu" or "Recommend attractions along the way"; when the target AI assistant is a family travel assistant, the prompt could include "Recommend attractions suitable for family travel" or "Plan a weekend trip with kids"; or, if the user triggers an AI dialogue on a hotel's product details page, it can generate a corresponding guiding prompt for that hotel, such as "Does the xx hotel have children's entertainment facilities?" Users can quickly initiate a trip planning request by clicking on any of the guiding prompts, without having to manually enter a complete description of their travel needs.

[0039] Based on user input, the electronic device can display the corresponding travel planning needs information in the dialogue content area of ​​the AI ​​dialogue window. Simultaneously, the device can parse this information to extract structured data, such as the user's core travel requests, including destination information (e.g., "Shanghai" or "West Lake Scenic Area in Hangzhou"), travel time information (e.g., "next Monday" or "July 20th-July 22nd"), travel duration information (e.g., "three days" or "two nights"), and travel companion information (e.g., "traveling with my 5-year-old daughter" or "traveling with three colleagues"). In some feasible implementations, the electronic device can use a Natural Language Understanding (NLU) model to perform intent recognition and slot filling on the user's input text, extracting the aforementioned structured information. This NLU model can employ a pre-trained language model based on the Transformer architecture (e.g., BERT, ERINE, etc.), fine-tuned and trained on a large corpus of dialogues from various travel service scenarios, enabling it to identify intents from user input (e.g., "travel planning intent," "attraction recommendation intent," etc.) and extract corresponding slots (e.g., "destination," "travel date," "travel duration," etc.). The specific implementation of the above natural language understanding model can adopt any applicable method known to those skilled in the art, and this application does not impose any restrictions on it.

[0040] After the electronic device extracts the relevant structured information, it uses it together with the aforementioned determined personalized travel parameters as constraints. This information is then filtered and matched within the available travel service product data to generate at least one itinerary plan that matches all conditions in the constraint set. Optionally, the electronic device can first determine the range of travel service products to be queried based on the destination and travel time information in the itinerary planning requirements (e.g., flights and high-speed rail services operating from Hangzhou to Shanghai on a specified date, hotels in the specified destination and surrounding areas available for stay on the travel date, etc.). Then, the electronic device obtains travel service product data within the aforementioned range from the travel service platform's server, including transportation product information (such as flight number, departure time, arrival time, price, cabin class, airline, etc.; high-speed rail number, departure time, arrival time, price, seat class, etc.) and accommodation product information (such as hotel name, address, room type, price, brand, facilities, membership benefits, etc.). Next, the electronic device can use trip planning requirements (such as arrival time no later than 9:00 AM) as a hard constraint and various parameters in the target personalized travel parameters (such as transportation budget limit, accommodation budget limit, hotel brand preference, frequent flyer membership, etc.) as adjustable constraints (where the budget limit is a hard constraint, and brand preference and membership are soft constraints) to perform multi-level filtering and scoring of travel service products within the above range.

[0041] In the multi-level filtering and scoring process, the electronic device first uses hard constraints (must-arrive time, budget limit) to filter out non-compliant products, obtaining candidate sets of transportation and accommodation products. Then, for each product in the candidate sets, the electronic device calculates a suitability score based on soft constraints (brand preference, frequent flyer membership, special accommodation needs, etc.). Bonus points are awarded for hotels that match brand preferences and flights that meet frequent flyer membership benefit accumulation conditions. Finally, the device selects at least one combination of transportation and accommodation products with the highest overall score as the corresponding itinerary plan. After generating the itinerary plan, the electronic device can display detailed information about the itinerary plan in the dialogue content area of ​​the AI ​​dialogue window (e.g., "Plan 1: Outbound high-speed train G1234, 7:00-8:15, ticket price 300 yuan; Hotel XX Hotel, double room 480 yuan / night, 2 km from the meeting venue") and the corresponding booking portal (e.g., a "Book" button). Users can click the booking portal to directly enter the booking process for the corresponding travel service product and complete the order.

[0042] Furthermore, in some feasible implementations, after viewing itinerary options in the AI ​​dialogue window, users may wish to directly express their specific purchase requests within the current dialogue without manually clicking a booking entry and being redirected to a separate booking page. To this end, electronic devices support users directly inputting natural language commands containing their purchase requests into the input controls of the AI ​​dialogue window. For example, after seeing a Disney-related itinerary in the AI ​​dialogue window, a user can input "I want to book Disney tickets for 3 adults and 2 children" into the input controls. The electronic device can respond to the purchase request information entered by the user in the AI ​​dialogue window, perform natural language understanding on the user's input text, and extract the traveler information contained within the purchase request information. This traveler information can include at least one of the number of adults and the number of children. Further, after extracting the traveler information, the electronic device uses this traveler information as a query condition to initiate a product inventory query request to the travel service platform's server to search for the existence of a single travel service product that perfectly matches the traveler information. For example, if a user inputs a request for "3 adults and 2 children", and the server's product inventory contains a product that exactly matches the 3 adults + 2 children configuration, the electronic device will use that single product as a candidate travel service product and directly display the product information and booking entry in the AI ​​dialogue window.

[0043] If the electronic device does not find a single travel service product that perfectly matches the number of travelers in the travel service application, the electronic device obtains the number of people supported by each travel service product in the product library (e.g., "2 adults and 2 children" supports 2 adults + 2 children, "1 adult" supports 1 adult, "2 adults and 1 child" supports 2 adults + 1 child, etc.), and uses the number of travelers as the target constraint to filter at least two candidate travel service products from the product library, so that the quantity configuration information and / or traveler type configuration information of the candidate travel service products match the number of travelers as a whole, and there is a complementary relationship between the candidate travel service products. For example, for a travel group of "3 adults and 2 children," if the product library does not contain a single item for "3 adults and 2 children," but does contain items for "2 adults and 2 children" and "1 adult," the electronic device will use both items as candidate travel service items. These two items are complementary in terms of group size: "2 adults and 2 children" covers the needs of 2 adults and 2 children, while "1 adult" covers the needs of the remaining 1 adult. Together, they cover the travel group of "3 adults and 2 children." Similarly, if the product library does not contain a single item for "3 adults and 2 children," but does contain items for "2 adults and 1 child" and "1 adult and 1 child," the electronic device can also use these two items as candidate travel service items.

[0044] After identifying multiple candidate travel service products, the electronic device displays product information and corresponding booking portals for each product in an AI dialogue window. The product information may include product name, target audience type, number of people allowed, price, and usage date. The booking portals for each product may be presented as interactive controls near or below the corresponding product information; this application does not impose any restrictions on this.

[0045] In some feasible implementations, electronic devices can also display prompts within an AI dialogue window that represent the relationships between multiple candidate travel service combinations. For example, a message like "The following combinations have been found, covering the needs of 3 adults and 2 children for park entry; we recommend booking them together" helps users understand why multiple options are displayed simultaneously. Users can confirm the booking entry for each candidate travel service separately within the same AI dialogue window and complete the booking sequentially. This allows the combination of multiple options to meet the needs of multiple travelers without having to manually break down the group size requirements and search for different options one by one. Therefore, when the required number of travelers cannot be met by a single option in the product library, the electronic device can automatically perform intelligent matching and recommendation of product combinations. This avoids the tedious operations of users manually performing multiple rounds of searches outside the dialogue window, checking the inventory of different options separately, and manually comparing whether the group size configuration matches. This reduces the interaction cost and cognitive burden for users when handling group travel booking needs and improves booking efficiency in scenarios with complex travel group requirements.

[0046] Optionally, when displaying itinerary options, the electronic device can also sort and display multiple generated itinerary options according to a certain sorting strategy. For example, it can sort them from low to high total price, from short to long total travel time, or from high to low overall rating. In addition, the electronic device can also annotate the reasons for recommending each option in the displayed itinerary content (such as "This option fits your travel budget and can accumulate airline miles" or "This hotel offers free breakfast and room upgrade for gold card members"). It should be understood that the above sorting and annotation strategies can be configured according to actual needs, and this application does not impose any restrictions on them.

[0047] Furthermore, in other feasible implementations, since the itinerary may include at least one attraction product, after the electronic device displays the itinerary and booking entry in step 102, the user may want to learn more about the target attraction product before deciding whether to complete the booking. In response, the electronic device can jump to and display the product details page of the target attraction product in response to the user's trigger action on the target attraction product in the itinerary (e.g., the user clicks on the attraction name or attraction card displayed in the itinerary), and display the corresponding AI component (e.g., an AI component embedded, floating, or displayed in parallel on the details page). After the user clicks on the AI ​​component, the electronic device displays the AI ​​component in an expanded form, and displays at least one attraction introduction guide and input controls associated with the target attraction product. The attraction introduction guide can be automatically generated according to the type of the target attraction product. For example, for zoos, it can generate questions such as "What activities are suitable for children in the zoo?", "Where is the panda enclosure?", and "Are there stroller rentals?"; for historical sites, it can generate "Recommended tour routes", "Best photo spots", and "How to book guided tours?". After a user clicks on any attraction's information or enters their needs in the input controls, the electronic device determines the attraction's information requirements for that specific attraction, displays an AI dialogue window, and shows the attraction's description information tailored to that request. This description information can include at least one point of interest (e.g., "Panda Pavilion: located in the southwest of the park, feeding time is 10:00-10:30 AM, suitable for children; Children's Ranch: located in the central area of ​​the park, allowing close contact with animals, suitable for parent-child interaction") and booking portals for travel services corresponding to each point of interest (e.g., "Panda Pavilion VIP Experience Ticket," "Ranch Feeding Experience Ticket," etc.). Thus, users can directly obtain targeted attraction introductions and recommendations while browsing attraction details without leaving the current page to search again or switching to other applications.

[0048] Furthermore, after the electronic device returns attraction descriptions based on the user's attraction information needs, the user can continue to input supplementary needs or ask follow-up questions in the same AI dialogue window, such as "Is the first point of interest suitable for the elderly?", "Are there any closer to the entrance?", "Does the boat tour for the second attraction require advance booking?", etc. Responding to the user's continued input in the AI ​​dialogue window, the electronic device combines the newly input content with the aforementioned itinerary planning needs and target personalized travel parameters as an updated set of constraints. It then re-searches, filters, and sorts the point-of-interest data for the target attraction, generating updated attraction descriptions, which are then displayed in the AI ​​dialogue window. These updated descriptions include descriptions of at least one newly matched point of interest, and information on the satisfaction status of each point of interest with the updated set of constraints (e.g., "There is a stroller access"). "Suitable for the elderly" "There are cruise ship activities" The system includes icons such as "[Icons]" and booking portals for travel services corresponding to each point of interest. This allows users to refine their needs through multiple rounds of interaction with the AI ​​dialogue window. The AI ​​dynamically adjusts the recommended points of interest based on each user input, ensuring the final recommendation closely matches the user's true intent. This achieves conversational, interventionist, and iterative intelligent POI recommendations in scenic spot scenarios, avoiding the limitations of traditional solutions where users only receive a one-time recommendation and cannot flexibly adjust it according to their needs. Furthermore, the booking portals for each point of interest remain displayed in each updated scenic spot description, allowing users to directly book experiences corresponding to any point of interest after any round of dialogue, without switching back and forth between the recommendation results and the booking page.

[0049] In another feasible implementation, when displaying attraction descriptions based on user needs, electronic devices can also personalize sorting and filtering by incorporating the user's historical behavioral data (such as browsing history, favorites, and order history) from the travel service platform. For example, if a user's historical orders repeatedly include attraction experiences themed "suitable for children," then when returning to the attraction descriptions, child-related points of interest will be prioritized and labeled with reasons such as "matches your previous preferences." Furthermore, while displaying descriptions of each point of interest, the electronic device can also indicate whether each point of interest meets the user's constraints, such as displaying "stroller access available" on the point of interest card. "There is a mother and baby room" "Suitable for 5-year-old children" "There are cruise ship activities" The use of labels such as "" allows users to quickly determine whether each point of interest matches their needs. The application of the aforementioned historical behavioral data complements the real-time multi-turn interaction in the AI ​​dialogue window—the former provides an initial ranking based on the user's long-term profile, while the latter dynamically adjusts based on the user's immediate feedback in the current dialogue, jointly improving the personalization and interaction efficiency of POI recommendation results.

[0050] Furthermore, in some feasible implementations, after the electronic device displays the itinerary plan and booking entry in step 102, if it detects a user's trigger operation on a certain booking entry, it can respond to the trigger operation by placing an order for the target travel service product corresponding to that booking entry and redirecting to display the order details page corresponding to the target travel service product. Simultaneously, the corresponding AI component is displayed on the order details page. The AI ​​component can be in the form of a floating icon, a fixed bottom bar, a floating button, etc., embedded in the order details page, allowing users to quickly initiate service requests for that order within the order details page.

[0051] In one possible implementation, after a user places an order for a target travel service, if they need to cancel the order due to itinerary changes, plan adjustments, or other reasons, they can initiate a cancellation request through an AI-powered dialogue window. Specifically, the electronic device identifies the user's cancellation request for the target travel service, determines the corresponding cancellation request information, and displays cancellation confirmation information in the AI ​​dialogue window, including order-related information for the target travel service (such as product name, travel date, order amount, etc.) and the expected refund amount. Once the user confirms the cancellation information, the electronic device executes the cancellation operation.

[0052] Furthermore, after a cancellation is completed or during the cancellation process, the electronic device can automatically retrieve related orders of other categories under the same itinerary (such as hotel orders, ticket orders, etc.) using the itinerary information of the cancelled target travel service as an anchor (e.g., destination city, travel date range). If no related orders are found, the corresponding query feedback (e.g., "No other orders related to this itinerary were found") and an order viewing entry (e.g., "View My Orders" button) will be displayed in the AI ​​dialogue window, allowing users to view the entire order list. If related orders are found, the order cancellation guidance will be displayed in the AI ​​dialogue window. This order cancellation guidance will include at least the related order cancellation guidance information (e.g., "We detected that you have booked a stay at XX Hotel in Shanghai on July 20th. Do you want to cancel it as well?") ​​and a related order cancellation entry (e.g., "Cancel it as well" button), allowing users to choose whether to cancel the related orders as well. If a user chooses to cancel all related orders, the electronic device performs a batch cancellation operation on the associated orders and provides real-time feedback on the cancellation result of each order in the AI ​​dialogue window (e.g., "Hotel order cancellation successful, estimated refund of XX yuan"), allowing the user to complete the cancellation of the entire trip within a single AI dialogue window. Furthermore, the electronic device can also display separate cancellation entry points for each associated order within the AI ​​dialogue window, allowing users to individually confirm or cancel the cancellation for each associated order. For example, when two associated orders (Hotel A and Ticket B) are found, two separate entry points, "Cancel Hotel A" and "Cancel Ticket B," are displayed, allowing the user to choose to cancel only a portion of the order.

[0053] In another feasible implementation, after placing an order for a target travel service, users can initiate related order service requests for the ordered product on the order details page. For example, for hotel orders, users can initiate personalized post-booking service requests such as reserving late check-out, reserving early check-in, or specifying a preference for non-smoking rooms. Specifically, the electronic device can respond to the user's triggering operation on the AI ​​component in the order details page, determine the order service request information for the target travel service product (i.e., the ordered product) associated with the order details page, and display an AI dialogue window. The content displayed in the AI ​​dialogue window includes at least: travel service information for at least one travel service item that matches the order service request information, as well as the processing status information corresponding to that travel service item.

[0054] The order service request information refers to the service requirements for a specific travel service product order. If the target travel service product is a hotel product, the order service request information includes sub-requirements such as reserving late check-out, reserving early check-in, specifying non-smoking room preference, specifying high floor preference, specifying quiet room preference, specifying bed type preference, reserving extra bed, reserving extra breakfast, reserving airport transfer service, and supplementing personalized requirements; the travel service information is the hotel name and room type information for the hotel product; the processing status information is the processing result corresponding to the order service request information, such as "accepted", "processing", "confirmed", "unable to meet", etc.

[0055] If the target travel service product is a transportation product (such as air tickets or train tickets), the order service request information includes requests for cancellation, rescheduling, seat selection, adding baggage allowance, and issuing an itinerary; the travel service information includes the departure point, destination, and travel time of the transportation product; and the processing status information includes the progress of cancellation, rescheduling, or itinerary issuance.

[0056] If the target travel service product is an attraction ticket, the order service request information includes requests for cancellation, rescheduling, and invoice issuance; the travel service information is the name of the attraction and the date of entry for the attraction ticket; and the processing status information includes the progress of cancellation, rescheduling, or invoice issuance.

[0057] In some examples, taking hotel bookings as an example, the electronic device displays guidance prompts for hotel orders in an AI dialogue window (such as "Reserve late check-out," "Please specify non-smoking room," "Reserve extra bed," etc.). After the user clicks on the corresponding prompt or enters specific requirements in the input controls (such as "I want to reserve late check-out until 2 PM and add a crib"), the electronic device determines that the order service requirements include two sub-requirements: "Reserve late check-out" and "Reserve extra bed." Based on the hotel booking information (hotel name, contact information, etc.), the electronic device makes a phone call to the hotel front desk via an AI outbound calling platform or sends a message to the hotel to communicate the requirements, combining the two sub-requirements into a single call. After the call, the electronic device obtains the processing results returned by the hotel front desk and displays the processing status information corresponding to each sub-requirement in the AI ​​dialogue window, such as "Late check-out - Confirmed," "Extra bed - Pending confirmation," etc. Thus, users can initiate and track the progress of multiple personalized hotel services after booking in a single AI dialogue window without having to make phone calls themselves or submit different requests through multiple pages.

[0058] It should be noted that in the above embodiments, the AI ​​component provided on the order details page is used as the trigger entry point for illustrative purposes. This means that when the AI ​​dialogue is triggered, the user can input their order service requests. The corresponding AI assistant can then locate the relevant travel service product and execute the processing operation associated with that product. For example, the user can trigger the AI ​​dialogue through the AI ​​entry point provided on other traditional function pages and input "I want to book a late check-out until 2 PM and add a crib" in the AI ​​dialogue window. The AI ​​assistant can automatically query the user's unused orders and display the query results in the AI ​​dialogue window, as well as confirm with the user whether to execute the corresponding order service operation (such as late check-out, adding a crib, etc.). After the user confirms, the electronic device can then call the hotel front desk via the AI ​​outbound calling platform or send a message to the hotel to communicate the request, merging the two sub-requests into a single call to convey the message. Furthermore, in some feasible implementations, electronic devices can determine the corresponding service processing method based on the type of the target travel service product. If the target travel service product is a hotel product, the electronic device, based on the aforementioned determined order service request information, initiates a voice call to the merchant user device to which the target travel service product belongs, or sends a request communication message to the merchant user account to which the target travel service product belongs, and obtains the processing result returned by the merchant user device or merchant user account. The voice call can be implemented through an integrated AI outbound calling platform (such as Xbot), which integrates TTS (text-to-speech) and ASR (automatic speech recognition) capabilities. This platform can automatically call the hotel front desk in the official capacity of the OTA platform, conveying the user's needs and collecting feedback from the hotel through voice interaction. Alternatively, sending a request communication message to the merchant user account can be achieved through a messaging channel between the OTA platform and the merchant (such as the merchant's backend message center, instant messaging IM channel, etc.).

[0059] After the electronic device receives the corresponding processing result, it can display the result in the AI ​​dialogue window. If the target travel service product is a transportation product or an attraction ticket product, the electronic device sends the order service request information (e.g., submitting a cancellation or rebooking request through an API interface) to the service system corresponding to the transportation product or attraction ticket product (such as an airline system, railway ticketing system, attraction ticketing system, etc.), receives the processing result returned by the service system, and then displays the processing result in the AI ​​dialogue window.

[0060] Furthermore, in some feasible implementations, electronic devices can also provide automatic invoicing services after a user completes the fulfillment of a target travel service. Specifically, in response to detecting that the user has completed the fulfillment of the target travel service (e.g., the check-out time for a hotel order has passed, or the flight for a flight order has landed), the electronic device displays invoicing guidance content for the target travel service in an AI dialogue window. If the electronic device finds that the user has already saved invoicing information (such as company name, taxpayer identification number, etc.) on the travel service platform, it displays order-related information for the travel service (such as hotel name, check-in date, check-out date, and cost details) and automatically fills in the corresponding invoice information (such as invoice title, tax number, and amount) in the AI ​​dialogue window. After the user confirms, they can submit an invoice request. If no invoice information is found, the AI ​​dialogue window displays information entry guidance to help the user supplement the invoice information (such as invoice title input controls, tax number input controls, etc.). After the user completes the invoicing information in the AI ​​dialogue window, the electronic device associates the invoice information with the corresponding order and submits the invoice request. After the invoice is issued, the electronic device can collect the electronic invoices according to the travel itinerary, making it convenient for users to view and manage them in a unified manner.

[0061] Based on the content of the above embodiments, the following provides an exemplary description of the above interaction process in conjunction with corresponding specific application scenarios, for example: Scenario 1: Travel assistant usage process in a business trip scenario: User Xiao Wang plans to travel from Hangzhou to Shanghai next Monday for a two-day business meeting. He opens a travel service app on his electronic device and sees three travel planning assistant cards on the main interface of the AI ​​assistant: Business Travel Assistant, Self-Driving Assistant, and Family Travel Assistant. Xiao Wang clicks on the "Business Travel Assistant" card. The device responds to this selection, displaying the AI ​​dialogue window for the Business Travel Assistant. It then queries the server database using the user ID and the "Business Travel Assistant" key to check if the user has already associated personalized travel parameters for a business trip. Since Xiao Wang has previously set personalized travel parameters (transportation budget under 1500 yuan, accommodation budget under 500 yuan, hotel brand preference XX series, frequent flyer membership XX Gold Card, special accommodation requirement non-smoking room), the device, after retrieving these parameters, displays "Your travel preferences have been loaded: transportation budget 1500 yuan, accommodation budget 500 yuan, preference for XX series hotels, XX Gold Card membership, non-smoking room" in the AI ​​dialogue window, while also providing an option to "Modify travel preferences." After confirming the parameters were correct, Xiao Wang entered "Please plan my itinerary for a two-day meeting in Shanghai next Monday, arriving no later than 9:00 AM" into the input control. The electronic device used a natural language understanding model to parse the input text, extracting the destination "Shanghai," travel date "next Monday," trip duration "two days," and required arrival time "before 9:00 AM." Subsequently, the electronic device initiated a transportation product query request to the server, obtaining all high-speed rail and flight schedules from Hangzhou to Shanghai next Monday with an arrival time no later than 9:00 AM. It then filtered these options using a transportation budget of 1500 yuan to obtain a candidate transportation product set. Simultaneously, it retrieved hotel products in the Shanghai area with an available check-in date next Monday and a price not exceeding 500 yuan per night. It then used hotel brand preference ("XX series") and frequent flyer membership ("XX Gold Card") for soft scoring, prioritizing hotels within the XX series that offer points accumulation or membership benefits. The electronic device combines filtered and rated transportation and accommodation options to generate two travel plans: Plan 1 is high-speed rail G1234 (7:00-8:15, fare 300 yuan, travel time 1 hour 15 minutes) + a hotel (double room, 480 yuan / night, gold card members enjoy free breakfast), total price 1260 yuan; Plan 2 is flight CA1234 (7:30-8:20, fare 900 yuan, travel time 50 minutes, can accumulate airline miles) + a hotel (double room, 490 yuan / night, gold card members enjoy room upgrade), total price 1380 yuan. The electronic device displays the two plans in the AI ​​dialogue window according to the comprehensive rating from high to low, and each plan has a "Book" button. After comparing, Xiao Wang believes that the high-speed rail plan is more cost-effective, clicks the "Book" button for Plan 1, and enters the booking process to complete the order.

[0062] Scenario 2: User flow of the parent-child travel assistant in the context of family outings: User Xiao Li plans to take her 5-year-old daughter to a zoo this weekend. Xiao Li opens a travel service app and selects "Family Travel Assistant" from the AI ​​assistant. Since this is Xiao Li's first time using the Family Travel Assistant, the device cannot find the corresponding personalized travel parameters. Therefore, the AI ​​dialogue window displays the guidance message, "Welcome to the Family Travel Assistant! Please set your and your child's travel preferences so that we can generate a more suitable travel plan for you," and provides an entry point for adding child information and controls for family travel preferences. Xiao Li enters "5-year-old girl" in the child information entry point, selects "Animal Theme" through the theme preference control, "Relaxed and Leisurely" through the activity intensity preference control, checks "Stroller Rental" and "Mother and Baby Room" through the facility needs preference control, and selects "Moderate Crowds" through the safety needs preference control. The device stores these parameters as personalized family travel parameters in the server database using the user ID + "Family Travel Assistant" as the key. Then, Xiao Li enters "Recommended Zoo Itinerary" in the input control. The electronic device identifies the destination as the zoo and combines it with the saved personalized parameters for family travel (child is 5 years old, relaxed and casual, requires stroller rental and mother and baby room). It then filters out attractions unsuitable for young children (such as roller coasters and other thrill rides) from the various points of interest within the zoo, prioritizing routes that include stroller rental locations and mother and baby rooms. This generates a suitable travel route for young children, and the AI ​​dialogue window displays the descriptions of each point of interest and the booking portals for the corresponding experiences at each point of interest in sequence.

[0063] Scenario 3: Personalized services after hotel booking: After booking a hotel through a travel assistant, user Zhang's electronic device redirected to the hotel order details page. A floating AI component icon appeared in the lower right corner of this page. Clicking the icon triggered an AI dialogue window, displaying guidance information specific to the hotel order, including options such as "Reserve late check-out," "Reserve early check-in," "Please specify non-smoking room," "Please specify higher floor," and "Reserve extra bed." Zhang clicked the "Please specify non-smoking room" prompt and entered "Add another crib" in the input field. The electronic device identifies two sub-demands for the user's hotel booking: "non-smoking room preference" and "extra bed." It then uses an integrated AI outbound calling platform to call the hotel's front desk, merging the two sub-demands into a single call via TTS (Text-to-Speech) synthesis: "Hello, this is customer service from XX Travel Platform. My user, Mr. Zhang, has booked room type XX at your hotel from July 20th to July 22nd. He has the following two requests: First, he needs a non-smoking room; second, he needs an extra crib. Please confirm if this is possible." After the call, the AI ​​outbound calling platform converts the hotel's response back into text using ASR (Automatic Speech Recognition). The electronic device then displays the processing status information for each sub-demand in the AI ​​dialogue window: "Non-smoking room - Confirmed" and "Extra bed - Pending confirmation (hotel needs to confirm inventory)." It also displays a follow-up prompt: "If the extra bed request cannot be fulfilled, you can choose to change to 'extra pillow' or abandon this request."

[0064] Scenario 4: Cross-category related cancellation scenario: User Xiao Wang booked a flight from Hangzhou to Shanghai and a hotel in Shanghai through a travel assistant. Due to a cancelled meeting, Xiao Wang needed to cancel the trip. On the flight order details page, Xiao Wang clicked the AI ​​component and entered "Cancel this flight" in the AI ​​dialogue window. The electronic device recognized the cancellation request, executed the cancellation, and displayed the flight order information and cancellation status in the AI ​​dialogue window: "Cancellation in progress... Cancellation successful, estimated refund of 800 yuan, refund will arrive in 3-5 business days." Subsequently, the electronic device automatically used the flight's itinerary information (destination: Shanghai, travel date: July 20th) as the anchor point to initiate a related order query request to the server, finding Xiao Wang's hotel booking for July 20th in Shanghai. The electronic device continued to display guidance information in the AI ​​dialogue window: "Detected that you have booked a hotel in Shanghai for July 20th (double room, 1 night, 480 yuan), do you want to cancel it as well?" along with two options: "Cancel both" and "Do not cancel for now." Xiao Wang clicked "Cancel All Bookings Together," and the electronic device processed the hotel booking cancellation. The AI ​​dialogue window updated to display "Hotel cancellation in progress... Cancellation successful, estimated refund of 480 yuan," and "You have successfully cancelled all related orders, with a total refund of 1280 yuan." Xiao Wang completed the cancellation of both the flight and hotel bookings within the same AI dialogue window, without needing to exit the current page and switch to the hotel booking details page for a second cancellation.

[0065] Scenario 5: Intelligent Recommendation of Points of Interest (POIs) for Tourist Attractions User Xiao Zhao booked tickets for an animal-themed amusement park through a parent-child travel assistant. After entering the ticket order details page, he clicked on the AI ​​component, which displayed information such as "What's fun in the park?", "What activities are suitable for children?", and "Recommended tour routes." Xiao Zhao clicked on "What's fun in the park?", and the electronic device determined the user's need for information about the park's attractions. It then used the RAG (Retrieval Enhanced Generation) engine to retrieve data on points of interest for the park, combining this with personalized travel parameters linked to the parent-child travel assistant (child aged 5, relaxed / casual, facility preferences, etc.) to filter and sort the retrieved points of interest. Subsequently, the AI ​​dialogue window returned descriptions of each point of interest, indicating how well each point of interest met the user's preference constraints. For example, the AI ​​dialogue window displayed: "Panda Pavilion (1.2km from the entrance, accessible by shuttle bus, suitable for children, with stroller access)." —Experience program 'Panda Feeding VIP Experience' is available for booking; "Ocean Theater (800m away, within walking distance, suitable for children, with a mother and baby room)" —The 'Dolphin Interactive Show' experience is available for booking; "Hiking trail (3km away, requires walking, not suitable for young children, no stroller access)" "Not recommended." Each attraction comes with a detailed description and a booking link for the corresponding experience. Based on the recommendations, Xiao Zhao chose the Panda Pavilion and the Ocean Theater and completed the purchase of the experience through the booking link.

[0066] It should be noted that the embodiments of this application include, but are not limited to, the examples described above. It is understood that those skilled in the art can make further settings according to actual needs under the guidance of the ideas in the embodiments of this application, and this application does not impose any restrictions on this.

[0067] In this embodiment, the electronic device presents a travel service AI assistant containing at least two trip planning assistants in a graphical user interface. These different trip planning assistants are associated with personalized travel parameters for the same user in different travel scenarios. Then, in response to a trigger operation that selects a target AI assistant from at least one trip planning assistant, an AI dialogue window is displayed, and the target personalized travel parameters associated with that target AI assistant are automatically determined. Next, in response to an input operation to the AI ​​dialogue window, the trip planning requirement information corresponding to the input operation is displayed in the AI ​​dialogue window, along with at least one trip plan matching the trip planning requirement information and the target personalized travel parameters, and a booking entry corresponding to the trip plan. This achieves the goal of pre-binding the personalized travel parameters of the same user in different travel scenarios with different trip planning assistants and automatically loading the matching mechanism of the corresponding scenario parameters when the user triggers a selection. The system enables rapid access to personalized travel parameters and precise scenario-based matching. Furthermore, by associating personalized travel parameters for different travel scenarios with corresponding itinerary planning assistants, users can select appropriate personalized travel parameters based on their actual needs. This reduces the burden of repetitive operations when switching between scenarios. It also significantly reduces the number of human-computer interaction rounds required for users to supplement constraints through multiple dialogues, lowering the interactive input cost of itinerary planning. Moreover, the system-generated itinerary plans highly match the user's true intentions in the current travel scenario, avoiding the mixed features of generalized recommendations. This enhances the scenario awareness and service consistency of travel service applications when facing differentiated travel needs of the same user. Additionally, by directly providing the booking entry point corresponding to the itinerary plan in the AI ​​dialogue window, the system shortens the decision-making and operation chain from itinerary planning to product booking, improving the overall interactive efficiency of itinerary planning and the efficiency of user travel decisions.

[0068] To further enable those skilled in the art to better understand the technical solutions in the embodiments of this application, the following exemplary description is provided in conjunction with corresponding schematic diagrams: In some examples, the electronic device runs a travel service application and displays the corresponding application page through a graphical user interface, such as... Figure 2The diagram illustrates a schematic of the AI ​​operation interface provided in this embodiment. When a user opens the AI ​​operation interface, the interface can be viewed as follows: Figure 2 As shown, by default, the trip planning assistant can be collapsed. Upon detecting a corresponding expand operation, multiple trip planning assistants will be displayed in the AI ​​interface. These different assistants are associated with personalized travel parameters for the same user in different travel scenarios. The trip planning assistants can include the AI ​​assistant provided by the application by default, or user-defined AI assistants. For example, refer to... Figure 3 The illustration shows a schematic diagram of the assistant page provided in this application embodiment. When a user opens the assistant page, they can view the currently existing trip planning assistants and the assistant description information corresponding to each trip planning assistant. Assuming that the trip planning assistants include family travel assistants, business travel assistants, and self-driving assistants, the corresponding assistant description information can be "a one-stop travel planning partner specially designed for families with children, making it easy to handle food, accommodation, transportation, entertainment, and learning", "an efficient and compliant corporate travel expert, focusing on saving time, controlling costs, and seamless reimbursement", "focusing on family self-driving travel planning, providing route optimization, family-friendly stop recommendations, and implementing road condition services to make every journey safer, smoother, and more fun", etc., so that users can understand the strengths of different trip planning assistants.

[0069] Therefore, by configuring different trip planning assistants with their own assistant description information, users can quickly understand the travel scenarios and service directions that each assistant is good at without having to explore or memorize the functions of each assistant. They can learn about the applicable scenarios of multiple trip planning assistants at once in the graphical user interface, which reduces the cognitive cost and trial and error cost for users when choosing a target AI assistant, and enhances the discoverability and ease of use of the application.

[0070] Furthermore, such as Figure 3 As shown, the application can also provide corresponding assistant creation controls on the assistant page, allowing users to create corresponding AI assistants using these controls. The creation process can be as follows: Figure 4As shown, when a user clicks the assistant creation control, the application displays the assistant creation page. This page provides controls for setting an avatar, entering a name, defining a role, and setting personal preferences. Users can set the avatar's image, the name, the role description, and preferences. The "polishing" control allows users to input initial descriptions and then refine them, structuring the logic and format to more accurately and clearly express the role's positioning. Finally, once the user has completed the setup, they can create a corresponding custom assistant through the confirmation control, such as "Zhang San's Food Exploration Assistant." This assistant can be positioned as a travel food explorer who can find local delicacies for the user's destination and is familiar with the streets and alleys, authentic flavors, food culture, and dietary needs of families / special needs in each city. It can also combine the user's preferences, itinerary, and real-time information (such as business status, queue status, and average budget) to recommend restaurants that are truly worth visiting and provide practical tips. Furthermore, after a user creates a custom assistant, the application can display a corresponding AI dialogue window, showing the assistant's introduction and guidance information. For example, the assistant's introduction could be something like, "Hello! I'm your food consultant~ I see you like strong flavors, Sichuan cuisine, and are allergic to dairy products, with a budget of around 80 yuan per person. Great, this information is very useful! Next, tell me which city or specific area you want to go to, such as 'Yuhang District, Hangzhou,' and I can accurately recommend several authentic and safe Sichuan restaurants for you, guaranteeing no dairy, enough spiciness, and no bad experiences!" Guidance information could include "Top 10 Authentic Sichuan Restaurants in Yuhang District, Hangzhou," "Recommended Hidden Gems from Locals in Hangzhou," "Selected Restaurants in Hangzhou, From Local Cuisine to Private Kitchens," etc. Users can trigger the corresponding AI dialogue through the input controls provided in the AI ​​dialogue window or by selecting the appropriate guidance information.

[0071] Therefore, users can quickly create AI assistants tailored to their individual needs using natural language descriptions, without relying on predefined assistant templates or training the AI ​​to understand their preferences through multiple rounds of dialogue. Simultaneously, by providing polishing controls, the unstructured character descriptions input by users can be structured into accurate and clear character positioning descriptions, improving the AI ​​assistant's accuracy in understanding user intentions and thus establishing a stronger consistency in the user's persona with the AI ​​assistant.

[0072] Accordingly, assuming the user selected the travel assistant on the aforementioned AI operation page, then as follows: Figure 5 As shown in ①, upon initial activation of the travel assistant, the application can display relevant travel itinerary constraint guidance information and provide different types of standard travel controls in the AI ​​dialogue window. For example, the travel itinerary constraint guidance information could be: "Hello! I am your travel assistant. It only takes 10 seconds to complete the initial preference settings. In the future, you only need to say one sentence (such as 'Going to a meeting in Beijing's CBD next Wednesday'), and I will generate a flight and hotel plan that meets your company's travel standards, is the closest to your location, and perfectly matches your schedule. I will also handle your reimbursement after the trip." Correspondingly, the standard travel controls can include controls related to transportation budgets, hotel budgets, hotel brands, etc., allowing users to set corresponding personalized travel parameters. After setting, the application can refresh the page, such as... Figure 5 As shown in ②, the AI ​​dialogue window can display corresponding itinerary planning guidance information, parameter modification entry, and guidance prompts. For example, the itinerary planning guidance information could be something like, "Hello! I'm your travel assistant. I can generate a flight and hotel plan that meets your company's travel standards, is the closest to your destination, and perfectly suits your schedule in just one sentence. I'll also handle your reimbursement after the trip." Users can modify the corresponding personalized travel parameters through the parameter modification entry, or directly input controls or select the corresponding guidance prompts so that the travel assistant can plan the itinerary based on the pre-set personalized travel parameters and current itinerary planning needs, and display the corresponding itinerary plan in the AI ​​dialogue window. It should be noted that for... Figure 5 ② can also be a page displayed when the user triggers the travel assistant a second time, and this application does not restrict this.

[0073] Therefore, users only need to set preferences once when they first activate the travel assistant. Subsequent uses eliminate the need to repeatedly input travel standards, hotel brand preferences, frequent flyer memberships, and other information. The electronic device automatically loads the saved personalized travel parameters to match itinerary plans, significantly reducing the number of interactions where users need to manually add constraints in multiple rounds of dialogue, thus lowering the input cost of itinerary planning. Furthermore, the parameter modification entry allows users to quickly update their saved personalized travel parameters when travel standards change (such as when a company adjusts its travel standard policy), without needing to recreate the assistant.

[0074] Furthermore, such as Figure 6As shown, based on the user's travel planning needs and personalized travel parameters, the travel assistant can plan corresponding travel schemes for the user. These schemes can include transportation and accommodation options, and each scheme can recommend relevant travel services such as high-speed rail schedules, flights, and hotels, and provide corresponding booking portals so that users can quickly book the relevant travel services through these portals.

[0075] Therefore, because the matching process automatically incorporates the user's pre-set personalized travel parameters (such as transportation budget, accommodation budget, hotel brand preference, frequent flyer membership, etc.), the generated transportation and accommodation plans are highly aligned with the user's actual needs in terms of price, brand, and benefits, avoiding the mixed features of plans caused by generalized recommendations. Simultaneously, by providing a direct booking portal for travel services within the AI ​​dialogue window, the user's workflow from confirming the itinerary to completing the booking is shortened.

[0076] Accordingly, assuming the user selects the self-driving assistant, then as follows: Figure 7 As shown, upon initial activation of the self-driving assistant, the electronic device can display corresponding self-driving trip constraint guidance information and provide different types of self-driving preference controls in the AI ​​dialogue window. For example, the self-driving trip constraint guidance information could be: "Hi, I'm your self-driving assistant! The most beautiful scenery is always on the road. To plan the perfect self-driving route for you, in addition to avoiding congestion, we've arranged charging stations / gas stations, hotels, etc. Please tell me your driving habits." Correspondingly, self-driving preference controls can include vehicle type controls (such as "gasoline vehicle," "new energy vehicle"), driving intensity preference controls (such as "compact tour (more than 6 hours per day)," "leisure tour (3-4 hours per day)"), route style preference controls (such as "scenic tour (provincial / national highway)," "rushing (highway priority)"), and companion controls (such as "children," "elderly," "pets," "unrestricted"), etc. Users can set corresponding personalized parameters for their self-driving trip through these self-driving preference controls. After setting, the electronic device can refresh the page, such as... Figure 7As shown in ②, the AI ​​dialogue window can display corresponding itinerary planning guidance information, parameter modification entry, and guidance prompts. For example, the itinerary planning guidance information could be: "Hi, I'm your self-driving assistant! The most beautiful scenery is always on the road. The system has understood and remembered your self-driving preferences. Click on personal preferences to view and modify relevant information." Users can modify the corresponding personalized travel parameters through the parameter modification entry, or directly input their self-driving planning needs through the input controls (such as "a 3-day self-driving tour from Chengdu"), or select the corresponding guidance prompts (such as "a 2-day self-driving tour route in Chengdu suitable for spring"). This allows the self-driving assistant to plan the itinerary based on the pre-set personalized travel parameters and current itinerary planning needs, and display the corresponding itinerary plan in the AI ​​dialogue window. It should be noted that for... Figure 7 ② It can also be a page displayed by the user when the self-driving assistant is triggered a second time, and this application does not restrict this.

[0077] Therefore, the self-driving assistant provides scenario-specific preference controls (such as vehicle type, driving intensity, route style, and supply preferences) that differ from those of the business travel assistant. This ensures that the personalized travel parameters collected in self-driving scenarios are scenario-specific, avoiding the problem of inapplicable parameters when using general parameters (such as business travel standards) for self-driving trip planning. Furthermore, the personalized travel parameters set in this scenario are pre-bound to the self-driving assistant and persistently stored, so users do not need to reset these parameters when using the self-driving assistant multiple times at different times.

[0078] Furthermore, such as Figure 8 As shown, based on the user's travel planning needs and personalized parameters for self-driving travel, the self-driving assistant can plan corresponding itinerary schemes for the user. These itinerary schemes can include daily routes and attractions, as well as corresponding travel service products (such as hotels), so that users can quickly book the corresponding travel service products through this booking portal.

[0079] Therefore, the trip plans generated by the self-driving assistant comprehensively consider vehicle type (e.g., prioritizing routes with high charging station coverage when there is a preference for new energy vehicles), driving intensity preference (e.g., appropriately shortening the daily driving time and increasing rest stops along the way when there is a preference for leisure travel), route style preference (e.g., prioritizing provincial / national highways when there is a preference for sightseeing), and companion type (e.g., prioritizing stops with service facilities suitable for the corresponding groups when there are children or the elderly in the group). This makes the generated trip plans highly consistent with the user's real intentions in the self-driving travel scenario. At the same time, the daily driving arrangements in the trip plans are reasonable and the rest stops are well-equipped.

[0080] Accordingly, assuming the user selected the parent-child travel assistant on the aforementioned AI operation page, then as follows: Figure 9As shown in ①, upon initial activation of the parent-child travel assistant, the electronic device can display relevant parent-child travel constraint guidance information and provide an entry point for adding children's information in the AI ​​dialogue window. For example, the parent-child travel constraint guidance information could be something like, "Hi, I'm your little helper for traveling with your child. I can help you filter safe and age-appropriate hotels with one click, intelligently recommend travel plans that match your family's preferences, and automatically avoid travel risks. To give you more thoughtful advice, let's confirm your child's information first. Click on 'Frequently Used Travelers' or manually enter @ to enter it." Correspondingly, the electronic device can also display guidance prompts related to parent-child travel in the AI ​​dialogue window, such as "8 fairytale-like destinations that 4-year-old girls love" or "Which hotels in Hangzhou offer children's beds / safety rails and other supplies?", allowing users to initiate corresponding itinerary planning or attraction recommendation requests with one click.

[0081] When activating the Parent-Child Travel Assistant for the first time, users can first enter the relevant child information through the child information addition portal; while when activating the Parent-Child Assistant for the second time, users can... Figure 9 The second option allows users to modify parameters such as child information. During the process of adding or modifying child information, users can click... Figure 9 The app allows users to add children's information, such as identity information and contact details, under the "Frequent Travelers" section. After adding this information, the app can display corresponding preference guidance information and parent-child travel preference controls, such as... Figure 10 As shown, the AI ​​dialogue window displays guidance information such as "Great, your child's information has been recorded. Next, select the themes you are interested in and your preferred modes of transportation, and I will recommend the most suitable travel plans for you." It also includes parent-child travel preference controls related to theme preferences and modes of transportation. After users select these controls to set corresponding personalized parameters for parent-child travel, the AI ​​dialogue window displays corresponding feedback information, such as "The system has understood and remembered your travel preferences. Click on your personal preferences to view and modify relevant information. Want to start a parent-child trip now? Just tell me your destination." It also provides corresponding guidance prompts, such as "Theme parks that kids love to visit in Hangzhou," "Recommended natural history museums suitable for parent-child interactive experiences," and "Weekend destinations suitable for children aged 3+ with strollers," etc.

[0082] Accordingly, based on the personalized parameters for family travel set by the user and combined with the user's current travel planning needs, the family travel assistant can plan corresponding travel schemes for the user. The planning process can be referenced from the process of other AI assistants, and will not be elaborated here.

[0083] Therefore, the parent-child travel assistant, through the entry point for adding children's information and the parent-child travel preference control, has achieved refined preference collection for parent-child travel scenarios (such as children's age, theme preferences, activity intensity, facility needs, etc.). Furthermore, electronic devices can automatically adapt differentiated recommendation strategies according to children's age (for example, prioritizing interest points that focus on observation and interaction for preschool children, and prioritizing interest points with educational significance for school-age children). This ensures that the generated itinerary plan fully considers the special needs of younger children, avoiding problems such as mismatched attraction selection with children's age, itinerary intensity exceeding children's physical capabilities, and lack of supporting facilities found in general itinerary planning schemes, thus improving the safety and comfort of parent-child travel scenarios.

[0084] Furthermore, during the process of a user conversing with any of the aforementioned AI assistants to plan their trip, if the user wants to learn more about the travel services or attractions involved in the itinerary provided by the AI ​​assistant, they can select the corresponding attraction card or name, and the application can redirect to the corresponding details page, such as... Figure 11 As shown, the details page displays corresponding AI components, which users can use to trigger AI dialogues. These AI dialogues can be similar to the previous interaction with the AI ​​assistant, where users can input their needs to receive recommendations from the AI ​​assistant. For example, if a user inputs "Where are the best places to visit around West Lake?", the AI ​​assistant can provide information about West Lake and offer corresponding attraction ticket cards in its response, allowing the user to purchase tickets.

[0085] Therefore, users do not need to leave the current page to search for attraction guides or Q&A while browsing the attraction details page. They can directly obtain targeted attraction introductions and recommendations on the current details page through the AI ​​component. At the same time, the AI ​​reply content directly includes the attraction ticket booking entry, allowing users to complete the entire operation from information acquisition to product booking within one page, avoiding switching back and forth between multiple pages.

[0086] Furthermore, assuming a user places an order for a hotel through the booking portal provided in the itinerary, the application can redirect them to the corresponding order details page, such as... Figure 12In addition to displaying hotel order information, the order details page also provides a corresponding AI component 1210. This AI component 1210 can include guidance prompts associated with the order and input controls. Users can select a guidance prompt or enter their order service requests in the input controls. For example, if a user wants to book a late check-out, after entering the corresponding order service request, the application can switch the AI ​​component to the corresponding AI dialogue window, displaying the order service request, such as "Help me book a late check-out," and corresponding feedback information, such as "Okay, I've communicated with the hotel in advance for you~ Late check-out request received, what time would you like to check out later?" Based on user dialogue, after clarifying the user's order service needs, the electronic device can make a phone call to the hotel front desk or send a message to the hotel to communicate the request. The communication results are displayed in the AI ​​dialogue window, for example, "A late check-out request has been submitted for you. We are contacting the hotel to negotiate. We will notify you of the results as soon as possible." If there is a processing result, the corresponding processing result can be displayed, such as "Hello, your request for check-in at Beijing XXXX Hotel on July 15, 2026 has been communicated with the hotel. The hotel's feedback is as follows: • Late check-out: We were unable to contact the hotel this time. We suggest you try again later or contact customer service. We wish you a pleasant stay."

[0087] Therefore, users no longer need to exit the order details page after completing a hotel booking to find the hotel's contact information or call the front desk, nor do they need to submit different requests (such as late check-out, non-smoking rooms, extra beds, etc.) on multiple pages. They can initiate multiple personalized post-booking service requests in a single AI dialogue window through the AI ​​component on the order details page. The electronic device then proactively contacts the hotel via an outbound call platform to communicate the requests and confirm the results. Simultaneously, the processing results for each sub-request are presented independently in the AI ​​dialogue window. If a request cannot be met, the user can adjust that sub-request individually (e.g., changing "late check-out until 3 PM" to "late check-out until 2 PM") without resubmitting all requests, thus improving the efficiency of post-booking service processing.

[0088] In addition to the aforementioned method of users placing orders for travel services directly through the booking portal, users can also input their specific purchase requests in the AI ​​dialogue window, such as... Figure 13As shown in the diagram, the user enters their purchase request information in the input controls of the aforementioned AI dialogue window, such as "Buy me tickets for Shanghai XX tomorrow, 3 adults and 2 children." The electronic device responds to the user's purchase request information in the AI ​​dialogue window, displays the information in the dialogue content area, and performs natural language understanding on the user's input text to extract the traveler information. For example, for the user's input of "3 adults and 2 children," the traveler information includes 3 adults and 2 children. After extracting the traveler information, the electronic device uses this information as a query condition to initiate a product inventory query request to the travel service platform's server to check if there is a single travel service product that perfectly matches the traveler information. If the server's product database contains a product with the exact traveler configuration of "3 adults + 2 children," the electronic device selects this single product as a candidate travel service product and directly displays its information and booking entry in the AI ​​dialogue window. If the electronic device cannot find a single travel service product that perfectly matches the traveler information, it retrieves the passenger capacity configuration information for each travel service product in the product library. Using the traveler information as the target constraint, it filters at least two candidate travel service products from the library, ensuring that these candidate products complement each other in terms of passenger capacity configuration and, when combined, match the overall traveler information. For example, for a traveler requirement of "3 adults and 2 children," if the product library does not contain a single product for "3 adults and 2 children," but contains products for "2 adults and 2 children" and "1 adult," the electronic device will use these two products as candidate travel service products. The electronic device displays feedback in the AI ​​dialogue window such as "Tomorrow XX will have 3 adults and 2 children, let me check ticket availability for you..." and displays the product information for each candidate travel service product and its corresponding booking entry. For example, ... Figure 13As shown, the electronic device displays product cards for "Ticket 1" (corresponding to 1 adult ticket) and "Ticket 2" (corresponding to a family ticket for 2 adults and 2 children) in an AI dialogue window. Each product card includes information such as the product name, usage date, applicable group, and price. Each product card also provides an entry point for adding traveler information, such as a "Please Add Traveler" control, for adding traveler identity information for the corresponding product. Simultaneously, the electronic device also displays the total price information (e.g., "¥2222") and a "Confirm Booking" control in the AI ​​dialogue window, allowing users to confirm the total price of the combined products and complete the order for all items at once. Users can add traveler information for each product card separately in the same AI dialogue window and complete the booking for all candidate travel services at once using the "Confirm Booking" control, without needing to be redirected to a separate booking page to fill in traveler information and confirm payment for each item individually after confirming the product combination. Therefore, when the number of travelers required by a user cannot be met by a single product in the product library, the electronic device can automatically perform intelligent matching of product combinations and present the combined product and traveler information together in the AI ​​dialogue window. This eliminates the need for users to manually perform multiple rounds of searches outside the dialogue window, check the inventory of different products separately, and manually compare whether the number of travelers matches. It also eliminates the need to fill in traveler information and complete payments on multiple pages separately, reducing the interaction cost and cognitive burden for users when dealing with the needs of booking tickets for multiple people, and improving the booking efficiency in scenarios with complex traveler requirements.

[0089] In addition, after a user completes an order, a one-click invoice function can be provided, or if a user's itinerary changes and they need to cancel the travel service, they can cancel the order with one click by talking to the AI ​​assistant. The relevant process can be referred to the description in the aforementioned embodiments, and will not be repeated here.

[0090] In summary, by pre-binding personalized travel parameters of the same user in different travel scenarios with different itinerary planning assistants, and automatically loading the matching mechanism of corresponding scenario parameters when the user triggers a selection, the system achieves rapid retrieval and accurate scenario-based matching of target personalized travel parameters. This not only significantly reduces the number of human-computer interaction rounds for users to supplement constraints through multiple rounds of dialogue, thus lowering the interactive input cost of itinerary planning, but also ensures that the itinerary plans generated by the system highly match the user's true intentions in the current travel scenario. This avoids the mixed features of plans caused by generalized recommendations, enhances the scenario perception capability and service consistency of travel service applications when facing differentiated travel needs of the same user, and shortens the decision-making and operation link from itinerary planning to product booking by providing the booking entry directly in the AI ​​dialogue window, thereby improving the overall interactive efficiency of itinerary planning and the user's travel decision-making efficiency.

[0091] It should be noted that, for the sake of simplicity, the method embodiments are all described as a series of actions. However, those skilled in the art should understand that the embodiments of this application are not limited to the described order of actions, because according to the embodiments of this application, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of this application.

[0092] In addition, this application also provides an electronic device, including: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the various processes of the above-described AI-based trip planning method embodiments and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0093] This application also provides a readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the various processes of the above-described AI-based route planning method embodiments and achieves the same technical effects. To avoid repetition, it will not be described again here. The readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0094] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0095] Those skilled in the art will understand that embodiments of this application can be provided as methods, apparatus, or computer program products. Therefore, embodiments of this application can take the form of entirely hardware embodiments, entirely software embodiments, or embodiments combining software and hardware aspects. Furthermore, embodiments of this application can take the form of computer program products implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, EEPROM, Flash, and eMMC, etc.) containing computer-usable program code.

[0096] This application describes embodiments with reference to flowchart illustrations and / or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0097] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing terminal device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0098] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal equipment, causing a series of operational steps to be performed on the computer or other programmable terminal equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable terminal equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0099] Although preferred embodiments of the present application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the embodiments of the present application.

[0100] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal device. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes said element.

[0101] The above provides a detailed description of an AI-based trip planning method provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the method and its core ideas. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. An AI-based itinerary planning method, characterized in that, A graphical user interface is provided via an electronic device, the content presented in the graphical user interface including a travel service AI assistant provided by a travel service application. The travel service AI assistant includes at least two trip planning assistants, each associated with personalized travel parameters for different travel scenarios of the same user. The method includes: In response to a triggering operation that selects a target AI assistant from at least one of the trip planning assistants, an AI dialogue window is displayed, and target personalized travel parameters associated with the target AI assistant are determined; In response to an input operation to the AI ​​dialogue window, the AI ​​dialogue window displays the itinerary planning requirement information corresponding to the input operation, and displays at least one itinerary plan that matches the itinerary planning requirement information and the target personalized travel parameters, as well as the booking entry corresponding to the itinerary plan.

2. The method according to claim 1, characterized in that, The trip planning assistant includes a business travel assistant, a self-driving assistant, and a family travel assistant. In response to a trigger operation that selects a target AI assistant from at least one of the trip planning assistants, an AI dialogue window is displayed, and target personalized travel parameters associated with the target AI assistant are determined, including: In response to a triggered action on the target AI assistant, an AI dialogue window is displayed; If no target personalized travel parameters are detected, and the target AI assistant is a travel assistant, then travel itinerary constraint guidance information and different types of standard travel controls are displayed in the AI ​​dialogue window. The standard travel controls include at least one of the following: transportation budget, accommodation budget, hotel brand, frequent flyer membership, travel compliance policy, and special accommodation needs information. If the target AI assistant is a self-driving assistant, then the self-driving trip constraint guidance information and different types of self-driving preference controls are displayed in the AI ​​dialogue window. The self-driving preference controls include at least one of the following: vehicle type, driving intensity preference, route style preference, companion type, waypoint preference control, and supply preference control. If the target AI assistant is a parent-child travel assistant, then parent-child travel constraint guidance information will be displayed in the AI ​​dialogue window, as well as an entry point for adding children's information and / or parent-child travel preference controls. The parent-child travel preference controls include at least one of theme preference, travel mode preference, activity intensity preference, facility need preference, and safety need preference. When the target AI assistant is detected as associated with a personalized travel parameter, if the target AI assistant is a business travel assistant, the personalized travel parameters associated with the business travel assistant and the parameter modification entry will be displayed in the AI ​​dialogue window; if the target AI assistant is a self-driving assistant, the personalized self-driving parameters associated with the self-driving assistant and the parameter modification entry will be displayed in the AI ​​dialogue window; if the target AI assistant is a parent-child travel assistant, the personalized parent-child travel parameters associated with the parent-child travel assistant and the parameter modification entry will be displayed in the AI ​​dialogue window.

3. The method according to claim 1 or 2, characterized in that, Before displaying the AI ​​dialogue window in response to a triggering operation that selects a target AI assistant from at least one of the trip planning assistants, the method further includes: In response to a custom operation by the AI ​​assistant, a custom window is displayed, which includes at least a character description control and a personal preference control. In response to an input operation to the character description control, character description information is obtained; In response to an input operation to the personal preference control, personal preference information is obtained; In response to the assistant creation operation, a corresponding target AI assistant is created based on the role description information and the personal preference information.

4. The method according to claim 1, characterized in that, The itinerary plan includes at least one travel service product, and the booking portal is the booking portal for the travel service product. After displaying at least one itinerary plan that matches the itinerary planning requirements and the target personalized travel parameters, and the booking portal corresponding to the itinerary plan, the method further includes: In response to the trigger operation of the booking entry, the order for the target travel service product is completed, and the order details page corresponding to the target travel service product is displayed, as well as the corresponding AI component is displayed; In response to a trigger operation on the AI ​​component, the order service demand information for the target travel service product is determined, and an AI dialogue window is displayed. The content displayed in the AI ​​dialogue window includes at least: travel service information of at least one travel service item that matches the order service demand information, and processing status information corresponding to the travel service item.

5. The method according to claim 4, characterized in that, The AI ​​component includes at least input controls and guidance prompts. The step of determining the order service demand information for the target travel service product in response to a trigger operation on the AI ​​component includes: In response to a trigger operation on the input control or any of the guidance prompts, determine the order service demand information for the target travel service product associated with the order details page, wherein the guidance prompts are information generated based on the target travel service product; The order service demand information includes at least one of the following demand information: If the target travel service product is a hotel product, the order service request information includes at least one of the following: reservation for late check-out, reservation for early check-in, preference for non-smoking rooms, preference for high floors, preference for quiet rooms, preference for specific bed types, reservation for extra beds, reservation for extra breakfast, reservation for airport transfer service, and supplementary personalized request information. The order information is the hotel name and room type information of the hotel product, and the processing status information is the processing result corresponding to the order service request information. If the target travel service product is a transportation product, the order service request information includes at least one of cancellation, rescheduling, seat selection, adding baggage allowance, and issuing an itinerary. The order information is the departure point, destination, and travel time information of the transportation product. The processing status information is cancellation progress information, rescheduling progress information, or itinerary issuance progress information. If the target travel service product is an attraction ticket, the order service request information includes at least one of cancellation, rescheduling, and invoice issuance. The order information is the attraction name and entry date information of the attraction ticket. The processing status information is cancellation progress information, rescheduling progress information, or invoice issuance progress information.

6. The method according to claim 4 or 5, characterized in that, After determining the order service demand information for the target travel service product and displaying the AI ​​dialogue window in response to a trigger operation on the AI ​​component, the method further includes: If the target travel service product is a hotel product, then a voice call is initiated to the merchant user device to which the target travel service product belongs, or the order service request information is sent to the merchant user account to which the target travel service product belongs, and the processing result returned by the merchant user device or the merchant user account is obtained. If the target travel service product is a transportation product or an attraction ticket product, then the order service request information is sent to the service system corresponding to the transportation product or the attraction ticket product, and the processing result of the order service request information returned by the service system is received.

7. The method according to claim 1, characterized in that, The itinerary plan includes at least one travel service product. After displaying at least one itinerary plan that matches the itinerary planning needs information and the target personalized travel parameters, and the booking portal corresponding to the itinerary plan, the method further includes: In response to the purchase request information entered by the user in the AI ​​dialogue window, the number of travelers information is extracted from the purchase request information; If a travel service product matching the traveler information is found, the AI ​​dialogue window will display the candidate travel service product matching the traveler information and the booking portal. If no single travel service product matching the number of travelers is found in the travel service application, at least two candidate travel service products matching the number of travelers are selected, wherein the quantity configuration information and / or traveler type configuration information of the candidate travel service products match the number of travelers. The AI ​​dialogue window displays product information and a booking portal for the candidate travel services.

8. The method according to claim 1, characterized in that, The itinerary plan includes at least one attraction product. After displaying at least one itinerary plan that matches the itinerary planning needs information and the target personalized travel parameters, and the booking portal corresponding to the itinerary plan, the method further includes: In response to a triggered operation targeting a product at a specific attraction, the system redirects to display the product details page for that product and shows the corresponding AI component on the product details page. In response to a trigger operation on the AI ​​component, the AI ​​component is displayed in an expanded component form, and at least one scenic spot introduction guide information associated with the target scenic spot product and an input control are displayed in the AI ​​component; In response to an input operation to the input control or a trigger operation to any of the attraction introduction guidance information, determine the attraction introduction demand information for the target attraction product, display an AI dialogue window, and display attraction description information for the attraction introduction demand information in the AI ​​dialogue window. The attraction description information includes description information of at least one point of interest and booking entry points for travel service products corresponding to each point of interest.

9. The method according to claim 5, characterized in that, After completing the order for the target travel service product, the method further includes: The system identifies cancellation requests for the target travel service and displays an AI dialogue window. The AI ​​dialogue window displays order-related information and the expected refund amount for orders matching the travel service products.

10. The method according to claim 9, characterized in that, After displaying the order information and expected refund amount matching the travel service product order in the AI ​​dialogue window, the method further includes: In response to a cancellation request for the target travel service product, if no associated order is found that is associated with the target travel service product for which a cancellation has been initiated, the corresponding query feedback content is displayed in the AI ​​dialogue window. The query feedback content includes at least query feedback information and an order viewing entry. If a related order is found that is associated with the target travel service product for which a cancellation has been initiated, the AI ​​dialogue window will display order cancellation guidance content. The order cancellation guidance content includes at least the related order cancellation guidance information and the related order cancellation entry. The related order cancellation guidance information is used to inform the user that if there is a need to cancel the related orders of the travel service product order for which a cancellation request has been initiated, the user can enter the corresponding request information through the AI ​​dialogue.

11. The method according to claim 5, characterized in that, After completing the order for the target travel service product, the method further includes: In response to detecting that the user has completed the fulfillment of the target travel service product, invoicing guidance content for the target travel service product is displayed in the AI ​​dialogue window; The invoicing guidance includes: if invoicing information is found, the order-related information of the travel service product and the corresponding invoice information will be displayed in the AI ​​dialogue window; if invoicing information is not found, information entry guidance will be displayed in the AI ​​dialogue window to guide the user to supplement the invoicing information.

12. An electronic device, characterized in that, It includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; The memory is used to store computer programs; When the processor executes a program stored in the memory, it implements the method as described in any one of claims 1-11.

13. A computer-readable storage medium having instructions stored thereon that, when executed by one or more processors, cause the processors to perform the method as described in any one of claims 1-11.