Display method of travel planning content and computer program product

Abstract

One or more embodiments of the present specification provide a travel plan content display method and a computer program product. The method is applied to a travel service platform, and comprises the following steps: determining at least one virtual human device matched with a travel planning request according to the received travel planning request; calling a travel planning model to carry out travel planning processing, and determining a thinking process of each virtual person in the travel planning processing; and respectively displaying the thinking process corresponding to each virtual person through a first client of the travel service platform.

Description

Technical Field

[0001] One or more embodiments of this specification relate to the field of travel services, and more particularly, to a method for displaying travel planning content and a computer program product. Background Art

[0002] Currently, travel service platforms can provide users with travel planning services using AI (Artificial Intelligence) models. Furthermore, to fully understand the specific travel planning process, the platform will display the AI ​​model's Chain of Thought (CoT) for users to review and trace.

[0003] In related technologies, platforms typically directly display large sections of text output by AI models to users. Consequently, users are required to read and understand the text line by line to understand the detailed process of travel planning. This approach results in a monotonous output format and is difficult to understand. This makes it difficult for users to fully, accurately, and intuitively understand the detailed process of travel planning. Consequently, the authenticity and credibility of travel planning results are low for users, and this urgently needs improvement. Summary of the Invention

[0004] In view of this, one or more embodiments of this specification provide the following technical solutions:

[0005] According to a first aspect of one or more embodiments of this specification, a method for displaying travel planning content is proposed, which is applied to a travel service platform. The method includes:

[0006] Determining, based on the received travel planning request, at least one virtual persona matching the travel planning request;

[0007] Invoking the travel planning model to perform travel planning processing, and determining the thinking process of each virtual person in the travel planning processing;

[0008] The thinking process corresponding to each virtual persona is displayed respectively through the first client of the travel service platform.

[0009] According to a second aspect of one or more embodiments of this specification, a method for replaying travel planning content is proposed, which is applied to a travel service platform. The method includes:

[0010] Based on the received travel planning request, the travel planning model is called to perform travel planning processing;

[0011] Displaying the thought process of the travel planning process and the corresponding travel planning results on the first client of the travel service platform, and recording the thought process and travel planning results as corresponding travel planning playback content;

[0012] A trigger control corresponding to at least one recorded travel planning playback content is displayed on at least one second client of the travel service platform, and when the trigger control corresponding to any travel planning playback content is selected, any travel planning playback content is played back on the second client.

[0013] According to a third aspect of one or more embodiments of this specification, an electronic device is proposed, comprising: a processor; a memory for storing processor-executable instructions; wherein the processor implements the steps of the method described in the first aspect or the second aspect by running the executable instructions.

[0014] According to a fourth aspect of one or more embodiments of this specification, a computer-readable storage medium is provided, on which computer instructions are stored. When the instructions are executed by a processor, the steps of the method described in the first aspect or the second aspect are implemented.

[0015] According to a fifth aspect of one or more embodiments of this specification, a computer program product is proposed, comprising a computer program / instruction, which, when executed by a processor, implements the steps of the method described in the first aspect or the second aspect.

[0016] From the first aspect mentioned above, it can be seen that this specification proposes a new method for displaying travel planning content, that is, after receiving a travel planning request initiated by a user, the travel service platform first determines at least one virtual person that matches the travel planning request, then calls the travel planning model to perform travel planning processing, and separately determines the thinking process of each virtual person in the travel planning processing; then, through the first client of the travel service platform, the thinking process corresponding to each virtual person is displayed separately for the user to view.

[0017] As can be seen, this solution, through at least one virtual persona and its thought process, splits the complete processing process for a travel planning request into professional processes completed separately by professionals in various fields (such as transportation, hotels, routes, strategies, etc.); further, by displaying the corresponding thought processes of each virtual persona through the first client, it creates for users the feeling of using the product, where a professional travel planning team composed of professionals in various fields collects data and executes plans for users. It can be understood that this solution can ensure the accuracy and reliability of travel planning results based on the travel planning model; on this basis, by displaying the thought processes of each virtual persona separately, it not only enriches the display form of the thinking process, enhances the fun of use and significantly reduces the difficulty of understanding for users, making it easier for them to fully, accurately and intuitively understand the specific process of the travel planning model generating travel planning results; it can also effectively improve the credibility of the travel planning process and enhance users' trust in the travel planning results. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a schematic diagram of the architecture of a travel service platform provided by an exemplary embodiment.

[0019] Figure 2 It is a flowchart of a method for displaying travel planning content provided by an exemplary embodiment.

[0020] Figure 3 It is a schematic diagram of an entry interface of a travel planning service provided by an exemplary embodiment.

[0021] Figure 4 This is a schematic diagram of an input interface for a travel planning request provided by an exemplary embodiment.

[0022] Figure 5 It is a schematic diagram of an input effect of a travel planning request provided by an exemplary embodiment.

[0023] Figure 6 This is a schematic diagram of a task splitting process of a main character provided by an exemplary embodiment.

[0024] Figures 7 to 11 It is a schematic diagram of the branching thinking process of various travel expert personas provided by multiple exemplary embodiments.

[0025] Figure 12 This is a schematic diagram of a branching thinking process of a budget analysis personality provided by an exemplary embodiment.

[0026] Figure 13 This is a schematic diagram of a task aggregation process of a main character provided by an exemplary embodiment.

[0027] Figure 14 It is a schematic diagram of a travel planning result provided by an exemplary embodiment.

[0028] Figure 15 It is a schematic diagram of the unfolding effect of a thinking process provided by an exemplary embodiment.

[0029] Figure 16 This is a schematic diagram of a parameter value adjustment interface provided by an exemplary embodiment.

[0030] Figure 17 This is a schematic diagram of a content sharing interface provided by an exemplary embodiment.

[0031] Figure 18 It is a flowchart of a method for replaying travel planning content provided by an exemplary embodiment.

[0032] Figure 19 It is a schematic diagram of a playback content plaza interface provided by an exemplary embodiment.

[0033] Figures 20 to 24 It is a schematic diagram of a playback process of a travel planning playback content provided by multiple exemplary embodiments.

[0034] Figures 25 to 27 FIG. 4 is a schematic diagram of a playback process of another travel planning playback content provided by multiple exemplary embodiments.

[0035] Figure 28 It is a structural diagram of a device provided by an exemplary embodiment.

[0036] Figure 29 It is a block diagram of a device for displaying travel planning content provided by an exemplary embodiment.

[0037] Figure 30 It is a block diagram of a playback device for travel planning content provided by an exemplary embodiment. DETAILED DESCRIPTION

[0038] Currently, travel service platforms can provide users with travel planning services using AI big models (hereinafter referred to as travel planning models). Furthermore, to fully understand the specific travel planning process and how AI arrives at its results, the platform displays the big model's thought chain for users to review and trace.

[0039] In related technologies, platforms typically directly display large text output from large models to users. Consequently, users are required to read and understand the text line by line to understand the detailed process of travel planning. This approach results in a monotonous output format and is difficult to understand. This makes it difficult for users to fully, accurately, and intuitively understand the details of the travel planning process. Consequently, the authenticity and credibility of travel planning results are low, leading to generally low user trust and acceptance of travel planning results, which urgently needs improvement.

[0040] To address the aforementioned technical issues, this specification proposes a solution for displaying travel planning content. By showing users the thought process of a virtual persona, this solution creates the feeling of a professional participating in a specific travel plan, thereby enhancing users' trust in the travel planning results output by the model and improving their acceptance of the results. This solution is described below with reference to the accompanying figures and related embodiments.

[0041] Figure 1 This is a schematic diagram of the architecture of a travel service platform provided by an exemplary embodiment. Figure 1 As shown, the platform may include a server 11 and several electronic devices, such as mobile phone 14, mobile phone 15, mobile phone 16, etc.

[0042] During operation, the server 11 can run a program on the server side of a certain application to implement the relevant functions of the application. For example, when the server 11 runs a program for a travel planning service, it can be implemented as the server end of the service (hereinafter referred to as the planning server end). In addition to the server 11 and several electronic devices, the travel service platform can also connect to or include servers that implement specific functions, such as server 12 and / or server 13, etc., wherein the server 12 can run a program for a travel service to implement a corresponding travel server end for providing travel services to the planning server end; and the server 13 can run a program for a model management service to implement a model server end, such as locally deploying and running a trained travel planning model, and providing corresponding reasoning services to the planning server end through the calling interface of the model. Of course, both the aforementioned planning server end and the travel server end can maintain corresponding travel data so that the planning server end can query and obtain it. The aforementioned travel planning model can also be run locally on the server 11 (in this case, the travel service platform may not include the aforementioned server 13) so that the planning server end can quickly call it, which will not be repeated here.

[0043] Any of the servers 11, 12 and 13 may be a physical server including an independent host, or the server 11 may be a virtual server hosted by a host cluster.

[0044] Any electronic device can run a client-side program of a certain application to implement the relevant functions of the application. For example, when the electronic device runs a program for a travel planning service, it can be implemented as a planning client for the service (such as mobile phone 14 is implemented as a first client, mobile phone 15 and mobile phone 16 are respectively implemented as second clients, etc.). Among them, the application of any of the above clients can be started and run on the corresponding electronic device. Exemplarily, the program can be a native application installed on the electronic device, or it can also be a small program, quick application or other similar forms. Of course, when using web technologies such as HTML5 or similar, the relevant functions can be implemented through the page displayed by the browser. The browser here can be an independent browser application or a browser module embedded in certain applications. In addition, mobile phones are only some types of electronic devices that users can use. In fact, users can obviously also use electronic devices such as PCs (Personal Computers), tablet devices, laptops, PDAs (Personal Digital Assistants), wearable devices (such as smart glasses, smart watches, etc.), etc., and one or more embodiments of this specification are not limited to this.

[0045] It can be understood that, from a hardware perspective, the travel service platform includes terminal devices (i.e., the aforementioned electronic devices, such as mobile phones 14 to 16, etc.) and servers (such as the aforementioned server 11, etc.); and from a software perspective, the platform includes planning clients (such as the client-side program of the travel planning service running in each mobile phone, i.e., the first client running in mobile phone 14, the second client running in mobile phone 15 and mobile phone 16, etc.) and shopping service terminals (such as the server-side program of the online shopping service running in server 11).

[0046] The users described in this specification (such as the first user and the second user described below) can be either legitimate users who have completed registration on the travel service platform or unregistered temporary users (or "visitors" of the platform), and the embodiments of this specification are not limited to this. Any user can access the platform through a corresponding client (such as the first client and the second client described below).

[0047] Furthermore, the network 10 for interaction between electronic devices such as mobile phones 14 and server 11 can be implemented using either a wired or wireless network, depending on the communication methods supported by the electronic devices. This specification does not limit this. For example, if mobile phones 14 typically only support wireless communication, wireless networks can be used for communication. Similarly, PCs can communicate with server 11 using either a wired or wireless method, which will not be discussed further.

[0048] In one embodiment, the first user can initiate a travel planning request to the server 11 (planning server) through the mobile phone 14 (planning client), and the latter can determine at least one virtual persona that matches the request, and query the corresponding travel data from the local or server 12 (travel server), and then call the travel planning model running in the server 13 (model server) to perform travel planning processing; and, respectively determine the thinking process of each virtual persona in the travel planning processing and send it to the mobile phone 14, so that the mobile phone 14 can display the thinking process corresponding to each virtual persona.

[0049] In another embodiment, after receiving a travel planning request initiated by the first user through the mobile phone 14, the server 11 can call the travel planning model to perform travel planning processing, and then display the thinking process and corresponding travel planning results of the travel planning processing on the mobile phone 14, and record the thinking process and travel planning results as the corresponding travel planning playback content (i.e., corresponding to the aforementioned travel planning request). Furthermore, if a sharing instruction for the aforementioned travel planning playback content is received from the first user, the content can be shared (published) within a preset range. Accordingly, for at least one travel planning playback content recorded by the server 11, the corresponding trigger control can be displayed on the mobile phone 15 used by the second user; and when the trigger control corresponding to any playback content is selected, the mobile phone 15 can locally replay the playback content (obtained from the server 11) to the second user, so that the user can know the specific process of the travel planning model generating the corresponding travel planning result by viewing the playback process.

[0050] Figure 2 This is a flow chart of a method for displaying travel planning content provided by an exemplary embodiment, which can be applied to the aforementioned travel service platform. Figure 2 As shown, the method includes the following steps 202 to 206.

[0051] Step 202: Determine, based on the received travel planning request, at least one virtual persona that matches the travel planning request.

[0052] The travel service platform described in this specification is used to provide travel planning services to users. Accordingly, users with travel planning needs can initiate corresponding travel planning requests to the travel service platform according to their own needs, and the platform will respond to the request and perform corresponding travel planning processing (i.e., call the travel planning model to perform inference and generate corresponding travel planning results). Exemplarily, a user (hereinafter referred to as the first user) can initiate a travel planning request to the planning server through the planning client (hereinafter referred to as the first client) used by the user (such as entering a travel-related question of interest in the conversation interface displayed by the planning client); accordingly, the planning server can respond to the received travel planning request and provide the corresponding travel planning service to the first user.

[0053] like Figure 3 As shown, the entry interface of the travel planning service includes a conversation box 301. After the first user triggers the conversation box 301, the interface switches to Figure 4 The input interface shown in FIG. 4 shows a first user inputting text or voice through the dialog box 401 of the input interface and clicking the “Send” button to initiate a travel planning request. Figure 5The conversation interface shown above displays a question 501 initiated by a first user (i.e., "What should I do in S2 during the May Day holiday?"). This question is the text entered by the first user in conversation box 401, or the text obtained by the first client through voice recognition of the first user's voice input in conversation box 401. As can be seen from the question, the first user is currently in city S1 and wants to consult the travel service platform about travel plans to city S2 during the May Day holiday.

[0054] Figure 1 The entry interface shown can also display quick entry controls 302 of commonly used planning functions to the first user, such as "Help me plan my itinerary", "Discover destinations", "Recommend special hotels", etc. After the first user clicks on the quick entry controls of any of the planning functions, he can specify key information for the planning function in the switched interface to initiate a travel planning request for the planning function. For example, after the first user clicks on the "Discover destinations" control, he can specify key information such as travel time, attraction preferences, budget, etc. in the switched interface, and the travel service platform will recommend destinations that meet the above key information to the first user. In addition, the first user can also invite other users in the travel service platform or even users of other platforms (such as social platforms) to use the travel planning services provided by the platform. For example, after clicking on the invitation control 303, you can select other users you want to invite in the invitation interface after the jump, which will not be repeated here.

[0055] First of all, it should be noted that the travel planning service provided by the travel service platform described in this specification is a planning service involving at least one aspect of the first user's travel (or trip). For example, it may include: integrating multiple travel data to plan (or formulate) travel-related plans for the user, providing travel-related suggestions for the user, answering the user's questions about any aspect of the travel, etc. For example, if the first user wants to travel to a certain place, a complete travel plan can be planned for the first user by integrating transportation, accommodation, routes, food, photography and other related content; or, if the first user inquires about transportation to a certain place, a low-cost travel plan can be planned for the first user by integrating multiple transportation routes such as civil aviation, railways, roads (self-driving, buses, etc.), and waterways (passenger ships); or, if the first user directly inquires about travel guides for a certain place, a high-quality travel plan can be planned for the first user by integrating multiple channels of public data such as maps and social platforms. For another example, the travel planning service may also include: individually answering questions raised by users regarding any travel-related area (such as flights, accommodation, catering, etc.), such as "What days are the May Day holiday?", "What time does check-in for flight xx start?", "What is the front desk phone number of hotel xx?", "Where is restaurant xx and what time does it open?", "What are some delicious barbecue restaurants within 1 km of me?", etc., which will not be repeated here.

[0056] In one embodiment, the planning server can pre-define multiple virtual personas related to travel planning, including but not limited to a travel director for planning the overall travel plan, a flight manager for creating flight plans, a route planner for creating itineraries, a travel expert for creating travel guides, a hotel tester for creating accommodation plans, and a budget manager for managing travel expenses. Each of these virtual personas can be used to perform detailed travel planning based on multiple options, including air tickets, attractions, routes, hotels, and budget.

[0057] Based on this, the planning server can determine at least one virtual persona that matches the travel planning request in a variety of ways. For example, the planning server can pre-formulate and maintain a mapping relationship between travel field keywords and the above-mentioned virtual persona (where any keyword is mapped to at least one virtual persona), so that for the received travel planning request, the target keyword contained in the request can be first extracted, and then the virtual persona corresponding to the target keyword can be determined as the virtual persona that matches the travel planning request based on the above-mentioned mapping relationship. Among them, if the travel planning request contains only one target keyword, the at least one virtual persona corresponding to the target keyword can be determined as the virtual persona that matches the travel planning request; if the travel planning request contains multiple target keywords, the at least one virtual persona corresponding to each target keyword can be removed and determined as the virtual persona that matches the travel planning request. The persona determination logic of this method is simple and can quickly determine the virtual persona. For another example, the planning server can directly input the travel planning request into the persona matching model for inference, and obtain the persona matching result output by the model, which is used to indicate at least one virtual persona corresponding to the travel planning request. The personality matching model can be pre-trained using standard sample data to ensure the accuracy of personality prediction.

[0058] In one embodiment, the at least one virtual persona that matches the travel planning request may include all predefined virtual persons related to travel planning, that is, each travel planning request initiated by the first user is planned from the perspective of all virtual persons, and a travel plan that is as detailed as possible is given. Alternatively, the at least one virtual persona that matches the travel planning request may also include a virtual persona that matches the identified planning intent of the travel planning request. Exemplarily, the planning intent of the travel planning request can be determined by extracting keywords or using a pre-trained intent recognition model, and then at least one virtual persona that matches the planning intent is determined from all predefined virtual persons related to travel planning. In this way, it can be ensured that the thinking process of the virtual persona displayed to the first user can be as consistent as possible with the planning intent of the first user, fully meeting the needs of the first user.

[0059] Step 204: call the travel planning model to perform travel planning processing, and determine the thinking process of each virtual person in the travel planning processing.

[0060] Step 206: Display the thought process corresponding to each virtual person through the first client of the travel service platform.

[0061] After determining at least one avatar that matches the travel planning request, the planning server can invoke the travel planning model to perform travel planning and determine the thought process of each avatar during the travel planning process. The determined thought process can then be sent to the first client for presentation to the first user.

[0062] It can be seen from the above embodiments that this specification proposes a new method for displaying travel planning content, that is, after receiving a travel planning request initiated by a first user, the travel service platform first determines at least one virtual person that matches the travel planning request, then calls the travel planning model to perform travel planning processing, and separately determines the thinking process of each virtual person in the travel planning processing; then, through the first client of the travel service platform, the thinking process corresponding to each virtual person is displayed separately for the first user to view.

[0063] As can be seen, this solution, through at least one virtual persona and its thought process, splits the complete processing of a travel planning request into professional processes completed separately by professionals in various fields (such as transportation, hotels, routes, strategies, etc.); further, by displaying the thought process corresponding to each virtual persona through the first client, it creates for the first user the feeling of using the product that a professional travel planning team composed of professionals in various fields collects data and executes the plan for the first user. It can be understood that this solution can ensure the accuracy and reliability of travel planning results based on the travel planning model; on this basis, by displaying the thought process of each virtual persona separately, it not only enriches the display form of the thought process, enhances the fun of use and significantly reduces the difficulty of understanding for the first user, making it easier for the first user to fully, accurately and intuitively understand the specific process of the travel planning model generating travel planning results; it can also effectively improve the credibility of the travel planning process and enhance the first user's trust in the travel planning results.

[0064] The travel planning service provided by the travel service platform is based on a travel planning model. The model can be trained on a basic model using sample data from travel-related fields. The basic model can be an LLM (Large Language Model) or a VLM (Vision-Language Model). This specification does not limit this.

[0065] In one embodiment, when the planning server calls the travel planning model to process the travel planning request, it can first construct a corresponding prompt word (Prompt) for the request and input the prompt word into the travel planning model for reasoning; then, obtain the thought chain and travel planning results output by the model, and determine the thinking process corresponding to each virtual person according to the thought chain.

[0066] When constructing the prompt word, the planning server may construct a prompt word that includes the functional description information of at least one of the virtual personalities, so as to indicate the specific functions of each virtual persona to the travel planning model through the prompt word, so that the model can accurately infer the specific role of each virtual persona in the travel planning process based on this information, thereby outputting a relatively more accurate thinking process and travel planning results. And / or, the travel data corresponding to the travel planning request may be first queried, and then the valid data corresponding to each virtual persona may be filtered out, and then the prompt word containing the above-mentioned valid data may be constructed. The above-mentioned travel data may be queried from the data locally maintained by the planning server, or from the aforementioned travel service or database associated with it; and the queried travel data may also include historical travel data (such as travel guide articles published on travel service platforms, evaluation information published on e-commerce platforms, etc.) and / or real-time travel data (such as real-time flight / ticket information, real-time hotel / ticket information, etc.). In this method, the travel service platform obtains the travel data required for the travel planning model in advance and directly includes it in the prompt words for the model to use. The model does not need to obtain data on its own. This is not only more in line with the current model operation logic, but also helps to simplify operations, improve the efficiency of travel planning processing, and reduce the waiting time of the first user.

[0067] Continuing from the above embodiment, for the question "How to play in S2 during the May Day holiday?" (i.e., travel planning request), the planning server can directly use the question as a prompt word, or construct prompt words in the above manner, such as constructing "I want to go to S2 during the May Day holiday. Please formulate a detailed travel plan for me from the perspectives of travel director (for planning the overall travel plan), flight manager (for formulating flight plans), travel expert (for formulating travel strategies), route customization engineer (for formulating travel routes), hotel test sleeper (for formulating accommodation plans), and budget manager (for managing travel expenses)" as prompt words; or constructing " I want to visit S2 during the May Day holiday. Please make a detailed travel plan for me based on the following travel data: <travel data list>” as a prompt; or, you can construct “I want to visit S2 during the May Day holiday. Please make a detailed travel plan for me based on the following travel data from the perspectives of travel director (for planning the overall travel plan), flight manager (for making flight plans), travel expert (for making travel strategies), route customizer (for making travel routes), hotel tester (for making accommodation plans) and budget manager (for managing travel expenses): <travel data list>” as a prompt.

[0068] In one embodiment, the virtual persona that matches the travel planning request may include a main persona and at least one branch persona, wherein the main persona is used to plan the travel plan as a whole, and each branch persona is used to plan part of the travel plan. Based on this, when the thinking process corresponding to each virtual persona is displayed separately through the first client, the main thinking process corresponding to the main persona and the branch thinking process corresponding to each branch persona can be displayed on the first client separately; wherein, the main thinking process includes a task splitting process and a task aggregation process; the task splitting process is used to: split the planning main task corresponding to the travel planning request into one or more planning subtasks, and output the information of the branch persona corresponding to each planning subtask; the task aggregation process is used to: summarize the branch thinking results of each branch persona; and the branch thinking process corresponding to each branch persona is used to: think about the planning subtasks assigned to the main persona. In this way, the effect of multiple people participating in travel planning processing can be presented to the first user, where "the main persona performs the main planning task (i.e., the overall planning of the travel plan), and each branch persona performs the split planning sub-tasks respectively", further enhancing the authenticity and credibility of the thinking process and travel planning results.

[0069] Among them, the branch personas can include a budget analysis persona and at least one travel expert persona; at this time, the planning subtask corresponding to any of the travel expert personas includes: querying the corresponding travel data through the travel data query logic that is consistent with the specific settings of the travel expert persona; and the planning subtask corresponding to the budget analysis persona includes: performing budget analysis on the travel data queried by each travel expert persona. It can be understood that the specific settings of each travel expert persona belong to a dimension related to travel, so the at least one travel expert persona can query travel data from at least one dimension such as flights, attractions, hotels, souvenirs, etc., and the budget analysis persona can perform budget analysis on the travel data of each of the above dimensions separately, and obtain the specific budget (i.e., expected expenditure) of each dimension and the total budget of the travel planning results. In this way, the budget analysis persona can accurately and comprehensively evaluate various expenses for the first user, which helps the first user to accurately and comprehensively understand the spending situation of the travel planning results, helps to avoid overspending and improves the first user's acceptance of the travel planning results.

[0070] In one embodiment, the planning server can split and semantically analyze the thought chain in text form output by the travel planning model, and determine the correspondence between the various parts of the thought chain and each virtual persona based on the analysis results, and thereby determine the thought process corresponding to each virtual persona. Alternatively, in the case where the aforementioned prompt words contain the personality description information of each virtual persona, the travel planning model can also output the corresponding parts of each virtual persona while outputting the thought chain, so that the planning server can determine the thought process corresponding to each virtual persona accordingly (that is, the travel planning model itself determines and outputs the thought process of each virtual persona).

[0071] In one embodiment, the planning server can continuously receive the thought chain output by the travel planning model, and continuously send the thinking process corresponding to each virtual person determined according to the thought chain to the first client; accordingly, the first client can display the thinking process of each virtual person received in real time, so as to present to the first user the effect that the virtual person thinks in chronological order and finally gives the travel planning result.

[0072] In one embodiment, when the thinking process corresponding to each virtual persona is displayed separately through the first client, different ways of display can be used according to the number of virtual persons. For example, when the number of virtual persons that match the travel planning request is one, the display area corresponding to the virtual persona can be shown in the planning process display interface on the first client, and the thinking process corresponding to the virtual persona can be displayed in the display area. When the number of virtual persons that match the travel planning request is multiple, the display area corresponding to each virtual persona can be shown in the planning process display interface on the first client, and the thinking process corresponding to the corresponding virtual persona can be independently displayed in each display area, so as to simultaneously display the thinking processes of multiple virtual persons to the first user, thereby creating the effect that multiple virtual persons are planning travel plans for the user at the same time.

[0073] Among them, the first client can adopt at least one of the following methods to display the thinking process corresponding to any virtual person in the display area corresponding to the virtual person. Continuing from the above embodiment, for the travel planning request of "How to go to S2 during the May Day holiday?", it is assumed that the virtual personas matched with the request determined by the planning server include a main persona and multiple branch persons. Among them, the main persona can be a travel director, and the multiple branch persons can include: flight stewards, travel experts, route customization experts, hotel test sleepers and budget managers. Among them, the travel planning process includes four stages in sequence: "task splitting", "travel expert query information", "budgetist splitting the budget" and "information aggregation". Figures 6 to 13 This is an exemplary display of the thinking process corresponding to each virtual character, which is explained below in conjunction with the embodiments.

[0074] For example, the thought process corresponding to any virtual persona may include the thought content and / or thought progress of the virtual persona. In this case, the first client may dynamically display the thought content and / or thought progress in the display area corresponding to the virtual persona so that the first user can accurately understand the specific content and current progress of the thought process. The thought content of any virtual persona may include at least one of the following: travel data query logic consistent with the virtual persona (such as flight experts querying online travel platforms and / or airlines for valid flight information publicly available, travel experts querying travel guides for a certain attraction on social platforms, hotel testers querying user reviews of a certain hotel on online group buying platforms, etc.), statistical information of travel data retrieved based on the travel data query logic (such as the number of valid flights retrieved by flight experts, the number of travel guides retrieved by travel experts, the number of user reviews retrieved by hotel testers, etc.), and detailed information of the queried travel data (such as each valid flight information retrieved by flight experts, each travel guide retrieved by travel experts, each user review retrieved by hotel testers, etc.).

[0075] For example, in the "task splitting" stage, Figure 6 As shown, the display area 602 of the travel director 601 displays the thought content 603 in text form (ie, "The user asks... to collect information."). In the "travel expert query information" stage, as shown in FIG. Figure 7 As shown, the display area 701 of the four branch characters such as the flight expert respectively displays their respective thinking progress (all are "preparing for work..."); Figure 8 As shown, the display areas of the four branch personas, such as the flight expert, respectively display their respective thinking content and thinking progress. For example, the display area 801 of the travel expert displays thinking content, specifically including text content 802 (i.e., "Combining May weather... popular attractions list") and image content 803 (i.e., a map in image form), and also displays thinking progress 804 (i.e., "1 / 5" in text form, indicating that the travel expert needs to specify 5 planning subtasks in sequence and is currently specifying the first planning subtask); Figure 9 As shown, each planning subtask of the flight expert, travel expert, and route planner has been completed (for example, the travel expert's display area 1001 displays the thinking progress 1002, i.e., the text "5 / 5"), while the four planning subtasks of the hotel tester and budget manager are currently executing the first one; Figure 9 As shown in the figure, the five branch characters, including the flight expert, have completed their respective planning subtasks, so their respective thinking progress is displayed in their respective display areas (such as "4 / 4", "5 / 5" and "Route query completed"). Figure 12As shown, the budget analyst's display area 1201 displays thinking content, specifically including text content 1202 (i.e., "sorting... back and forth") and animation content 1203 (i.e., the optimal solution calculation progress being dynamically displayed). Figure 13 As shown, the travel director's display area 1301 displays thinking content (i.e., the text "combining thinking...travel strategy").

[0076] For another example, the first client can adjust the size of the display area corresponding to any virtual persona according to the thinking progress corresponding to the virtual persona, and display a corresponding amount of thinking content in the display area. This solution can adjust the size of the display area according to the thinking progress, and adjust the amount of displayed thinking content, so as to realize the automatic expansion and closing of the display area, thereby presenting a dynamic display effect. Taking the aforementioned travel expert as an example, by comparing Figures 7 to 10 It can be seen that when the character starts thinking, its display area 701 is small; as the thinking deepens, the display area 801 enlarges and displays more thinking content; and after the thinking is completed, the display area 1001 (and 1101) shrinks and stops displaying the thinking content (the shrunken display area 1101 is the same size as the display area 701 before enlargement).

[0077] For another example, the first client can display part of the thinking process corresponding to any virtual person in the display area corresponding to the virtual person; and in response to a browsing trigger operation on the display area, switch the display content in the display area to other thinking processes corresponding to the virtual person. In this way, during the display of the thinking process corresponding to any virtual person, the first user can flexibly adjust the part of the thinking process he wants to view according to his own wishes, and the display method is more flexible. Figure 8 As shown, the first user slides his finger on the flight information displayed in the flight expert's display area, which can trigger the first client to scroll and display the flight list; slides his finger on the map displayed in the travel expert or route customizer's display area, which can trigger the first client to adjust the center position of the map; slides his finger on the hotel information displayed in the hotel tester's display area, which can trigger the first client to scroll and display the hotel list, etc., which will not be repeated here. Figure 15 As shown, it is difficult to display all the flight information retrieved by the flight expert on the same screen interface, so a sliding button 1501 can be displayed. At this time, the first user can control the first client to scroll up and down to display each flight information by dragging the button left and right.

[0078] For another example, when the first client displays the thinking process corresponding to any virtual character in the display area corresponding to the virtual character, in response to the trigger operation for viewing the details of the display area, a floating window is generated on the first client, and the thinking process corresponding to the any virtual character is displayed in the floating window. In this way, the first user can trigger the further display of the details of the thinking process in the floating window, so as to view the thinking process in more detail and comprehensively. Figure 8 As shown, a zoom control 805 is displayed in the upper right corner of the flight information displayed in the display area of ​​the flight expert. If the first user clicks the control or directly clicks on the flight information, the user can jump from the current interface to the flight details interface, or generate a floating window above the current interface, and display a richer flight list or more comprehensive flight information in the floating window. For example, Figure 9 The floating window 901 shown displays flight expert's thoughts 902 and a complete flight list 903 .

[0079] For another example, the first client can display the thinking process of any virtual person in the display area corresponding to the virtual person when the thinking process corresponding to the virtual person has not yet ended; and when the thinking process corresponding to the virtual person has ended, stop displaying the thinking process and display the corresponding thinking end prompt information in the display area corresponding to the virtual person. In this way, the first client will actively prompt the first user after the thinking process of any virtual person ends, so that the user can accurately know the current status of the thinking process, which helps to improve the first user's perception of the travel planning process. Figures 7 to 10 As shown, before the thinking process is finished, the display area of ​​each virtual character displays the thinking process of the character; Figure 11 As shown, after the thinking process is completed, each virtual character's display area will display its own thinking completion prompt information, such as "Air ticket query completed" corresponding to the flight expert, "Attraction query completed" corresponding to the travel expert, etc., which will not be repeated here.

[0080] For another example, the thinking process of any virtual person includes thinking content, which contains interactive content related to the preset travel service, so the first user can perform corresponding interactive operations on the interactive content. In this regard, the first client can display the related content of the travel service and / or execute the related instructions of the travel service in response to the interactive content being triggered. This method allows the first user to trigger the travel service platform to call its associated travel service to achieve richer corresponding functions. Figure 8As shown, the display area of ​​the flight expert displays the words "5000 flights", where the number "5000" is a triggerable text 806. If the first user clicks on the text, the first client can jump to another interface or display a floating window, and display the detailed information of the 5000 flights in the other interface or floating window; if the first user clicks on a flight 807, the first client can further display the detailed information of the flight, etc., which will not be repeated. For another example, the display area of ​​the hotel tester displays a list of hotels. If the first user clicks on a hotel 808, he can jump to the hotel's website homepage or booking page, or directly send a purchase instruction for the hotel to the online purchase platform, execute a collection instruction for the hotel (i.e., collect the hotel), etc., which will not be repeated.

[0081] In one embodiment, the first client can also display the processing progress of the travel planning process, that is, the processing progress corresponding to the above-mentioned multiple stages. Figure 6 As shown, the processing progress 605, namely the text "Task 1 / 4", is displayed on the right side of the statistical results of the above travel data, indicating that it is currently in the first stage of the above four stages (ie, the "task splitting" stage). Figures 7 to 11 As shown, the processing progress, i.e., the text "Task 2 / 4", is displayed on the right side of the statistical results of the travel data, indicating that the current stage is the second of the four stages mentioned above (i.e., the "Travel Expert Query Information" stage). Figure 12 As shown in the figure, the processing progress is displayed on the right side of the statistical results of the travel data, that is, the text "Task 3 / 4", indicating that it is currently in the third stage of the above four stages (i.e., the "Budget Analyst Splits the Budget" stage). Figure 13 As shown, the processing progress, namely the text "Task 4 / 4", is displayed on the right side of the statistical results of the travel data, indicating that it is currently in the fourth stage (ie, the "information summary" stage) of the above four stages.

[0082] In one embodiment, when the planning server queries the corresponding travel data in response to the travel planning request, the server can also send the statistical results of the queried travel data to the first client for display. For example, the planning server can count the data in real time during the process of querying the travel data, and continuously send the corresponding statistical results to the first client so that the client can display the statistical results in real time. In this way, the amount of real data queried can be fully displayed to the first user, showing that the subsequent travel planning results are planned based on a lot of real data, thereby improving the credibility of the travel planning results. Figures 6 to 13 As shown, the statistical results of the queried travel data are displayed above the conversation interface, including Figures 6 and 7 It displays "2611 routes" and "15362 real travel data"; Figures 8 to 13It displays "3046 routes" and "20140 real travel data".

[0083] By displaying the thinking process of each virtual persona in the aforementioned manner, the first client can fully demonstrate to the first user the specific process of multiple professionals participating in generating travel planning results, thereby effectively enhancing the first user's trust and acceptance of the travel planning results generated by the travel planning model.

[0084] In one embodiment, the planning server can also determine the travel planning result corresponding to the thinking process obtained by the travel planning process and send it to the first client for display. In this way, the result of the travel planning process performed by the travel planning model can be output to the first user so that the user can check whether the result meets his or her requirements.

[0085] Among them, after the thinking process is completed, the first client can display the total amount of travel data queried during the thinking process, and / or the number of valid travel data in the travel planning results filtered from the queried travel data, so that the first user can intuitively perceive the real data used in the travel planning process of the travel planning model, which helps to reflect the authenticity of the travel planning results and enhance the first user's acceptance of the results.

[0086] And / or, the first client may display the thought process in thumbnail form and the travel planning result in detailed form; thereafter, in response to a viewing trigger operation for the thought process, the thought process may be switched from thumbnail form to detailed form, wherein when displaying the thought process in detailed form, the first client is provided with selection controls corresponding to multiple virtual persons and a shared display area for the multiple virtual persons; and, when the selection control corresponding to any one of the multiple virtual persons is selected, the thought process corresponding to the any one of the multiple virtual persons is displayed in the shared display area. In this way, the first client can automatically retract the corresponding display area after the thought process ends, and display the relatively more important travel planning result in detailed form for the first user to view more conveniently.

[0087] like Figure 14 As shown, the process display area 1401 in the conversation interface displays a summary of the thinking process, along with a corresponding detail viewing control 1402. The area below displays the travel planning results, including text results 1403 (i.e., a general description of the travel plan) and table results 1404 (i.e., the specific content of the travel plan, such as the itinerary).

[0088] If the user triggers the detail view control 1402, the first client can further display the complete content of the above thought process, see Figure 15 .like Figure 15 As shown, the conversation interface displays the entire thought process for the travel planning process, specifically, the thought processes for each of the four aforementioned stages of the travel planning process, displayed from top to bottom. Furthermore, to facilitate the user's intuitive understanding of the speed of the thought process at each stage, time information is displayed at appropriate locations for each stage. For example, the total duration 1502 for each stage is displayed to the right of the start of each stage, e.g., "Task Splitting" (2 seconds), "Travel Expert Information Query" (3 seconds), and "Information Aggregation" (3 seconds).

[0089] For multiple branch personas such as the flight expert, the first client can display selection controls 1503 for each virtual persona and display the thought process of the first branch persona (i.e., the flight expert) by default in the shared display area 1504. If the first user selects the selection control for "travel expert," the first client can display the travel expert's thought process in the shared display area 1504, allowing the first user to switch between the thought processes of different virtual persons.

[0090] Below the abbreviated thought process, relevant information 1405 about the travel data is also displayed, such as the total amount of travel data retrieved and the number of valid travel data filtered into the travel planning results, such as "50 hotels were retrieved, and 15 were finally filtered out" and "26 hotels were retrieved, and 8 were finally filtered out." In addition, the first user can perform corresponding quick operations based on key information in the above travel planning results. For example, the first client can display a one-click collection control 1406 (i.e., a button with the words "One Favorite Route") for the itinerary in the above table results. At this time, the first user can trigger this control to collect the above itinerary with one click.

[0091] It is understood that the thought process and travel planning results displayed by the first client fully reflect the travel service platform's response to the travel planning request initiated by the first user. To this end, the planning service or first client may further record the thought process and travel planning results as corresponding travel planning playback content. Furthermore, a trigger control corresponding to at least one recorded travel planning playback content may be displayed on at least one second client of the travel service platform. When the trigger control corresponding to any travel planning playback content is selected, the travel planning playback content may be played back on the second client.

[0092] If the first user is satisfied with the above response process, that is, recognizes the above thinking process and travel planning results, he or she may have the need to share the above response process with other users. Figure 16As shown, for the displayed travel planning result 1601, if the first user wants to share, the sharing control 1602 can be triggered; accordingly, the first client can jump to Figure 17 The sharing interface shown allows the first user to select the user / platform (i.e., one or more of A to E) to share with in the interface, and can choose whether to share the aforementioned thinking process at the same time. Among them, if the first user chooses to share the aforementioned thinking process, the first client can notify the planning server to generate corresponding travel planning playback content based on the thinking process and travel planning results. Of course, after obtaining full authorization from the first user and relevant parties, the planning server can also generate travel planning playback content on its own after obtaining the travel planning results, without the need for the first user to actively share.

[0093] It is worth noting 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, stored data, displayed data, etc.) involved in this manual are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0094] In one embodiment, the travel service platform can support sharing and playback of travel planning content. For example, the planning service can maintain a content collection, then place all generated travel planning playback content in this collection. Based on preset recommendation / retrieval logic, each travel planning playback content in this collection is distributed to a client (hereinafter referred to as the second client) used by a corresponding user (hereinafter referred to as the second user) for playback, thereby replaying (i.e., reproducing) the aforementioned thinking process and travel planning results on the second client (see the embodiments below for details).

[0095] Based on the above-mentioned content collection, the travel service platform can also maintain a public space jointly realized by multiple users, which may be called an "inspiration square" or "inspiration space". For example, after the first user uses the first client to generate a relatively satisfactory travel planning result, he can choose to share the travel planning result and the corresponding thinking process in the square: the planning service end or the first client can record the thinking process and travel planning result as travel planning playback content, and publish the content in the square; further, the second user can browse the various travel planning playback contents published to the square through the second client, and when he is interested in any content, he can replay the content locally through the second client. In this way, the first user can share his travel-related inspirations and the corresponding travel planning playback content with other users in the square, which helps to expand the dissemination range of high-quality travel planning content and enhance the user activity and fun of the travel service platform.

[0096] In one embodiment, the question posed by the first user may be too simple, resulting in the travel planning model receiving too little valid information. Therefore, the first user may be dissatisfied with the travel planning result output by the travel planning model after the travel planning model processes the travel planning, and the first user may need to re-plan the travel plan. In one embodiment, to reduce the operational complexity of the first user's re-planning and improve their user experience, the travel service platform can allow the first user to modify parameter values ​​based on the current travel planning result and quickly re-plan the travel plan. For example, the first client can display the parameter values ​​of the travel planning result in the target query dimension, as well as a parameter value adjustment control for the target query dimension; at this time, the first user can perform a parameter adjustment operation on the control, and the travel service platform can respond to the parameter value adjustment control being triggered, update the travel planning request according to the corresponding adjusted parameter values, and re-call the travel planning model to process the travel planning for the updated travel planning request. The user can adjust the parameter values ​​of many parameters, including but not limited to the number of people, age of people (children, middle-aged people, elderly people, etc.), travel date and location, attraction type (such as cultural and historical, natural scenery, etc.), budget, etc.

[0097] Take budget as an example, Figure 16As shown, a budget adjustment control 1603 can be displayed below the travel planning result in the conversation interface, and the first user can drag the control to adjust the current budget. In addition, during the budget adjustment process, the first client can display a prompt message 1604 corresponding to the adjusted current budget in real time, such as a prompt of "need to change to economy class" when adjusted to 2,400 yuan; a prompt of "need to take a red-eye flight" when adjusted to 1,500 yuan, etc. After the adjustment is completed, the first user can click the confirmation button 1604 to submit the adjusted budget to the travel service platform and instruct the platform to re-process the travel planning according to the adjusted budget and generate the corresponding travel planning results. Of course, similar to the previous generation process, the thinking process of each virtual person can also be displayed separately in the manner described in the previous embodiment during the regeneration process, which will not be repeated here.

[0098] This concludes the introduction to the travel planning content display solution (i.e., a solution in which the travel service platform invokes the travel planning model to perform travel planning processing and displays the corresponding thought processes of each virtual persona through the first client). Furthermore, corresponding to the aforementioned content playback embodiment, this specification also proposes a travel planning content playback solution, which is described in detail below with reference to the accompanying figures and corresponding embodiments.

[0099] In related technologies, platforms typically provide travel planning services to individual users, specifically responding to each user's travel planning requests and developing corresponding travel plans. Because platforms provide travel planning services to different users, it's likely that different users will ask the same (or similar) questions. In these cases, the platform needs to independently plan and provide answers for each user's questions. Consequently, duplicate questions can significantly waste the platform's computing resources (such as the operating resources of the travel planning model). Furthermore, travel service platforms lack interaction among the multiple users they serve, resulting in low user retention and activity levels, which urgently require improvement.

[0100] To address the aforementioned technical issues, this specification proposes a travel planning content playback solution, building upon the previously described travel planning content display solution. This solution aims to enhance user interaction and improve user retention and activity by providing recording, sharing, and playback capabilities to users of the travel service platform. It also aims to reduce the number of times the platform and travel planning models answer the same or similar questions, thereby conserving platform resources. This solution is described below with reference to the accompanying figures and corresponding embodiments.

[0101] Figure 18 This is a flowchart of a method for replaying travel planning content provided by an exemplary embodiment, which is applied to a travel service platform. Figure 18 As shown, the method includes the following steps 1802 to 1806.

[0102] Step 1802: Based on the received travel planning request, call the travel planning model to perform travel planning processing.

[0103] The specific process of the travel service platform receiving the travel planning request and calling the travel planning model to perform travel planning processing can be found in the detailed description of the above embodiment and will not be repeated here.

[0104] Step 1804: Display the thought process of the travel planning process and the corresponding travel planning results on the first client of the travel service platform, and record the thought process and travel planning results as corresponding travel planning playback content.

[0105] In one embodiment, when displaying the thought process of the travel planning process on the first client of the travel service platform, at least one target virtual persona corresponding to the travel planning request and the thought process of each target virtual persona during the travel planning process can be first determined, and then the thought process corresponding to each target virtual persona can be displayed separately through the first client. This method, by displaying the thought process corresponding to each virtual persona separately through the first client, can create for users the feeling of using the product, where a professional travel planning team composed of professionals in various fields collects data and executes plans for users. This not only enhances the fun and significantly reduces the difficulty for users to understand, allowing them to fully, accurately, and intuitively understand the specific process by which the travel planning model generates travel planning results. It can also effectively enhance the credibility of the travel planning process and increase users' trust in the travel planning results.

[0106] The thought process corresponding to any virtual persona may include the thought content and / or thought progress corresponding to the virtual persona. Furthermore, the thought content corresponding to any virtual persona may include at least one of the following: travel data query logic corresponding to the virtual persona, statistical information on travel data retrieved based on the travel data query logic, and detailed information on the retrieved travel data. For details, please refer to the detailed description of the previous embodiments and will not be repeated here.

[0107] It should be noted that the travel planning playback content described in this specification is not a video recorded and generated by the process of displaying the thinking process and travel planning results on the first client, and playing back the travel planning playback content is not a video. In fact, the travel planning playback content described in this specification is a data set obtained by organizing the thinking process and travel planning results in an orderly manner according to the order of generation, and the set contains structured data in one or more formats. In addition, the process of "playing back" the travel planning playback content by the second client is the process of displaying each data in the order of generating the thinking process and travel planning results - because the various data in the travel planning playback content are organized in an orderly manner according to this order, therefore, displaying each data in this order can ensure that the display effect of the travel planning playback content when it is played back in the second client is consistent with the display effect when the thinking process and travel planning results are displayed in sequence in the first client, thereby achieving "replay" of the display content and display effect. In short, the "replay" described in the embodiment of this specification is essentially to re-display the thinking process and travel planning results in the same display method (that is, to re-display each structured data in the travel planning playback content according to the structured information).

[0108] In one embodiment, the first client can record the thinking process and travel planning results locally as travel planning playback content; alternatively, the planning service can also record the travel planning playback content locally and send the content or its description information to the first client. Furthermore, a sharing authorization control for the travel planning playback content recorded on the first client can be displayed on the first client so that the first user can decide whether to share the content. Furthermore, in response to the authorization confirmation operation for the sharing authorization control, the travel planning playback content recorded on the first client can be shared on the travel service platform so that the travel planning playback content is allowed to be played back on the second client. It is understandable that because the thinking process and travel planning results recorded as travel planning playback content are generated by the travel planning model when the travel planning request initiated by the first user is processed, the travel planning playback content can be regarded as the creative product of the first user (i.e., data generated based on the user's creativity / inspiration), so the content is shared by the first user to the travel service platform for the second user to view.

[0109] As mentioned above, the travel service platform can maintain a public space: "Inspiration Square", so that the aforementioned travel planning playback content can be shared (or published) to the square by the first user so that the second user who enters the square can view (i.e., replay), thereby expanding the dissemination scope of high-quality travel planning content and enhancing the user activity and fun of the travel service platform.

[0110] It should be noted that the first user and the second user referred to in this specification can be any user on the travel service platform, and the first client and the second client are the clients used by the respective users. For example, if the first user and the second user are the same user, then when the second client plays back the travel plan playback content, it is actually the user viewing the travel plan playback content that they themselves shared on the travel service platform. If the first user and the second user are different users, then when the second client plays back the travel plan playback content, it is actually the second user viewing the travel plan playback content that the first user shared on the travel service platform.

[0111] Step 1806: Display the trigger control corresponding to at least one recorded travel planning playback content on at least one second client of the travel service platform, and when the trigger control corresponding to any travel planning playback content is selected, replay any travel planning playback content on the second client.

[0112] As can be seen from the above embodiments, this platform-centric solution provides users with a sharing function for travel planning replay content: a first user generates a thought process and a corresponding travel planning structure using a first client. This thought process and the corresponding travel planning results can then be recorded as travel planning replay content and shared on the platform. A second user can then replay the travel planning replay using a second client to view the travel planning results and their generation process.

[0113] It can be understood that in this solution, the first user can use the travel service platform to share their travel-related inspirations and corresponding travel planning replays with the second user, thereby enhancing user interaction and making the platform more interesting, helping to improve the platform's user retention and user activity. Furthermore, by viewing the diverse replays shared by other users on the platform, on the one hand, it helps to strengthen the second user's willingness to travel and use the travel planning service, and on the other hand, it can inspire the second user to generate more inspiration, thereby asking more valuable questions to the travel planning model and generating corresponding travel planning replays. Furthermore, these high-value questions and travel planning replays can inspire more users' creativity and inspiration, and attract more users to use this feature, achieving an overall virtuous cycle. Furthermore, these increasingly rich high-value questions and travel plans can also serve as new training samples for continuous iterative updates of the travel planning model, thereby achieving high-quality evolution of the travel planning model.

[0114] In one embodiment, a second client can obtain at least a portion of the travel planning playback content shared to the travel service platform from the planning server and playback it. For example, upon opening the travel planning service portal interface, the second client can obtain at least a portion of the travel planning playback content from the planning server and extract its brief information to display a corresponding trigger control. When the second user selects a trigger control for any of the travel planning playback content, playback of the content begins. For another example, upon opening the travel planning service portal interface, the second client can also obtain at least a portion of the travel planning playback content from the planning server and display a corresponding trigger control. Then, when the second user selects a trigger control for any of the travel planning playback content, playback of the content begins again from the planning server. It is understood that the former method obtains a larger amount of data from the planning server but responds faster to user operations, while the latter method obtains a smaller amount of data from the planning server but responds slower to user operations. Therefore, when implementing the solution, the appropriate content acquisition method can be flexibly selected based on actual business needs.

[0115] In one embodiment, at least one travel planning playback content corresponding to the trigger control displayed by the second client may include one of the following: for example, the request content of the corresponding travel planning request meets the travel planning playback content of the preset request screening conditions; wherein the preset request screening conditions include at least one of the following: the request content is typical (meaning that the request content can accurately reflect the statistical properties of similar request content, emphasizing the consistency of the sample and the whole in structure or characteristics, such as "how to book a hotel", "how to book a plane ticket" and other requests for inquiries about the operation process), the request content is representative (meaning that the request content can concentrate on reflecting the core characteristics or commonalities of similar request content, but there is no need to preset clear overall boundaries, such as "how to play from xx to yy", "xx scenic spot tour route" and other requests for travel strategies), and the request content is interesting (that is, the request content is relatively interesting and usually has a strong appeal to viewers, such as "how to stay in a five-star hotel with 300 yuan", "how to make up for expired tickets" and other requests for skills in special scenarios).

[0116] In other words, the request content for each travel planning replay corresponding to the trigger control displayed by the second client can be typical, representative, and / or interesting. By displaying this replay content (especially replay content with interesting request content) to platform users, it helps to enhance other users' willingness to travel and use the travel planning service. It can also inspire other users to generate more inspiration, thereby asking more valuable questions to the travel planning model and generating corresponding travel planning replay content. Furthermore, these high-value questions and travel planning replay content can inspire more users' creativity and inspiration, and attract more users to use this function, thus achieving a virtuous cycle overall.

[0117] For another example, the at least one travel planning playback content may include travel planning playback content that meets preset recommendation conditions; wherein the preset recommendation conditions include at least one of the following: the playback popularity reaches a preset popularity (such as the total number of playbacks is higher than 1,000 times, the total number of playbacks is higher than 800 people, the number of playbacks in the past week is higher than 200 times, etc.), travel planning playback content recorded within a preset time period (such as playback content generated within the past week, etc.), travel planning playback content shared within a preset time period (playback content shared to the square within the past month, etc.), travel planning playback content recorded within a preset geographical range (such as playback content triggered by users located in city S1, etc.), and travel planning playback content shared within a preset geographical range (such as playback content shared by users located in city S2, etc.).

[0118] For another example, the at least one travel planning playback content may include the request content of the corresponding travel planning request and / or travel planning playback content whose content matches the user feature data of the logged-in user (i.e., the second user) on the second client. Exemplarily, the user characteristic data may be positioning data. For example, when the second user is located in city S3, the at least one travel planning playback content may include playback content with requested content related to city S3, such as "What fun places are there in S3?", "How to visit xx attractions in S3?", etc.; or, the user characteristic data may also be occupational data. For example, when the second user sets his occupation as a student on the travel service platform, the at least one travel planning playback content may include playback content with requested content containing keywords such as "saving money", "winter vacation", and "summer vacation"; or, the user characteristic data may also be preference data. For example, when the second user sets the preferred attraction type as "natural landscape" on the travel service platform, the at least one travel planning playback content may include playback content with requested content containing keywords such as "xx Forest Park", "xx Lake", and "xx Mountain", and when the second user sets the preferred attraction type as "humanities and history" on the travel service platform, the at least one travel planning playback content may include playback content with requested content containing keywords such as "xx Museum", "xx Art Gallery", and "xx Exhibition", etc., which will not be repeated here.

[0119] In one embodiment, when a trigger control corresponding to at least one recorded travel planning replay content is displayed on any second client, a selection control corresponding to at least one classification dimension can be displayed on the second client for the second user to select. Furthermore, in response to the selection control corresponding to any of the classification dimensions being triggered (by the second user), at least one travel planning replay content belonging to any of the classification dimensions can be further displayed; wherein each travel planning replay content belongs to one or more classification dimensions. In this way, the second user can actively and centrally view travel planning replay content of a specific classification dimension of interest through operation, which helps to avoid problems such as difficulty in locating the focus and low viewing efficiency that may occur when multiple types of replay content are displayed together.

[0120] like Figure 3As shown, the entry interface of the travel planning service displays an inspiration space, and the replay content published in this space is divided into multiple classification dimensions such as "Featured", "How to Book a Hotel", "How to Book a Flight", and "How to Plan a Tour Route". The interface displays selection controls corresponding to each classification dimension, and by default displays multiple travel planning replay contents under the "Featured" dimension, such as "My First AI Travel Partner After Retirement", "How to Stay in a Four-Star Hotel with the Company's 300 Yuan Daily Allowance", and "AI Easily Manages Travel for Three Generations". At this time, if the user clicks the "How to Book a Flight" control 305, the first client can refresh the various travel planning replay contents belonging to the "Featured" dimension below to at least one travel planning replay content belonging to the "How to Book a Flight" dimension. Other controls are similar and will not be repeated here.

[0121] In addition, the display area of ​​any travel planning replay content can display a trigger control for that content. For example, for the replay content "How to stay in a four-star hotel with the company's 300 yuan daily allowance," its display area 306 displays brief information such as a main table 307, the content name 308 (which can be the same as the travel planning request, or a keyword extracted from the travel planning request or a synonym for the travel planning request), the publisher's (i.e., the first user's) account name 309, and the current number of replays 310.

[0122] In one embodiment, the second client can immediately retrieve and play back any travel planning playback content in response to the trigger control being triggered, thereby simplifying the second user's operation. Of course, it is also possible to wait for the second user to trigger the playback so that the second user can control the playback rhythm.

[0123] In one embodiment, a planning playback interface can be displayed on the second client, and any triggered travel planning playback content can be played back in this interface. The content can be played back in the order of each part, so that the second user can fully view the thinking process and the complete presentation of the travel planning results.

[0124] Alternatively, given that the travel planning playback content is obtained by recording the thinking process and travel planning results, and the thinking process may include thinking process parts corresponding to multiple virtual personalities, the planning playback interface can also be divided into a first area for replaying any of the travel planning playback content, and a second area different from the first area; wherein, when any of the travel planning playback content is divided into multiple content segments in sequence, the second area is used to display the playback controls corresponding to each content segment; and then, in response to the playback control corresponding to any content segment being triggered, the playback of any of the travel planning playback content can be started from the any content segment. In this way, the second user can adjust the playback progress of the travel planning playback content by himself, such as directly jumping to the content segment of his interest to start playback, without waiting for all the content segments to be played back in sequence.

[0125] Following the above embodiment, Figures 6 to 15 The thinking process display process shown includes four stages, so the travel planning playback content generated according to the thinking process and the travel planning result includes five stages; for this purpose, the travel planning playback content can be divided into five content segments (the first four segments correspond to the four stages of the thinking process, and the fifth segment corresponds to the process of displaying the travel planning result), and the travel planning playback content is played back in the first area of ​​the planning playback interface (that is, the above five content segments are played back in sequence), and the playback controls corresponding to each content segment are displayed in sequence in the second area. The second user Figure 1 Swipe up on the interface shown to view more selected content, such as Figure 19 As shown. Let's assume that the user triggers the playback of the travel planning playback content corresponding to the above embodiment, then the second client can jump to the planning playback interface. Figures 20 to 24 In the planned playback interface shown, the top and middle parts of the interface constitute the first area for playing back each content segment, and the bottom part of the interface serves as the second area to display each playback control. Figure 20 For example, from left to right, the process playback control 2001 (including the four controls of "task splitting", "travel expert query information", "budgetist split budget" and "information summary") and the result playback control 2002 (i.e., "view result" control) are displayed in sequence. If the second user clicks the "task splitting" control, the first stage of the thinking process will be directly played back in the first area (the task splitting results can be seen in the Figure 20 ); If the second user clicks the "travel expert query information" control, the second stage of the thinking process is directly replayed in the first area (the process of the four travel experts each performing the planning subtask can be seen in Figure 21); If the second user clicks the "Budget Analyst Split Budget" control, the third stage of the thinking process will be directly replayed in the first area (the specific process of the budget analyst performing budget analysis can be found in Figure 22 ); If the second user clicks the "Information Summary" control, the fourth stage of the thinking process will be directly replayed in the first area (the information summary results can be found in Figure 20 ); If the second user clicks Figures 20 to 23 Click the "View Results" control on any image in the second area to directly play back the travel planning results. The playback effect is visible. Figure 24 .

[0126] Among them, when showing any content segment of any travel planning playback content, the corresponding playback progress can be shown at the playback control corresponding to the content segment. For example, the corresponding playback progress bar (such as Figure 20 The progress bar in 2003, Figure 21 The progress bar 2101, Figure 22 The progress bar 2201, Figure 23 Based on this, the second user can see the corresponding playback progress in the playback control, improving their perception of the playback progress. And / or, in response to the progress adjustment operation on any playback control, the content segment corresponding to the playback control can be played back, and the playback progress of the content segment can be controlled based on the progress adjustment operation. Based on this, the second user can flexibly and accurately adjust the playback progress of each content segment, making it easier for them to view the corresponding content. Figure 21 In the interface shown, the second user can drag the progress bar 2101 left and right to accurately adjust the playback progress in the content segment of "Travel Expert Query Information".

[0127] and Figures 20 to 24 Similar to the embodiment shown, if the second user triggers playback Figure 19 Any content 1901 shown (of course, you can also Figure 1 The second client can display Figure 25 The planning playback interface shown in the figure is as follows. Figure 25 As shown, the first area 2501 at the top of the interface is used to play back the travel planning playback content, and the second area at the bottom is used to display various playback controls, such as the process playback control 2502 (including the three controls of "split task", "all-round hotel query" and "output summary") and the result playback control 2503 (i.e., "view result" control). Accordingly, the second user can control the playback progress of the travel planning playback content by triggering the above-mentioned controls, or drag the playback progress bar (such as Figure 26 The progress bar 2601) shown adjusts the playback progress and will not be described in detail.

[0128] Among them, if the user triggers the result playback control (such as clicking Figure 23 or Figure 26 In addition to displaying the corresponding travel planning results, the second client can also update the control to a review control, such as Figure 24 The control 2401 shown and Figure 27 The control 2701 shown. Obviously, if the user clicks the replay control, the second client can restart the display of the travel planning playback content from the starting position of the travel planning playback content.

[0129] In addition, the second client can also pause playback of the travel plan playback content when a pause operation is detected for the playback content; or it can display a pause control for the playback content of the travel plan and pause playback of the content in response to the control being triggered. This method allows the user to forcibly pause playback of the content so that they can view certain information in the playback content in detail.

[0130] In one embodiment, any of the travel planning playback contents being played back may include interactive content related to the travel service; in this regard, the second client may also display the related content of the travel service and / or execute the related instructions of the travel service in response to the interactive content being triggered. For details, please refer to the previous embodiment.

[0131] In one embodiment, the planning server can divide the travel planning playback content into multiple content segments and send the segmentation results to the second client; alternatively, the second client can perform the segmentation itself. Specifically, the planning server or the second client can divide the travel planning playback content into multiple content segments in various ways. For example, the multiple content segments can be determined based on segmentation information output by a travel planning model, or the travel planning playback content can be input into a content segmentation model and the segmentation results output by the model can be received. This will not be further described.

[0132] Figure 28 This is a schematic structural diagram of a device provided by an exemplary embodiment. Figure 28At the hardware level, the device includes a processor 2802, an internal bus 2804, a network interface 2806, a memory 2808, and a non-volatile memory 2810. Of course, it may also include hardware required for other functions. One or more embodiments of this specification can be implemented based on software, such as the processor 2802 reading the corresponding computer program from the non-volatile memory 2810 into the memory 2808 and then running it. Of course, in addition to software implementation, one or more embodiments of this specification do not exclude other implementation methods, such as logic devices or a combination of software and hardware, etc., that is, the execution subject of the following processing flow is not limited to each logic unit, but can also be hardware or logic devices.

[0133] Please refer to Figure 29 , the display device of travel planning content can be used for Figure 28 The device shown in the figure is used to implement the technical solution of this specification. The device is applied to a travel service platform and includes:

[0134] A persona determining unit 2901 is configured to determine, based on the received travel planning request, at least one virtual persona matching the travel planning request;

[0135] A planning processing unit 2902 is configured to call a travel planning model to perform travel planning processing and determine the thinking process of each virtual person in the travel planning processing;

[0136] The process display unit 2903 displays the thinking process corresponding to each virtual person through the first client of the travel service platform.

[0137] Optionally, the personality determination unit 2901 is specifically configured to:

[0138] All predefined avatars related to travel planning;

[0139] Alternatively, a virtual persona that matches the identified planning intent of the travel planning request.

[0140] Optional,

[0141] The virtual persona matching the travel planning request includes: a main persona and at least one branch persona;

[0142] The process display unit 2903 is specifically used for:

[0143] Displaying the main thinking process corresponding to the main character and the branch thinking process corresponding to each branch character on the first client;

[0144] The main thinking process includes a task splitting process and a task aggregation process; the task splitting process is used to split the main planning task corresponding to the travel planning request into one or more planning subtasks, and output the information of the branch persona corresponding to each planning subtask; the task aggregation process is used to aggregate the branch thinking results of each branch persona;

[0145] The branch thinking process corresponding to each branch persona is used to think about the planning subtasks assigned to the main persona.

[0146] Optional,

[0147] The branch personas include: at least one travel expert persona, and a budget analysis persona;

[0148] The planning subtasks corresponding to any travel expert persona include: querying the corresponding travel data through the travel data query logic consistent with the specific settings of the travel expert persona;

[0149] The planning subtask corresponding to the budget analysis persona includes: performing budget analysis on the travel data queried by each travel expert persona.

[0150] Optionally, the process display unit 2903 is specifically configured to:

[0151] When there is only one avatar matching the travel planning request, displaying a display area corresponding to the avatar in the planning process display interface on the first client, and displaying the thinking process corresponding to the avatar in the display area;

[0152] When there are multiple virtual characters that match the travel planning request, the display area corresponding to each virtual character is shown separately in the planning process display interface on the first client, and the thinking process corresponding to the corresponding virtual character is independently displayed in each display area.

[0153] Optionally, the process display unit 2903 is specifically used for at least one of the following:

[0154] Dynamically displaying the thinking content and / or thinking progress corresponding to any of the virtual characters in the display area corresponding to the virtual character;

[0155] Adjusting the size of the display area corresponding to any of the virtual characters according to the thinking progress corresponding to the virtual character, and displaying a corresponding amount of thinking content in the display area;

[0156] Displaying a portion of the thought process corresponding to any of the virtual characters in a display area corresponding to the virtual character; and, in response to a browsing trigger operation on the display area, switching the display content in the display area to another thought process corresponding to the virtual character;

[0157] When a thought process corresponding to any one of the virtual characters is displayed in the display area corresponding to the virtual character, in response to a trigger operation for viewing details of the display area, a floating window is generated on the first client, and the thought process corresponding to the any one of the virtual characters is displayed in the floating window;

[0158] When the thinking process corresponding to any virtual persona has not yet ended, the thinking process is displayed in the display area corresponding to the any virtual persona; when the thinking process corresponding to any virtual persona has ended, the thinking process is stopped from being displayed and corresponding thinking end prompt information is displayed in the display area corresponding to the any virtual persona.

[0159] Optionally, also include:

[0160] The result display unit 2904 is used to determine the travel planning result corresponding to the thinking process obtained by the travel planning process, and display the travel planning result through the first client.

[0161] Optionally, also include:

[0162] The partition display unit 2905 is used to display the thinking process in a thumbnail form and the travel planning result in a detailed form on the first client; in response to a viewing trigger operation for the thinking process, the thinking process is switched from a thumbnail form to a detailed form, wherein when the thinking process is displayed in the detailed form, the first client is provided with selection controls corresponding to multiple virtual personalities and a shared display area for the multiple virtual personalities; and when the selection control corresponding to any virtual persona among the multiple virtual personalities is selected, the thinking process corresponding to the any virtual persona is displayed in the shared display area; and / or,

[0163] The data statistics unit 2906 is used to display the total amount of travel data queried during the thinking process on the first client after the thinking process is completed, and / or the number of valid travel data in the travel planning results filtered from the queried travel data.

[0164] Optionally, the system further includes a content playback unit 2907, configured to:

[0165] Recording the thought process and the travel planning results as corresponding travel planning playback content;

[0166] A trigger control corresponding to at least one recorded travel planning playback content is displayed on at least one second client of the travel service platform, and when the trigger control corresponding to any travel planning playback content is selected, any travel planning playback content is played back on the second client.

[0167] Please refer to Figure 30 , the playback device of travel planning content can be applied to Figure 28 The device shown in the figure is used to implement the technical solution of this specification. The device is applied to a travel service platform and includes:

[0168] The planning processing unit 3001 is used to call the travel planning model to perform travel planning processing according to the received travel planning request;

[0169] A content recording unit 3002 is configured to display the thought process of the travel planning process and the corresponding travel planning results on the first client of the travel service platform, and record the thought process and travel planning results as corresponding travel planning playback content;

[0170] The content playback unit 3003 is used to display the trigger control corresponding to at least one recorded travel planning playback content on at least one second client of the travel service platform, and to play back any travel planning playback content on the second client when the trigger control corresponding to any travel planning playback content is selected.

[0171] Optionally, the system further includes a content sharing unit 3004, configured to:

[0172] Displaying, on the first client, a sharing authorization control for the travel planning playback content recorded on the first client;

[0173] In response to the authorization confirmation operation for the sharing authorization control, the travel planning playback content recorded on the first client is shared on the travel service platform, so that the travel planning playback content is allowed to be played back on the second client.

[0174] Optionally, the at least one travel plan playback content includes at least one of the following:

[0175] The request content of the corresponding travel planning request meets the travel planning playback content of the preset request screening conditions; wherein the preset request screening conditions include at least one of the following: the request content is typical, the request content is representative, and the request content is interesting;

[0176] Travel planning playback content that meets preset recommendation conditions; wherein the preset recommendation conditions include at least one of the following: playback popularity reaches a preset level, travel planning playback content recorded within a preset time period, and travel planning playback content shared within a preset time period;

[0177] The request content and / or the included content of the corresponding travel planning request match the user feature data of the logged-in user on the second client and the travel planning playback content.

[0178] Optionally, the content playback unit 3003 is specifically configured to:

[0179] Displaying a selection control corresponding to at least one classification dimension on any second client;

[0180] In response to a selection control corresponding to any one of the classification dimensions being triggered, at least one travel planning playback content belonging to any of the classification dimensions is displayed; wherein each travel planning playback content belongs to one or more classification dimensions.

[0181] Optionally, the content playback unit 3003 is specifically configured to:

[0182] Displaying a plan playback interface on the second client, the plan playback interface including a first area for playing back any of the travel plan playback content, and a second area distinct from the first area; wherein, if any of the travel plan playback content is sequentially divided into a plurality of content segments, the second area is used to display playback controls corresponding to each content segment;

[0183] In response to a playback control corresponding to any content segment being triggered, playback of any travel planning playback content begins from the any content segment.

[0184] Optionally, a playback progress unit 3005 is further included, which is used to:

[0185] When displaying any content segment of any travel planning playback content, showing the corresponding playback progress at the playback control corresponding to the content segment; and / or,

[0186] In response to a progress adjustment operation on any playback control, the content segment corresponding to the playback control is played back, and the playback progress of the content segment is controlled based on the progress adjustment operation.

[0187] Optionally, any of the travel planning playback contents being played back includes interactive content related to travel services; the apparatus further includes an interaction triggering unit 3006, configured to:

[0188] In response to the interactive content being triggered, content associated with the travel service is displayed and / or instructions associated with the travel service are executed.

[0189] Optionally, the content recording unit 3002 is specifically configured to:

[0190] Determining at least one target virtual persona corresponding to the travel planning request and a thought process of each target virtual persona in the travel planning process;

[0191] The thought process corresponding to each target virtual persona is displayed separately through the first client.

[0192] Optional,

[0193] The thinking process corresponding to any virtual persona includes at least one of the following: the thinking content corresponding to the virtual persona, the thinking progress corresponding to the virtual persona; and / or,

[0194] The thinking content corresponding to any virtual persona includes at least one of the following: travel data query logic consistent with any virtual persona, statistical information of travel data queried based on the travel data query logic, and detailed information of the queried travel data.

[0195] Based on the same concept as the above method, this specification also provides an electronic device, including: a processor; a memory for storing processor-executable instructions; wherein the processor implements the steps of the method described in any of the above embodiments by running the executable instructions.

[0196] Based on the same concept as the above method, this specification also provides a computer-readable storage medium on which computer instructions are stored. When the instructions are executed by a processor, the steps of the method described in any of the above embodiments are implemented.

[0197] Based on the same concept as the above method, this specification also provides a computer program product, including a computer program / instruction, which implements the steps of the method described in any of the above embodiments when executed by a processor.

Claims

1. A method for displaying travel planning content, characterized in that: Applied to a travel service platform, the method includes: Determining, based on the received travel planning request, at least one virtual persona matching the travel planning request; Invoking the travel planning model to perform travel planning processing, and determining the thinking process of each virtual person in the travel planning processing; The thinking process corresponding to each virtual persona is displayed respectively through the first client of the travel service platform.

2. The method according to claim 1, characterized in that The at least one virtual persona matching the travel planning request includes: All predefined avatars related to travel planning; Alternatively, a virtual persona that matches the identified planning intent of the travel planning request.

3. The method according to claim 1, characterized in that The virtual persona matching the travel planning request includes: a main persona and at least one branch persona; The first client of the travel service platform displays the thought process corresponding to each virtual persona, including: Displaying the main thinking process corresponding to the main character and the branch thinking process corresponding to each branch character on the first client; The main thinking process includes a task splitting process and a task aggregation process; the task splitting process is used to split the main planning task corresponding to the travel planning request into one or more planning subtasks, and output the information of the branch persona corresponding to each planning subtask; the task aggregation process is used to aggregate the branch thinking results of each branch persona; The branch thinking process corresponding to each branch persona is used to think about the planning subtasks assigned to the main persona.

4. The method according to claim 3, characterized in that The branch personas include: at least one travel expert persona, and a budget analysis persona; The planning subtasks corresponding to any travel expert persona include: querying the corresponding travel data through the travel data query logic consistent with the specific settings of the travel expert persona; The planning subtask corresponding to the budget analysis persona includes: performing budget analysis on the travel data queried by each travel expert persona.

5. The method according to claim 1, wherein The first client of the travel service platform displays the thought process corresponding to each virtual persona, including: When there is only one avatar matching the travel planning request, displaying a display area corresponding to the avatar in the planning process display interface on the first client, and displaying the thinking process corresponding to the avatar in the display area; When there are multiple virtual characters that match the travel planning request, the display area corresponding to each virtual character is shown separately in the planning process display interface on the first client, and the thinking process corresponding to the corresponding virtual character is independently displayed in each display area.

6. The method according to claim 5, characterized in that Displaying the thought process of any virtual character in the display area corresponding to the virtual character includes at least one of the following: Dynamically displaying the thinking content and / or thinking progress corresponding to any of the virtual characters in the display area corresponding to the virtual character; Adjusting the size of the display area corresponding to any of the virtual characters according to the thinking progress corresponding to the virtual character, and displaying a corresponding amount of thinking content in the display area; Displaying a portion of the thinking process corresponding to any of the virtual characters in the display area corresponding to the virtual character; and, in response to a browsing trigger operation on the display area, switching the display content in the display area to other thinking processes corresponding to the virtual persona; When a thought process corresponding to any one of the virtual characters is displayed in the display area corresponding to the virtual character, in response to a trigger operation for viewing details of the display area, a floating window is generated on the first client, and the thought process corresponding to the any one of the virtual characters is displayed in the floating window; When the thinking process corresponding to any virtual persona has not yet ended, the thinking process is displayed in the display area corresponding to the any virtual persona; when the thinking process corresponding to any virtual persona has ended, the thinking process is stopped from being displayed and corresponding thinking end prompt information is displayed in the display area corresponding to the any virtual persona.

7. The method according to claim 1, characterized in that Also includes: Determine a travel planning result corresponding to the thinking process obtained by the travel planning process, and display the travel planning result through the first client.

8. The method according to claim 7, characterized in that Also includes: On the first client, the thinking process is displayed in a thumbnail form, and the travel planning result is displayed in a detailed form; In response to a viewing trigger operation for the thought process, the thought process is switched from a thumbnail form to a detailed form, wherein when the thought process is displayed in the detailed form, the first client is provided with selection controls corresponding to a plurality of virtual personalities and a shared display area for the plurality of virtual personalities; and, when the selection control corresponding to any one of the plurality of virtual personalities is selected, the thought process corresponding to the any one of the virtual personalities is displayed in the shared display area; and / or, After the thinking process is completed, the total amount of travel data queried during the thinking process and / or the amount of valid travel data in the travel planning result filtered from the queried travel data are displayed on the first client.

9. The method according to claim 7, characterized in that Also includes: Recording the thought process and the travel planning results as corresponding travel planning playback content; A trigger control corresponding to at least one recorded travel planning playback content is displayed on at least one second client of the travel service platform, and when the trigger control corresponding to any travel planning playback content is selected, any travel planning playback content is played back on the second client.

10. A computer program product, characterized in that The method comprises a computer program / instruction, which, when executed by a processor, implements the steps of the method according to any one of claims 1 to 9.

Images