Travel management method and computer program

Abstract

The embodiment of the invention provides a travel management method and a computer program, the method is applied to a travel service platform, the travel service platform comprises a travel service client and a travel background server, and the method comprises the following steps: acquiring travel application information that a traveler applies for a travel; after the business travel application information passes the approval, recommending a business travel scheme to the traveler based on a recommendation strategy bound by the traveler under the condition that the business travel standard is met; wherein the recommendation strategy bound by the traveler is pre-selected from multiple types of recommendation strategies, the recommendation strategy is associated with multiple recommendation factors influencing the recommended travel scheme and the weight of each recommendation factor, the weights of at least one recommendation factor of different types of recommendation strategies are different, the recommendation factors at least comprise price factors, and the price factors are different from the price factors. The weight of the price factor is used for controlling the cost control degree of the recommended travel scheme for the travel cost. According to the embodiment of the invention, the flexibility and refinement degree of travel management can be improved.

Description

TECHNICAL FIELD

[0001] Embodiments of the present application relate to the technical field of computer Internet, in particular to a business travel management method and a computer program. BACKGROUND

[0002] Business travel, referred to as business travel, refers to the temporary travel of employees dispatched by an organization such as an enterprise, an institution or a government agency to a place away from home for work or official business and handling related business affairs. As a high-frequency expenditure scenario in the organization and operation of an organization such as an enterprise, business travel management not only relates to the efficiency of employees in performing their duties, but also directly affects the operating cost and management standard of the organization.

[0003] In the traditional business travel management scenario, an organization such as an enterprise usually passively intervenes in the behavior of a traveler booking travel services according to business travel standards (such as passively controlling mainly in terms of budget limit, reimbursement specification and approval process), lacks dynamic control means, and is difficult to meet the different business travel management needs of different organizations. Therefore, how to provide a technical solution to enable an organization to conduct personalized business travel management based on its business travel management needs and improve the flexibility of business travel management has become a technical problem that technicians in the field need to solve. SUMMARY

[0004] Therefore, embodiments of the present application provide a business travel management method and a computer program to improve the flexibility of business travel management.

[0005] To achieve the above-mentioned purpose, embodiments of the present application provide the following technical solutions.

[0006] In a first aspect, embodiments of the present application provide a business travel management method applied to a travel service platform, the travel service platform comprising a travel service client and a travel background server, and the method comprising:

[0007] obtaining business travel application information of a traveler applying for business travel;

[0008] when the business travel application information is approved, and in the case of meeting the business travel standards, recommending a business travel plan for the traveler based on a recommendation strategy bound by the traveler;

[0009] wherein the recommendation strategy bound by the traveler is preselected from multiple types of recommendation strategies, the recommendation strategy is associated with multiple recommendation factors affecting the recommended business travel plan and the weight of each recommendation factor, the weight of at least one recommendation factor of different types of recommendation strategies is different, and the recommendation factor at least includes a price factor, and the weight of the price factor is used to control the cost control degree of the recommended business travel plan on the business travel cost.

[0010] In a second aspect, an embodiment of the present application provides a computer program, comprising computer instructions, which, when executed by a processor, implement the business trip management method according to the first aspect.

[0011] The business trip management method provided by the embodiment of the present application obtains the business trip application information of a trip person applying for a business trip, and when the business trip application information is approved, a business trip solution is recommended for the trip person based on a recommendation strategy bound by the trip person in the case of meeting the business trip standard, wherein the recommendation strategy bound by the trip person is preselected from multiple types of recommendation strategies, the recommendation strategy is associated with multiple recommendation factors affecting the recommended business trip solution and the weight of each recommendation factor, and the weight of at least one recommendation factor in different types of recommendation strategies is different. That is, an organization such as an enterprise can preselect a recommendation strategy bound for an employee from multiple types of recommendation strategies based on the business trip management needs of the organization, so that the recommendation strategy bound for the employee can be dynamically adjusted, which makes the recommendation process of the business trip solution no longer a static and unified processing logic, but can be dynamically adjusted according to different recommendation strategies bound for different trip persons, and each type of recommendation strategy is associated with multiple recommendation factors affecting the recommended business trip solution and the weight of each recommendation factor, and the weight of at least one recommendation factor in different types of recommendation strategies is different, so that different types of recommendation strategies have different focuses when recommending the business trip solution, so that the organization such as an enterprise can adjust the type of the recommendation strategy bound for the employee to meet the business trip management needs of the organization for different focuses (such as business trip cost, employee travel experience, etc.), even to meet the business trip management needs of the organization for multiple focuses, and improve the flexibility of business trip management.

[0012] At the same time, since the recommendation factor at least includes the price factor, the organization such as an enterprise can control the degree of emphasis of different types of recommendation strategies on the business trip cost by configuring the weight of the price factor of different types of recommendation strategies, thereby meeting the management and control needs of the organization for the business trip cost. It can be seen that, by using the solution provided by the embodiment of the present application, the organization such as an enterprise can flexibly configure the recommendation strategy bound for the employee according to the business trip management needs of the organization, thereby realizing the differentiated management in the dimension of the business trip cost when recommending the business trip solution, and improving the flexibility and fine degree of business trip management. BRIEF DESCRIPTION OF DRAWINGS

[0013] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiments or the prior art description will be briefly introduced as follows. Obviously, the drawings in the following description are only embodiments of the present application, and those skilled in the art can obtain other drawings according to the provided drawings without creating any inventive labor.

[0014] Figure 1AA block diagram of a travel service platform.

[0015] Figure 1A A stage example diagram of a business travel management method provided by an embodiment of the present application.

[0016] Figure 2 A flowchart of a business travel management method provided by an embodiment of the present application.

[0017] Figure 3A A partial display content example diagram of a recommendation strategy configuration page provided by an embodiment of the present application.

[0018] Figure 3B An example diagram of a strategy setting page.

[0019] Figure 4 A recommendation flowchart of a hotel accommodation service provided by an embodiment of the present application.

[0020] Figure 5 An example diagram of a business travel standard priority configuration page.

[0021] Figure 6A A flowchart of a business travel necessity verification provided by an embodiment of the present application.

[0022] Figure 6B A flowchart of a business travel period suggestion provided by an embodiment of the present application.

[0023] Figure 7A An example diagram of a business travel solution recommendation page.

[0024] Figure 7B A hierarchical architecture diagram of a recommendation system of a travel service platform.

[0025] Figure 8A An example diagram of a board interface.

[0026] Figure 8B An example diagram of a recommendation result page. DETAILED DESCRIPTION

[0027] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of protection of the present application.

[0028] It should be noted that the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties, and the collection, use and processing of related data need to comply with relevant laws, regulations and standards of countries and regions, and provide corresponding operation portal for user to choose authorization or refusal.

[0029] With the continuous improvement of the digital level of travel management of enterprise and other organization units, AI (artificial intelligence) technology is gradually changing from an auxiliary tool to a core engine in travel management. Based on the big data and intelligent recommendation driven travel management mode, enterprises and other organization units expect to ensure the travel experience of employees while optimizing travel costs and improving management efficiency. However, in actual implementation, the traditional travel management method relying on fixed difference standards (difference standards refer to travel standards) and approval rules is difficult to adapt to the diversified management demands of enterprises and other organization units, such as being difficult to achieve fine travel cost control, and being difficult to balance the satisfaction of employees and compliance. Therefore, how to support enterprises and other organization units to flexibly realize travel management according to their own needs has become a challenge that enterprises and other organization units need to cope with in the travel management scenario.

[0030] To cope with the above challenges, enterprises and other organization units can choose to rely on a travel service platform and combine AI model capabilities to build an intelligent travel management mechanism. Among them, the travel service platform refers to an online service platform that provides travel services, such as an OTP (Online Travel Platform) platform, an OTA (Online Travel Agency) platform, etc. In the travel scenario, the travel service platform can serve as a basic carrier for travel service resource aggregation and intelligent recommendation, access travel approval, travel standard management and intelligent recommendation engine functions, and provide technical support for the whole process of travel management of enterprises and other organization units. On this basis, the improved travel management method is provided in the embodiments of the present application to realize the differentiated travel management and flexible regulation and control needs of enterprises and other organization units.

[0031] As an optional implementation, the trip management method provided by the embodiment of the present application can be applied to a travel service platform. The travel service platform referred to in the embodiment of the present application can be understood as a service platform that provides travel service booking, travel management and travel service support. The travel service platform can provide comprehensive travel service booking for individual users, such as travel service booking of traffic tickets, hotel accommodation, scenic spot tickets, etc. The travel service platform can also provide travel service booking, cost control, compliance review and other integrated travel management support for enterprise customers. That is, the travel service platform can be a travel service system for individual consumers, or a professional travel management system for organizations such as enterprises.

[0032] Specifically, from the perspective of service mode, the travel service platform can provide travel product services for individual consumers (such as online booking services for individual users of traffic tickets, hotel accommodation, etc.) and / or provide centralized travel management services for organizations such as enterprises. In actual application, the form and service mode of the travel service platform can be diversified; on the one hand, the travel service platform can mainly serve individual consumers, and integrate travel management functions on this basis, thereby providing travel management services for organizations such as enterprises with travel management needs. On the other hand, the travel service platform can also be a professional service platform with travel management as the core function, mainly serving organizations such as enterprises.

[0033] Optionally, Figure 1A An exemplary block diagram of the travel service platform is shown as follows. Figure 1A As shown, the travel service platform can include a travel service client 101 and a travel background server 102;

[0034] The travel service client 101 is mainly for users and is regarded as a front-end interface provided by the travel service platform for user interaction, including but not limited to: APP (Application) provided by the travel service platform, web pages, mini programs, etc. The travel service client can be loaded on the user device of the user, and the user device includes but is not limited to a smart phone, a tablet computer, a PC (Personal Computer), a smart TV, etc.

[0035] The travel background server 102 is a background server provided by the travel service platform, responsible for processing the requests of the travel service client and performing data management;

[0036] The travel service client 101 and the travel background server 102 jointly constitute the travel service platform, that is, the travel service client as the front end and the travel background server as the back end cooperate with each other to ensure that the user can use the services provided by the travel service platform.

[0037] The trip management method provided by the embodiments of the present application mainly involves three stages, and an example is shown in the following table. Figure 1B An example of a stage example diagram of the trip management method provided by the embodiments of the present application is shown, and the three stages mainly include: a trip application approval stage 110, a recommendation stage 120, and a post-trip evaluation stage 130.

[0038] The trip application approval stage refers to the stage of approving the trip application after the trip initiator initiates the trip application, mainly involving trip necessity verification, trip period suggestion and other processes, to ensure the necessity and compliance of the trip behavior. For example, the trip necessity verification, such as in combination with the trip background (such as business opportunity information), judges whether the trip is necessary or not, such as when the business opportunity success rate is low, it is suggested to use remote meeting instead; the trip period suggestion, such as analyzing the travel service price trend and off-season situation of the destination, suggesting the trip initiator to avoid the high price period to improve the overall cost performance.

[0039] The recommendation stage refers to the stage of recommending a trip solution for the trip initiator after the trip application initiated by the trip initiator is approved. The embodiments of the present application meet the flexible regulation and control requirements of the trip management of an enterprise or other organization unit, and are mainly realized in this stage. Specifically, the embodiments of the present application support pre-binding a recommendation strategy for the trip initiator from multiple types of recommendation strategies, and the weights of at least one recommendation factor associated with different types of recommendation strategies are different. The recommendation factors associated with the recommendation strategy are, for example, price factor, preference factor, efficiency factor, etc. Therefore, by pre-selecting and configuring the recommendation strategy for the trip initiator, the enterprise or other organization unit can flexibly balance the management goals of trip cost control, employee satisfaction and travel efficiency, and realize differentiated recommendation and management of trip solutions.

[0040] Further, the recommendation stage can also involve: booking opportunity prediction, trip standard control, cost optimization control, AI model target strategy and other processes; wherein the booking opportunity prediction refers to using a price trend prediction algorithm to judge the future price fluctuation of the travel service, and recommending an economical travel service booking time node for the user; the trip standard control refers to dividing the trip standard upper limit and the recommended trip standard, and recommending a compliant and cost-saving travel service when recommending a trip solution for the trip initiator; the cost optimization control refers to giving a saving solution and corresponding guidance from the trip itinerary, and even providing a saving incentive (such as points or honor mark, etc.) to encourage the trip initiator to adopt the saving solution; the AI model target strategy refers to that the enterprise or other organization unit can set an AI target such as a saving target (such as a trip cost reduction ratio), so that the AI model continuously optimizes the recommendation logic of the recommended solution according to the AI target.

[0041] The post-trip evaluation stage refers to, after the trip is completed and the actual expenses are formed, based on the behavior data, the trip rationality audit, cost analysis and supply chain optimization suggestions are carried out, mainly involving trip behavior analysis, purchase optimization suggestions and the like. Among them, the trip behavior analysis mainly judges whether the trip person has high-cost preference (such as repeatedly selecting over-standard hotels and other over-standard travel services), unnecessary business trips, frequent ticket service changes and other unreasonable behaviors, and outputs a behavior board for the administrator to review; the administrator refers to the administrator responsible for trip management in an enterprise or the like. The purchase optimization suggestion mainly identifies airlines, hotels and other travel service suppliers with price negotiation potential based on historical high-frequency trip routes and booking data, so as to propose agreement purchase suggestions, help enterprises and the like obtain preferential prices or exclusive policies of suppliers, and improve the overall input-output results.

[0042] As an optional implementation, Figure 2 An exemplary flowchart of a trip management method provided by the embodiments of the application is shown, which can be applied to a travel service platform such as an OTP, an OTA platform and the like. Referring to Figure 2 The trip management method provided by the embodiments of the application can include the following steps.

[0043] In step S210, trip application information of a trip person applying for a trip is obtained.

[0044] The trip application information refers to a structured application data set submitted by the trip person for trip approval and itinerary arrangement, such as a trip application form. The trip application information records the key elements of the trip planned by the trip person, and constitutes the basis input of the trip management process. For example, the trip application information includes but is not limited to the following contents: trip person information (such as the trip person's name, employee number, department, position and the like), trip purpose (such as visiting customers, attending meetings and the like), departure and destination, proposed departure time and return time and the like. Further, the trip application information can further include at least one item of trip information, one item of trip information can correspond to one trip task of the trip person (such as a meeting, a visit, a training and the like), and one item of trip information can include the item name, the item type, the item location, the item time and the like. Further, the trip application information can further include at least one of the following contents: trip expense attribution information (such as the department to which the expense reimbursement belongs), itinerary requirements or preferences (such as flight time period, hotel accommodation preferences and the like) and the like.

[0045] Specifically, the trip person can fill in and submit the trip application information through the front-end trip application interface of the travel service client of the travel service platform, so that the travel back-end server of the travel service platform receives and structures the storage, which is used as the basis for subsequent trip application approval, trip plan recommendation and post-trip audit.

[0046] Step S220, when the travel application information is approved, in the case of meeting the travel standard, a travel plan is recommended for the traveler based on the recommendation strategy bound to the traveler; wherein the recommendation strategy bound to the traveler is pre-selected from multiple types of recommendation strategies, the recommendation strategy is associated with multiple recommendation factors affecting the recommended travel plan and the weight of each recommendation factor, the weight of at least one recommendation factor of different types of recommendation strategies is different, and the recommendation factor at least includes a price factor, the weight of the price factor is used to control the cost control degree of the recommended travel plan to the travel cost.

[0047] The travel application information submitted by the traveler can be approved through the travel application approval stage, and when the travel application information is approved, it enters the recommendation stage; so in the recommendation stage, the travel service platform can recommend a travel plan for the traveler based on the recommendation strategy bound to the traveler in the case of meeting the travel standard. Specifically, the travel back-end server of the travel service platform can determine the travel plan recommended for the traveler and display it on the travel plan recommendation page of the travel service client of the travel service platform. Among them, the travel plan is a travel service combination recommended for the traveler based on travel application information, travel standards and recommendation strategies and other factors, covering transportation ticket services, hotel accommodation services and optional additional services (such as insurance, pick-up and drop-off machines, etc.) during the trip, thereby guiding or assisting the traveler to make travel service reservation decisions.

[0048] In the embodiments of the present application, the recommendation strategy is a strategy for guiding the recommendation logic of the travel plan based on the travel management goals of the enterprise or other organizational units; the recommendation strategy is associated with multiple recommendation factors affecting the recommendation result of the travel plan and the weight of each recommendation factor, and the weight of at least one recommendation factor of different types of recommendation strategies is different, that is, the focus of different types of recommendation strategies is different when recommending travel plans. Specifically, the multiple recommendation factors associated with the recommendation strategy are the dimensions of consideration affecting the recommendation decision of the travel plan, such as price factor, preference factor, efficiency factor, etc., and in each type of recommendation strategy, each recommendation factor is set with a corresponding weight, indicating the influence degree of each recommendation factor on the recommendation result of the travel plan in this type of recommendation strategy.

[0049] In an optional implementation, the recommendation factors associated with the recommendation strategy can at least include a price factor, which is a consideration dimension associated with the recommendation strategy for considering the economy (such as travel cost) of the travel scheme; and the weight of the price factor is used to control the cost control degree of the recommended travel scheme on the travel cost, i.e., the cost sensitivity degree of the travel scheme in the recommendation process, such as the weight of the price factor associated with the recommendation strategy is negatively correlated with the travel cost of the travel scheme recommended by the recommendation strategy, for example, the higher the weight value of the price factor (such as the higher the price weight level of the price factor), the lower the travel cost of the recommended travel scheme, thereby improving the control strength of the travel cost.

[0050] In a further optional implementation, the recommendation factors associated with the recommendation strategy can also include a preference factor, which is a consideration dimension associated with the recommendation strategy for considering the travel preference of the traveler matched by the travel scheme, such as the travel preference of the traveler for the transportation ticket service (for example, the preference for the type of transportation tool, the travel period, etc.), the preference for the hotel accommodation service (for example, the preference for the hotel brand, the hotel type, the accommodation geographical location, etc.), etc.; and the weight of the preference factor is used to control the preference consideration degree of the recommended travel scheme on the travel preference of the traveler, i.e., the personalized adaptation sensitivity degree of the travel scheme on the travel preference in the recommendation process, such as the weight of the preference factor associated with the recommendation strategy is positively correlated with the matching degree of the travel preference of the travel scheme recommended by the recommendation strategy, for example, the higher the weight value of the preference factor (such as the higher the preference weight level of the preference factor), the closer the recommended travel scheme to the preference of the traveler, thereby improving the adoption rate and user satisfaction of the recommended scheme.

[0051] In a further optional implementation, the recommendation factors associated with the recommendation strategy can also include an efficiency factor, which is a consideration dimension associated with the recommendation strategy for considering the timeliness of the travel scheme, such as whether the departure and arrival times are convenient, whether the overall travel time is short, whether it is a direct route, etc.; and the weight of the efficiency factor is used to control the trade-off degree of the recommended travel scheme in terms of timeliness, i.e., the time efficiency sensitivity degree of the travel scheme in the recommendation process. The weight of the efficiency factor associated with the recommendation strategy is positively correlated with the superior or inferior performance of the travel scheme in terms of travel time efficiency, for example, the higher the weight value of the efficiency factor (such as the higher the efficiency weight level of the efficiency factor), the more the recommended travel scheme tends to select a scheme with reasonable departure time and shorter time consumption, thereby improving the travel efficiency and reducing the time cost.

[0052] It should be noted that the price factor, the preference factor and the efficiency factor mentioned above are only exemplary recommendation factors. In actual implementation, multiple recommendation factors associated with a recommendation strategy can be flexibly configured according to the management needs of an organization unit, different sets of recommendation factors can be associated with different types of recommendation strategies, and different weights can be set for each recommendation factor to adapt to different emphases of different types of recommendation strategies.

[0053] As an optional implementation, the multiple types of recommendation strategies provided by the embodiments of the present application can include at least two types as follows:

[0054] The balanced recommendation strategy is used to comprehensively consider the travel cost, travel efficiency and travel preference of a traveler to recommend a travel plan at least in the case of meeting the travel standard. That is, the balanced recommendation strategy comprehensively considers the travel cost, travel efficiency and travel preference in the weight setting of multiple recommendation factors associated therewith, so that the weight distribution of each recommendation factor is relatively balanced. For example, the balanced recommendation strategy can be applied to a management scenario in which it is desired to maintain a balance between travel cost control, employee satisfaction and travel efficiency.

[0055] The saving recommendation strategy is used to preferentially consider the travel cost to recommend a travel plan at least in the case of meeting the travel standard. That is, the saving recommendation strategy sets the weight of the price factor to the highest in the weight setting of multiple recommendation factors associated therewith, and preferentially selects a low-cost travel service resource. For example, the saving recommendation strategy can be applied to a scenario in which an organization unit such as an enterprise needs to strengthen travel cost control.

[0056] The efficiency recommendation strategy is used to preferentially consider the travel efficiency to recommend a travel plan at least in the case of meeting the travel standard. That is, the efficiency recommendation strategy increases the weight of the efficiency factor in the weight setting of multiple recommendation factors associated therewith, and aims to recommend a travel plan with the shortest travel time and the most compact travel schedule. For example, the efficiency recommendation strategy can be applied to a scenario in which a travel task is urgent and time is required.

[0057] The comfort recommendation strategy is used to preferentially consider the travel preference and travel time of a traveler to recommend a travel plan at least in the case of meeting the travel standard. That is, the efficiency recommendation strategy increases the weight of the preference factor in the weight setting of multiple recommendation factors associated therewith, and pays more attention to the travel preference of a traveler when recommending a travel plan. For example, the comfort recommendation strategy can be applied to a scenario in which a high-ranking official travels or employee travel experience is improved.

[0058] It should be noted that the types of the above-mentioned recommendation strategies are only optional examples, and the types of the multiple recommendation strategies provided by the embodiments of the present application can be flexibly configured according to the travel management goals of the enterprise or other organization units. Specifically, the focus of different types of recommendation strategies corresponds to the associated recommendation factors and the weight settings of the recommendation factors, so that the enterprise or other organization units can flexibly adjust different weight combinations of multiple recommendation factors according to actual travel management demands (for example, more emphasis on travel cost saving, more emphasis on employee experience, or emphasis on travel efficiency guarantee, etc.), so as to form multiple types of recommendation strategies that adapt to different focuses. Of course, in order to meet the differentiated management needs of the enterprise or other organization units, the embodiments of the present application can also support the pre-configuration of multiple types of recommendation strategies by the travel service platform, and support the subsequent update or expansion of the types of recommendation strategies. Further, the administrator of the enterprise or other organization unit can bind different recommendation strategies for different employee groups based on specific differentiated travel management needs from the multiple types of recommendation strategies, or dynamically switch the recommendation strategies bound for all employees at different times, to realize the phased adjustment of the recommendation strategies, so as to realize the individualized and differentiated travel management of the enterprise or other organization unit, and improve the flexibility of travel management.

[0059] In a more specific optional implementation, taking the multiple recommendation factors associated with the recommendation strategy as an example, including the price factor and the preference factor, the price factor can have multiple price weight levels corresponding to multiple cost control degrees of the travel cost, and the preference factor can have multiple preference weight levels corresponding to multiple preference consideration degrees of the travel preference; so that by adjusting the price weight level of the price factor and the preference weight level of the preference factor associated with the recommendation strategy, the recommendation strategy can be adjusted in the emphasis of travel cost and travel preference, for example, increasing the price weight level of the price factor and reducing the preference weight level of the preference factor, the corresponding recommendation strategy preferentially recommends the travel scheme with lower travel cost, to realize the intensive control of travel cost; on the contrary, if the preference weight level of the preference factor is increased and the price weight level of the price factor is reduced, the corresponding recommendation strategy pays more attention to the matching of the travel preference of the traveler, and preferentially recommends the travel scheme that matches the travel preference of the traveler, so as to improve the employee satisfaction and travel experience. Further, different combinations of the preference weight level of the price factor and the preference weight level of the preference factor can correspond to different types of recommendation strategies, to form multiple types of recommendation strategies with different focuses.

[0060] For example, for the balanced recommendation strategy, the price weight level of the price factor can correspond to an intermediate degree of cost control, and the preference weight level of the preference factor can correspond to an intermediate degree of preference consideration; that is, the price weight level of the price factor associated with the balanced recommendation strategy is set to an intermediate degree, so as to maintain moderate control over travel costs, and at the same time, the preference weight level of the associated preference factor is also set to an intermediate degree, so as to achieve moderate matching of travel preferences, thereby achieving a balance between cost, efficiency and experience.

[0061] For example, for the saving recommendation strategy, the price weight level of the price factor corresponds to the highest degree of cost control, and the preference weight level of the preference factor corresponds to the lowest degree of preference consideration; that is, the price weight level of the price factor associated with the saving recommendation strategy is set to the highest degree, so as to preferentially select travel services with lower travel costs when recommending travel plans, thereby achieving enhanced cost control, and at the same time, the preference weight level of the associated preference factor is set to the lowest degree, that is, the degree of consideration of the individual preferences of employees is the lowest, and more emphasis is placed on the economy of the travel plan.

[0062] For example, for the comfortable recommendation strategy, the price weight level of the price factor corresponds to the lowest degree of cost control, and the preference weight level of the preference factor corresponds to the highest degree of preference consideration; that is, the price weight level of the price factor associated with the comfortable recommendation strategy is set to the lowest degree, so as to relax the constraint on travel costs, and at the same time, the preference weight level of the associated preference factor is set to the highest degree, so that the recommended travel plan is more in line with the preferences and experience of employees.

[0063] For further understanding, Figure 3A An example is shown in FIG. 1, which shows a part of the display content of a recommendation strategy configuration page provided by an embodiment of the present application. The recommendation strategy configuration page can be a sub-page of a strategy center page in a travel management interface provided by a travel service platform, wherein the travel management interface is logged in and operated by an administrator, and the strategy center page in the travel management interface is used by the administrator to configure various recommendation strategies and bind the recommendation strategies for employees. As shown in FIG. 1, the recommendation strategy configuration page can display multiple types of configured recommendation strategies, such as balanced, saving, efficient and comfortable. Figure 3A As shown in FIG. 1, the recommendation strategy configuration page can display the control dimensions of the price factor and the preference factor. Specifically, the price factor has a control slider, and taking the 5-level price weight level as an example, the control slider of the price factor has marking points corresponding to the 5-level price weight level, which are, in turn:

[0064] Compliance is OK, corresponding to the lowest price weight level (such as the first level of price weight level), indicating that the control requirement of travel cost is the lowest, as long as it does not exceed the travel standard.

[0065] slight saving, corresponding to a lower price weight level (e.g. level 2 price weight level), such as a control wish for a slight lower travel cost than the travel standard;

[0066] appropriate consideration, corresponding to an intermediate degree (i.e. moderate) price weight level (e.g. level 3 price weight level), representing a control of travel cost without affecting travel efficiency or travel experience;

[0067] low price consideration, corresponding to a higher price weight level (e.g. level 4 price weight level), representing a preference for recommending lower price travel services;

[0068] low price priority, corresponding to a highest price weight level (e.g. level 5 price weight level), representing a maximum compression of travel cost;

[0069] In Figure 3A In an example, the control slider of the price factor is slid from left to right, representing a price sensitivity from low to high, corresponding to a gradually increasing weight of the price factor.

[0070] The preference factor has a control slider, with a plurality of preference weight levels, taking level 5 preference weight level as an example, the control slider of the preference factor has marked points corresponding to the level 5 preference weight level, in turn:

[0071] preference priority, corresponding to a highest preference weight level (e.g. level 5 preference weight level), representing a priority matching of the travel preference of the traveler;

[0072] preference consideration, corresponding to a higher preference weight level (e.g. level 4 preference weight level), representing a moderate bias towards the travel preference of the traveler;

[0073] appropriate consideration, corresponding to an intermediate degree (i.e. moderate) preference weight level (e.g. level 3 preference weight level), thereby taking into account the travel preference of the traveler when recommending travel solutions.

[0074] slight consideration, corresponding to a lower preference weight level (e.g. level 2 preference weight level), representing that the travel preference of the traveler is considered when recommending travel solutions, but not a major consideration factor;

[0075] no consideration, corresponding to a lowest preference weight level (e.g. level 1 preference weight level), representing that the travel preference of the traveler is not considered when recommending travel solutions;

[0076] In Figure 3A In an example, the control slider of the preference factor is slid from left to right, representing a degree of preference consideration from high to low, corresponding to a gradually decreasing weight of the preference factor.

[0077] Thus, the administrator can set the weight level of each recommendation factor by dragging the slider on the control slider of each recommendation factor (such as the slider on the control slider of the price factor, the slider on the control slider of the preference factor), and then combine the weight levels of each recommendation factor to form a type of recommendation strategy corresponding to a certain focus, such as the Figure 3A In the example of the balanced recommendation strategy, the sliders on the control sliders of the price factor and the preference factor are both at the middle position, and the landing positions of the sliders on the control sliders of the price factor and the preference factor corresponding to other types of recommendation strategies can be combined in the same way as described above. Therefore, the details are not described here.

[0078] In a further optional implementation, the administrator can bind a recommendation strategy for the employees of an organization such as an enterprise through a strategy setting page. For example, for any type of recommendation strategy in multiple types of recommendation strategies, the administrator can select the employees to which the recommendation strategy is bound (such as binding all employees or selecting specific employees for binding), and further, the administrator can select the employees excluded from the recommendation strategy. An example of the strategy setting page is shown in the example of Figure 3B An example of the strategy setting page is shown in the example of Figure 3B As shown, the strategy setting page is a sub-page of the strategy center page, and is used at least for binding a recommendation strategy for an employee. The strategy center page is a sub-page of the travel management interface. Specifically, the travel management interface can have sub-pages such as a recommendation result page, a strategy center page, a user behavior analysis page, and a purchase optimization suggestion page, which can be accessed through the corresponding entry of the travel management interface.

[0079] In the strategy setting page, the strategy setting page includes a strategy list area for displaying and managing multiple types of recommendation strategies, and an operation control for setting the scope of employees to which the recommendation strategy applies. Specifically, the strategy list area of the strategy setting page can display multiple types of configured recommendation strategies (such as the balanced recommendation strategy, the saving recommendation strategy, etc.) in the form of cards. One card corresponds to one type of configured recommendation strategy, and the content displayed by one card includes the control information of the recommendation factors corresponding to the recommendation strategy, which is used to indicate the weight configuration state of each recommendation factor in the recommendation strategy, such as the weight levels of the price factor and the preference factor. One card can also be associated with an operation control for the scope of the recommendation strategy, which supports the administrator to select the scope of employees to which the recommendation strategy applies, such as all employees or selected employees. If selected employees are selected, the administrator can enter an employee selection interface to specify the employee group or specific employees to which the recommendation strategy applies. Meanwhile, the present application also supports the employees excluded from the recommendation strategy, such as through an exclusion entry, the administrator can set the employees excluded from the recommendation strategy, that is, the employees to which the recommendation strategy does not apply, so as to flexibly adjust the scope of the personnel bound by the recommendation strategy.

[0080] It can be seen that the travel service platform can provide the administrator with flexible recommendation strategy adaptation capability, and the administrator of an organization such as an enterprise can dynamically adjust the employees and employee groups to which the recommendation strategy is applicable according to the travel management target, realize directional configuration and differential management of the recommendation strategy, and perform differential recommendation of the travel plan to improve the fine and intelligent level of travel management.

[0081] The travel management method provided in the embodiments of the present application can obtain travel application information of a traveler applying for travel, and when the travel application information is approved, a travel plan is recommended for the traveler based on a recommendation strategy bound by the traveler in the case of meeting a travel standard. The recommendation strategy bound by the traveler is preselected from multiple types of recommendation strategies, the recommendation strategy is associated with multiple recommendation factors affecting the recommended travel plan and the weight of each recommendation factor, and the weight of at least one recommendation factor of different types of recommendation strategies is different. That is, an organization such as an enterprise can preselect a recommendation strategy bound for an employee from multiple types of recommendation strategies based on its own travel management needs, so that the recommendation strategy bound for the employee can be dynamically adjusted, which makes the recommendation process of the travel plan no longer a static and unified processing logic, but can be dynamically adjusted according to different recommendation strategies bound by different travelers, and each type of recommendation strategy is associated with multiple recommendation factors affecting the recommended travel plan and the weight of each recommendation factor, and the weight of at least one recommendation factor in different types of recommendation strategies is different, so that different types of recommendation strategies have different focuses when recommending the travel plan, so that the organization such as an enterprise can adjust the type of the recommendation strategy bound for the employee to meet the travel management needs of the organization such as an enterprise for different focuses (such as travel cost, employee travel experience, and different focuses), or even meet the travel management needs of the organization such as an enterprise for multiple focuses, and improve the flexibility of travel management.

[0082] At the same time, since the recommendation factor at least includes the price factor, the organization such as an enterprise can control the degree of emphasis of different types of recommendation strategies on the travel cost by configuring the weight of the price factor of different types of recommendation strategies, thereby meeting the management and control needs of the organization such as an enterprise for the travel cost. It can be seen that, by using the scheme provided in the embodiments of the present application, the organization such as an enterprise can flexibly configure the recommendation strategy bound for the employee according to its own travel management needs, thereby realizing differential management in the dimension of the travel cost and the like when recommending the travel plan, and improving the flexibility and fine degree of travel management.

[0083] In a further optional implementation, after determining the recommended strategy of the trip person binding, the embodiment of the present application recommends a travel plan for the trip person based on the recommended strategy of the trip person binding under the premise of meeting the travel standard. The travel standard is used to perform compliance constraints in the process of recommending the travel plan, to ensure that the recommended travel service meets the travel task requirements of the trip person while meeting the travel management rules of the enterprise or other organization unit. As an optional implementation, the travel standard can include, but is not limited to, the following two types of control standards:

[0084] The travel standard upper limit represents the highest allowable standard set for the price of the travel service (such as a ticket, a hotel, a train ticket, etc.), the cabin class of the ticket, and the like in the compliance level. The travel service exceeding the travel standard upper limit is considered to be out of standard and is not recommended or needs to go through a special approval process.

[0085] The recommended travel standard is a controllable part of the travel standard, mainly used for travel cost control, and represents a price interval that is preferentially considered in the process of recommending the travel plan. The recommended travel standard can be set to be lower than the travel standard upper limit, so that the recommended travel standard can guide the cost tendency of the travel plan recommendation result, and recommend a travel service with a more optimal travel cost under the premise of not reaching the travel standard upper limit.

[0086] In a further optional implementation, various travel services (such as tickets, train tickets, hotel accommodation services, and various travel services) can have corresponding travel standards, specifically involving the travel standard upper limit and the recommended travel standard of various travel services. For example, the travel standard of transportation ticket services such as tickets and train tickets includes the travel standard upper limit and the recommended travel standard, and the travel standard of hotel accommodation services includes the travel standard upper limit and the recommended travel standard. Further, a complete travel plan containing multiple types of travel services can also correspond to the setting of the travel standard, and involve the travel standard upper limit and the recommended travel standard.

[0087] In a further optional implementation, to meet the differentiated management goals of different organizational levels and post categories, the embodiment of the present application can also support the configuration of multi-dimensional travel standards based on multiple personnel grouping dimensions. Specifically, the personnel grouping dimension refers to an organizational basis for classifying employees of an enterprise or other organization unit. Different personnel grouping dimensions can reflect the organizational identity and management attributes of employees, facilitating the hierarchical configuration and personalized application of travel standards. For example, the multiple personnel grouping dimensions include, but are not limited to: an employee dimension, corresponding to the configuration of travel standards for specific employees; a job level dimension, corresponding to the configuration of travel standards for different employee job levels such as ordinary employees, supervisors, managers, and executives; a role dimension, corresponding to the configuration of travel standards for different employee positions such as sales positions, research and development positions, procurement positions, and project managers; a department dimension, corresponding to the configuration of travel standards for different departments such as the marketing department, the finance department, the legal department, and the technology center; and an all-employee dimension, corresponding to the configuration of travel standards for all employees.

[0088] Therefore, the embodiments of the present application support multi-dimensional configuration of the travel standard based on multiple personnel grouping dimensions, i.e., support configuration of the travel standard under each personnel grouping dimension or a combination of at least two personnel grouping dimensions, so that the enterprise can implement differentiated travel standards for different groups of employees. As a further optional implementation, one personnel grouping dimension or one combination of personnel grouping dimensions (one combination of personnel grouping dimensions can include at least two personnel grouping dimensions, such as the combination of the R&D department and the manager) can correspond to a one-dimensional travel standard, and the one-dimensional travel standard can involve the travel standard upper limit and the recommended travel standard; further, the one-dimensional travel standard can set the travel standard for different types of travel services (such as tickets, train tickets, hotel accommodation services, and various travel services) respectively, and involve the travel standard upper limit and the recommended travel standard.

[0089] It can be seen that the embodiments of the present application support multi-dimensional and hierarchical configuration of the travel standard, and the travel standard not only includes the travel standard upper limit, but also includes the recommended travel standard for travel cost control; and the travel standard can be set individually for different types of travel services, or can be set for the entire travel plan as a whole. At the same time, the embodiments of the present application can also support flexible configuration of the travel standard based on multiple personnel grouping dimensions such as employees, job levels, roles, departments, etc., to achieve fine-grained management of the travel standard.

[0090] In a further optional implementation, the recommended travel standard as the travel standard for controlling travel costs can be configured by the administrator as needed whether to enable in a specific recommended strategy. Specifically, the travel service platform can support configuring whether to enable the recommended travel standard in the recommended travel plan process for each type of recommended strategy (such as the balanced recommended strategy, the saving recommended strategy, etc.). If it is configured to enable, the strategy setting page can automatically display a prompt message indicating that the recommended strategy applies the recommended travel standard when recommending the travel plan. Further, the enablement of the recommended travel standard can also be refined to specific travel service types, such as transportation ticket services (air tickets, train tickets, etc.), hotel accommodation services, etc., and the administrator can specify the travel service types to which the recommended travel standard applies under the recommended strategy. For example, if the recommended travel standard for hotel accommodation services is enabled for the saving recommended strategy in the strategy setting page, the corresponding opening prompt can be displayed in the page, for example, the opening state of the recommended travel standard under the hotel accommodation recommendation is opened. Figure 3B Of course, in addition to hotel accommodation services, whether the recommended travel standard is opened can also be applicable to other travel service types, including but not limited to transportation ticket services such as air tickets and train tickets, which can be flexibly configured by the administrator according to needs.

[0091] As an optional implementation, in the recommendation strategy based on the travel person binding, when recommending the travel scheme for the travel person, if the recommended travel standard is enabled, the recommended travel standard also needs to be used as the travel standard when recommending the travel scheme. That is, in the recommendation strategy based on the travel person binding, when recommending the travel scheme for the travel person, if the recommended travel standard is enabled, the travel service platform also needs to judge whether the recommended travel standard meets the preset use condition, so that in the case that the recommended travel standard meets the use condition, the recommended travel standard is used as the travel standard when actually recommending the travel scheme, so as to control the travel cost. Specifically, the use condition of the recommended travel standard includes but is not limited to that the market average price of the travel service corresponding to the recommended travel standard is lower than or equal to the recommended travel standard, and the recommended travel standard is lower than the upper limit of the travel standard of the corresponding travel service. When the above use condition is met, the travel service platform can use the recommended travel standard as the execution standard when recommending the travel scheme, for screening and filtering the corresponding travel service resources, so as to improve the cost control strength; on the contrary, if the above use condition is not met, in order to avoid the recommended travel service resources being excessively compressed or unavailable, the travel service platform can use the upper limit of the travel standard as the execution standard when recommending the travel scheme, so as to ensure the availability and compliance of the recommended travel service.

[0092] That is, the recommended travel standard can have two layers of mechanisms of enabling and using, so as to balance flexibility and recommendation feasibility. The enabling of the recommended travel standard is actively set by the administrator in the strategy configuration stage whether to introduce the recommended travel standard as a cost control means into the recommendation process, which is used to reflect the travel management intention; and the use of the recommended travel standard is dynamically judged whether the recommended travel standard has actual executability in the specific recommendation process according to the current market situation and the travel standard rule. Thus, through the decoupling design of the enabling and use of the recommended travel standard, the cost control means can be flexibly configured according to the travel management target of the enterprise or other organization unit, and the upper limit of the travel standard can be returned in time when the market price fluctuates or the resources are scarce, so as to avoid the recommendation failure or the damage of the employee experience caused by too strict recommendation conditions, so as to realize the balance between flexible regulation and stable operation.

[0093] For the convenience of understanding, taking the travel service including the hotel accommodation service as an example, the recommendation process of the hotel accommodation service under the condition that the recommended travel standard meets the use condition is introduced below. That is, at this time, the recommended travel standard of the hotel accommodation service in the recommendation strategy of the travel person binding is enabled, and in the process of recommending the travel scheme for the travel person based on the recommendation strategy of the travel person binding, the recommended travel standard of the hotel accommodation service meets the use condition, for example, the market average price of the hotel accommodation service is lower than or equal to the recommended travel standard of the hotel accommodation service, and the recommended travel standard of the hotel accommodation service is lower than the upper limit of the travel standard of the hotel accommodation service. In an optional implementation, Figure 4An exemplary flowchart of a recommendation process of a hotel accommodation service provided by an embodiment of the present application is shown in FIG. 1, and the flowchart is described with reference to FIG. 1. Figure 4 The flowchart can include the following steps.

[0094] At step S410, when the recommendation travel standard meets the use condition, a recall price range of the hotel accommodation service is determined according to the recommendation travel standard.

[0095] The recall price range of the hotel accommodation service refers to a price interval used to filter (i.e., recall) the hotel accommodation service meeting the condition from the travel service platform when recommending the hotel accommodation service to the traveler, based on the used travel standard and in combination with certain upper and lower tolerance rules. In an optional implementation, the recall price range of the hotel accommodation service can be set as:

[0096] not lower than max(recommendation travel standard-lower tolerance, preset lower limit price) and not higher than min(recommendation travel standard+upper tolerance, upper limit of travel standard).

[0097] The lower tolerance refers to a downward price floating value set to prevent the recommended hotel accommodation service from being too low to cause poor user experience, for example, 250 yuan. The upper tolerance refers to an upward price floating value set, for example, 80 yuan, but the floating price must not exceed the upper limit of the travel standard. The preset lower limit price refers to the lowest accommodation reference line set, for example, 100 yuan, to prevent the recommended hotel accommodation service from being lower than the preset lower limit price to affect the employee experience or fail to guarantee the accommodation quality.

[0098] Therefore, when the recall price range of the hotel accommodation service is determined, the maximum value is selected from the result of subtracting the lower tolerance from the recommendation travel standard and the preset lower limit price as the lower limit value of the recall price range. The minimum value is selected from the result of adding the upper tolerance to the recommendation travel standard and the upper limit of the travel standard as the upper limit value of the recall price range. Further, the recall price range is formed by the price interval not lower than the lower limit value and not higher than the upper limit value.

[0099] At step S411, the hotel accommodation service in the recall price range is determined in the set geographical range of the desired accommodation location, and the hotel accommodation service is filtered based on the filtering rule of the hotel accommodation service and the hotel filtering rule to obtain the recalled hotel accommodation service.

[0100] In an optional implementation, the embodiments of the present application can filter the hotel accommodation services whose prices fall within the recalled price range according to the filtering rules of the hotel accommodation services and the hotel filtering rules in a set geographical range centered on the desired accommodation location (such as a range of 3 kilometers with the desired accommodation location as the center), so as to obtain the recalled hotel accommodation services. Optionally, the filtering rules of the hotel accommodation services, such as filtering the hotel accommodation services that are not bookable (such as filtering the hotel accommodation services that have no room inventory), filtering the hotel accommodation services that have resource control (such as filtering the hotel accommodation services that cannot be normally booked, cannot be deployed, or are banned due to enterprise travel management policies or travel service platform supply chain management restrictions), and the like; the hotel filtering rules, such as filtering the hotels whose hotel scores are lower than a hotel score threshold (such as filtering the hotels whose hotel scores are lower than 3 points), filtering the hotels whose hotel types are set removal types (such as filtering the hotels whose hotel types are hostel types, and the like), and further filtering the hotels that do not support enterprise contracts, cannot issue invoices, and the like that do not meet enterprise compliance requirements.

[0101] In an optional implementation, the desired accommodation location can be selected by the traveler or set by default, including but not limited to any of the following: the location of the travel matter, the central location of the destination city, the location of the arrival transportation station, and the like.

[0102] Step S412, for each of the recalled hotel accommodation services, scoring and ranking the hotel accommodation services according to the prices of the hotel accommodation services, and determining the hotel accommodation services recommended for the traveler based on the scoring and ranking results.

[0103] In a specific implementation, the score of any hotel accommodation service is obtained based on the current price of the hotel accommodation service, the highest price in the recalled hotel accommodation services, and the reference price intermediate value, the closeness between the current price of the hotel accommodation service and the reference price intermediate value is positively correlated with the score of the hotel accommodation service; the reference price intermediate value is an intermediate value corresponding to the average price and the highest price of the recalled hotel accommodation services, and if the reference price intermediate value is higher than the recommended travel standard, the reference price intermediate value is adjusted to the recommended travel standard minus a preset value.

[0104] That is, the core of the scoring logic of the hotel accommodation service is to compare the current price of the hotel accommodation service with the reference price median value, and determine the score of the hotel accommodation service according to the closeness of the two. For example, the score formula of a hotel accommodation service can be expressed as: score = 4-((X-μ) / (M-μ))x2; where X is the current price of the hotel accommodation service, M is the highest price in the recalled hotel accommodation service, and μ is the reference price median value; the reference price median value μ can be expressed as the average price of the recalled hotel accommodation service and the median value of the highest price, such as dividing by 2, and further, if the calculated reference price median value μ is greater than the recommended travel standard, the calculated reference price median value μ needs to be adjusted, specifically μ = recommended travel standard - preset value, such as reducing the recommended travel standard by 50 yuan, to achieve cost control adjustment. Thus, the closer the current price of the hotel accommodation service to the reference price median value μ, the higher the score of the hotel accommodation service, and the closer the current price of the hotel accommodation service to the highest price M, the closer the score of the hotel accommodation service to 0.

[0105] Thus, the scores of the recalled hotel accommodation services can be used for sorting, and then the top several hotel accommodation services or the one with the highest score can be selected to recommend to the traveler.

[0106] Further, if the recall result of the hotel accommodation service is empty (such as no recalled hotel accommodation service), or the scores of the recalled hotel accommodation services are all low, the upper limit of the travel standard can be used to re-recommend the hotel accommodation service according to the fallback mechanism; of course, using the upper limit of the travel standard to recommend the hotel accommodation service can also be triggered when the recommended travel standard is not enabled or the use condition is not met.

[0107] In an optional implementation, the process of using the upper limit of the travel standard to recommend the hotel accommodation service can be: in the set geographical range of the desired accommodation location, filtering the hotel accommodation services within the upper limit of the travel standard, and performing hotel accommodation service filtering and hotel filtering to obtain the recalled hotel accommodation services; scoring and sorting the recalled hotel accommodation services, and the specific way can be: scoring according to the closeness of the current price of the hotel accommodation service to the reference price median value μ, and the related introduction of the scoring can refer to the foregoing description, and the difference here is that if the calculated reference price median value μ is greater than the upper limit of the travel standard, the reference price median value μ is adjusted to μ = upper limit of travel standard - preset value, such as reducing the upper limit of the travel standard by 50 yuan.

[0108] In a further optional implementation, when the travel standards are divided into multi-dimensional travel standards according to the multi-person grouping dimension, the trip person can be bound with at least two-dimensional travel standards, such as a trip person being bound with travel standards of multiple dimensions at the same time (such as a trip person belonging to both a technical department and being a manager, and thus being bound with travel standards of both the technical department and the manager), or multiple trip persons being bound with different travel standards at the same time, in which case the application embodiment further needs to confirm the travel standard used when recommending the travel scheme for the trip person, and specifically, the travel standard used when recommending the travel scheme for the trip person can be confirmed through the priority order configured by the administrator on the travel standard priority configuration page, as shown in an example. Figure 5 An example of a travel standard priority configuration page is shown in an example diagram, which can be referred to. The travel standard priority configuration page can be entered through the priority configuration option of the policy setting page shown in Figure 3B

[0109] Specifically, for the first case, i.e., the case of a same trip person being bound with at least two-dimensional travel standards at the same time, in combination with Figure 5 shown, the application embodiment can perform the following processing:

[0110] If the trip person is bound with at least two-dimensional travel standards, the travel standard used when recommending the travel scheme is selected from the at least two-dimensional travel standards according to the priority of the at least two-dimensional travel standards bound by the trip person, wherein the priority of the multi-dimensional travel standards is related to the priority of the multi-person grouping dimension; for example, a same employee can belong to multiple groups (such as belonging to both a manager and a sales department), in which case the travel standard selection is performed according to the binding accuracy priority, such as the priority order being: employee > job level > role > department > all employees, and thus if a same employee hits both the travel standard of the employee dimension and the travel standard of the department dimension, the travel standard of the employee dimension with higher priority is used preferentially;

[0111] ​Further, if the at least two travel standards bound by the traveler have the same priority, the highest travel standard or the lowest travel standard is selected from the at least two travel standards according to a pre-configured difference standard processing option, the difference standard processing option is divided into a high processing option and a low processing option, the high processing option corresponds to selection of the highest travel standard, and the low processing option corresponds to selection of the lowest travel standard; that is, if the at least two travel standards bound by the traveler have the same priority (for example, the same employee hits two department-level travel standards), the travel standard selection is performed according to the high processing or low processing principle, the high processing is to select a more relaxed travel standard (for example, a higher cabin of an air ticket, a higher hotel price, etc.), that is, the highest travel standard in the at least two travel standards bound by the traveler, and the low processing is to select a more stringent travel standard (for example, an economy class of an air ticket, a lower hotel price, etc.), that is, the lowest travel standard in the at least two travel standards bound by the traveler; as an example, the high processing can be configured by default to facilitate guarantee of travel experience, and of course, the specific configuration can be set and adjusted by an administrator.

[0112] For the second case, that is, when multiple travelers travel at the same time and bind different travel standards, the embodiment of the application can perform the following processing, as shown in the figure. Figure 5

[0113] If the travel application information has multiple travelers (that is, multiple travelers travel at the same time, and at this time, traffic ticket service or hotel accommodation service is booked for multiple people by using a same order), the highest travel standard or the lowest travel standard is selected from the travel standards bound by the multiple travelers based on a pre-configured difference standard processing option, and is used as a travel standard used when a recommended travel plan is recommended; wherein, the traffic ticket service and the hotel accommodation service have corresponding difference standard processing options, and the difference standard processing option is divided into a high processing option and a low processing option. That is, the traffic journey (such as an air ticket, a train ticket, and the like) and the hotel journey (that is, the hotel accommodation service) are respectively provided with the high processing option and the low processing option, so as to configure, by an administrator, a travel standard processing mode under the traffic journey and a travel standard processing mode under the hotel journey; as an example, the traffic journey can be configured by default with the low processing, and the hotel journey can be configured by default with the high processing, and of course, the specific configuration can be set and adjusted by an administrator.

[0114] ​The mechanism for recommending travel standards of the embodiments of the present application aims to recommend more cost-advantageous travel services, thereby achieving fine control of travel costs by enterprises. Therefore, the travel standards are set as a travel standard upper limit and a recommended travel standard. The travel standard upper limit is an explicitly set upper limit of travel standard compliance, which is used to rigidly constrain the highest price or level of travel services selected by employees. The recommended travel standard corresponds to a travel standard saving target set by an enterprise or other organization unit. For example, an enterprise sets the travel standard upper limit of hotel accommodation services as 700 yuan, which is the highest upper limit price of hotel accommodation services. At the same time, the enterprise internally requires saving travel costs, and therefore, based on the travel standard saving target, the recommended travel standard of hotel accommodation services is set as 400 yuan. Thus, in the actual recommendation process, the travel standard upper limit filtering mechanism can be used to ensure that all recommended travel services meet the travel standard upper limit, i.e., travel service options that exceed the travel standard upper limit are not displayed. For example, the embodiments of the present application can mark travel services in the travel service resource library that exceed the corresponding travel standard upper limit as an out-of-standard state. Then, the recommended travel standard is used to preferentially recommend travel services with a price that meets the recommended travel standard, thereby achieving a tendency guide for travel cost control. Further, if the traveler adjusts the travel service in the recommended travel plan, and the price of the adjusted travel service exceeds the corresponding recommended travel standard, the out-of-standard reporting logic is triggered to prompt the traveler to fill in the out-of-standard reason. That is, when the traveler selects a travel service with a price higher than the recommended travel standard, the embodiments of the present application can further trigger a reporting process that exceeds the recommended travel standard, for example, requiring the traveler to explain the reason or submit additional approval, to strengthen the execution of cost control, and to flexibly adapt to the goals of travel compliance and cost reduction of enterprises or other organization units.

[0115] The above describes the recommendation strategy configuration, weight setting of recommendation factors, setting and use of travel standards (travel standard upper limit and recommended travel standard), and hotel accommodation service recommendation of the embodiments of the present application. Through the above content, the embodiments of the present application can achieve differentiated management in the dimensions of travel cost and the like when recommending travel plans based on differentiated recommendation strategies, thereby improving the flexibility and fine degree of travel management. Next, other content and implementation manners involved in the travel application approval stage, recommendation stage, and post-travel evaluation stage of the embodiments of the present application are described.

[0116] As an optional implementation, in the travel application approval stage, the embodiments of the present application can approve the travel application information, wherein the approval process can at least include travel necessity verification. Optionally, Figure 6A An exemplary flowchart of travel necessity verification provided by the embodiments of the present application is shown, referring to Figure 6A The flowchart can include the following steps.

[0117] Step S610, parsing the content of the travel application information.

[0118] As an optional implementation, the embodiments of the present application can automatically parse the travel application information submitted by the user by using NLP (Natural Language Processing) technology, such as a semantic understanding model relying on a BERT (Bidirectional Encoder Representations from Transformers) model and the like, to structurally extract and semantically classify the text content of the travel application information, so as to parse out content including but not limited to a departure place, a destination, a departure time, an arrival time, a travel purpose, a travel matter, a budget and the like. For example, the BERT model is a model using a bidirectional Transformer structure to understand the context semantics of words in a sentence at the same time.

[0119] Step S611, matching the parsed content with a preset travel necessity rule library to obtain a rule matching result, the travel necessity rule library including a rule set of travel necessary conditions and / or a rule set of travel non-essential conditions.

[0120] In an optional implementation, the embodiments of the present application can set a travel necessity rule library that can be dynamically maintained, which is configured by an enterprise or the like according to its own travel policy. The rule library can at least include a rule set of travel necessary conditions, which is a judgment standard set by the enterprise or the like to confirm the necessity of travel, that is, if the travel application information meets the rules in the rule set of travel necessary conditions, it means that the travel application has a clear purpose or is irreplaceable, so that the rule set of travel necessary conditions can provide a positive basis for travel approval and reduce the risk of misjudgment or interception of reasonable travel. For example, the content in the rule set of travel necessary conditions can be, for example: the business opportunity corresponding to the travel is in the later stage (such as having entered the contract negotiation or delivery stage); the object to be visited is a key customer or an important partner; the travel task is on-site technical support or bid evaluation, which cannot be completed remotely; the client strongly requests an offline meeting, and the online method is explicitly refused by the client, and the like.

[0121] Further, the travel necessity rule base can also include a rule set of travel non-essential conditions, which are judgment criteria set by an enterprise or the like for judging whether a travel behavior can not be essential or has alternative solutions, to provide a basis for identifying non-essential travel applications that can be replaced remotely; for example, the content in the rule set of travel non-essential conditions can be, for example: the object of the travel visit is the first contact, and the intention of cooperation has not been established; the success rate of business opportunities is low, and it is still in the early stage of information collection; the client's location supports online meetings, and the client accepts remote communication; there are other travel records for visiting the client or the project, and information can be shared without repeated travel, etc.

[0122] In an optional implementation, after the content of the travel application information is parsed, the parsed structured content can be matched with the travel necessity rule base rule by rule, and the hit result can be output to obtain a rule matching result. In a more specific implementation, the above rule matching process can be implemented based on a rule engine, which has the following capabilities: rule modeling and maintenance, supporting an enterprise or the like to flexibly configure the rule content in the travel necessity rule base based on dimensions such as job level, region, budget upper limit, project phase, etc.; semantic analysis adaptation, combining NLP and BERT model to perform semantic analysis on the travel application information and extract matchable content; dynamic update mechanism, supporting an enterprise or the like to dynamically adjust the rule content; matching result scoring mechanism: converting the hit positive rules (such as the rules in the hit rule set of travel essential conditions) and the hit negative rules (such as the rules in the hit rule set of travel non-essential conditions) into scoring factors for calculating the necessity risk score of the travel application.

[0123] Step S612, performing rationality analysis on the travel application information according to the historical travel applications related to the parsed content, to obtain a rationality analysis result.

[0124] Further, the present application embodiment can combine similar historical travel applications related to the current travel application, cross-compare, analyze whether there are repeated travel applications, mergable itineraries, or excessive frequent visits, etc. to obtain the rationality analysis result of the travel application information, for example, if the client has been communicated online recently and there is no new matter, it is determined that the current travel application is an unreasonable travel application that can be communicated online. Specifically, the present application embodiment can combine the existing travel record data of an enterprise or the like and use a machine learning comparison model to form a rationality judgment result of the current travel application.

[0125] Step S613, combining the rule matching result and the rationality analysis result to generate a necessity risk score of the travel application information.

[0126] Step S614, if the necessity risk score exceeds a preset risk score threshold, marking the travel application information as an abnormal travel application and outputting risk reminder information, the risk reminder information being accompanied by a risk level corresponding to the necessity risk score.

[0127] Step S615, if the necessity risk score does not exceed the preset risk score threshold, determining that the travel necessity verification of the travel application information is passed.

[0128] Based on the rule matching result and the rationality analysis result described above, the embodiment of the present application can determine the necessity risk score of the travel application information, such as weighting and integrating the rule matching result and the rationality analysis result to calculate the necessity risk score. Specifically, if the travel application information hits multiple rules in the rule set of the non-necessity condition of travel, and there is no rationality support, the necessity risk score is higher; if the travel application information hits the rule set of the necessity condition of travel, and / or there is rationality support, the necessity risk score is lower. Further, the necessity risk score can be divided into multiple risk levels, for example, risk levels 1 to 5, and the highest risk level is 5.

[0129] Therefore, if the necessity risk score of the travel application information is higher than the preset risk score threshold, the travel application information is marked as an abnormal travel application, i.e. no necessity of travel, and the risk reminder information with the risk level is generated and automatically pushed to the approver; if the score does not exceed the risk score threshold, it is determined that the travel necessity verification of the travel application information is passed.

[0130] As an optional implementation, during the travel application approval stage, the embodiment of the present application can suggest the travel period of the travel application, so as to intelligently recommend a travel period (i.e. travel time period) with better cost performance and higher compliance for the traveler. Optionally, Figure 6B An exemplary flow chart of the travel period suggestion provided by the embodiment of the present application is shown, referring to Figure 6B The flow can include the following steps.

[0131] Step S620, acquiring real-time price data of travel services related to travel, and combining dynamic influencing factors affecting the price of travel services to predict the price fluctuation trend corresponding to multiple candidate travel periods.

[0132] In an optional implementation, based on the current time point, the embodiment of the application can capture actual price data of each travel service that can be booked through an interface, such as actual price data of each hotel accommodation service in the travel destination that can be booked, actual price data of each train ticket from the departure place to the destination that can be booked, actual price data of each air ticket that can be booked, and the like. These data can be regarded as prices of each travel service that are open to display in a future period of time (such as 7 days or 14 days in the future), reflecting the instant supply and demand state of the travel service at the current time point.

[0133] The dynamic influencing factor that affects the price of the travel service refers to an external or market variable that can cause fluctuations in the price of the travel service such as a flight or a hotel accommodation in a future period of time, and has time sensitivity and dynamic changeability, and mainly includes but is not limited to: holiday information such as statutory holidays, arrangements for taking leave, weekends, and the like that can cause the price of the travel service to rise; large event information in the destination city such as conference and exhibition, concert, and the like that can cause the hotel and transportation service resources in the destination city to be in short supply; weather factors such as typhoon, rainstorm warning, and the like that can affect the flight and train and indirectly affect the price and supply of the transportation ticket; macroeconomic factors such as an increase in fuel surcharges, changes in airline policies, and the like.

[0134] Therefore, based on the real-time price data of the travel service, the embodiment of the application can further integrate the above dynamic influencing factors to construct a price prediction model to predict the price fluctuation trend of the travel service corresponding to a plurality of candidate travel periods. For example, by calling the API interface of the airline, the hotel accommodation, the train ticket, and the like, the listing price data of each travel service corresponding to the destination city is collected to form a real-time price sequence; and the holiday library, the weather forecast, the local event information of the destination, the macroeconomic changes, and the like are introduced as time-related external variables as the dynamic context for modeling; thereby, in the modeling process, the real-time price data is taken as an observation input, and the dynamic influencing factor is taken as a variable to participate in the modeling, and a price prediction trend curve corresponding to a plurality of candidate travel periods in the future is output, that is, a predicted price fluctuation trend corresponding to a plurality of candidate travel periods, to provide a quantitative basis for the recommendation of the recommended travel period.

[0135] In step S621, the travel compliance of each candidate travel period is judged according to the travel compliance rule, and the travel compliance result of each candidate travel period is obtained.

[0136] In an optional implementation, the travel compliance rule is a set of judgment basis for checking the reasonableness of travel time, which is set in advance by an organization such as an enterprise, so as to avoid potential unreasonable, illegal or high-risk travel time periods. For example, based on the travel compliance rule, the compliance of each candidate travel period is judged, including but not limited to the following: avoiding non-travelable time periods, such as the destination being in a major conference or exhibition period; avoiding time periods before and after holidays, such as the travel date being close to the National Day, Spring Festival, Tomb-Sweeping Day, etc. statutory holidays, or in the vicinity of the holiday; avoiding departure / arrival at night or non-working time, such as early morning flights, night hotel check-in, etc. time that does not comply with the company's travel policy; avoiding time periods restricted by the organization for travel, such as financial settlement period, concentrated meeting period, internal training period, etc. time periods that are not recommended for arranging travel. Therefore, the above rules are applied to each candidate travel period, and the travel compliance result of each candidate travel period is output as a reference index for recommending a recommended travel period together with the price fluctuation trend.

[0137] In step S622, a recommended travel period is recommended according to the price fluctuation trend and the travel compliance result of each candidate travel period; wherein when the emergency information related to travel is detected, the recommended travel period is re-recommended; and the predicted travel cost and the travel compliance result of each candidate travel period are also displayed in the visualization interface in chronological order.

[0138] In an optional implementation, the embodiments of the present application can comprehensively evaluate each candidate travel period based on two core dimensions (i.e. price fluctuation trend and travel compliance result of candidate travel period), and then recommend a recommended travel period. To realize comprehensive evaluation, as an optional implementation, the embodiments of the present application can use a weighted ranking algorithm to score and rank each candidate travel period, such as scoring the price fluctuation trend result, for example, based on the relative low degree of the predicted price of the candidate travel period among all candidate travel periods, to determine the score of the candidate travel period corresponding to the price fluctuation trend; scoring the travel compliance result of the candidate travel period, for example, if the candidate travel period is completely compliant, the score is full score, if the candidate travel period is optional but needs to be approved, the score is intermediate score, and if the candidate travel period is not compliant, the score is zero; further, the weighted ranking algorithm can be used to calculate the comprehensive score of each candidate travel period by combining the weight parameters of the price fluctuation trend and the travel compliance result; and the candidate travel periods are ranked in descending order of comprehensive score, and the candidate travel period with the highest comprehensive score is recommended as the recommended travel period.

[0139] In a further optional implementation, if the embodiments of the present application detect that there is an emergency related to the trip (such as a weather warning, a sudden public event, a large-scale cancellation of flights / trains, a traffic blockade, etc.) during the recommended trip period, the recommended trip period during which the emergency occurs can be excluded, the price fluctuation trend and the travel compliance result are re-predicted, and then the recommended trip period is updated. In a further optional implementation, the embodiments of the present application can display the predicted travel cost and travel compliance result of multiple candidate trip periods in the form of an interactive timeline chart, etc., so that the traveler can clearly compare the price change trend and the travel compliance risk of different trip periods, thereby improving the understandability and trustworthiness of the recommended trip period.

[0140] In an optional implementation, the embodiments of the present application can also predict the booking price and recommend the booking timing of the travel service in the recommendation stage. Specifically, a hybrid prediction model (such as an LSTM model and a random forest model) is used to predict the future price trend of the travel service (such as an air ticket, a hotel accommodation service, a train ticket, etc.) and intelligently recommend the booking time of the travel service, thereby automatically triggering a prompt or a price locking operation. In a specific implementation, the embodiments of the present application can use a hybrid prediction model to predict the future price trend of the travel service. When the price of the travel service in a certain period is lower than a price threshold, the period is taken as the recommended booking timing and a booking reminder or a price locking process is automatically triggered. The hybrid prediction model includes an LSTM (Long Short-Term Memory) model and a random forest model. The LSTM model is used to learn the change rule of the price of the travel service over time, and the random forest model is used to mine the influence of discrete factors in the historical booking data on the price of the travel service.

[0141] Specifically, the LSTM model is used to capture the time series characteristics of the travel service price, i.e., to learn the trend of the price change over time, such as the gradual increase of the flight price before departure, the fluctuation of the hotel price before and after the holiday, etc.; the random forest model is used to analyze the discrete influencing factors that are not time series, such as the influence of holidays, supplier promotions, weather changes, regional large-scale activities, etc. on the price, and the discrete influencing factors are nonlinear and irregular, but they have an effect on the price mutation. Further, the training data of the hybrid prediction model can be derived from historical transaction data, such as flight / hotel price records of a specific airline or OTA platform; at the same time, to support the training of the hybrid prediction model, the application embodiments can supplement external information such as network hotspot events and local events at the destination, to improve the model's early warning ability for sudden price increases; further, by using the trained hybrid prediction model, the context such as the prediction date, destination, and travel service type can be input, and then the future price curve of the travel service can be predicted. At the same time, the application embodiments can set the price threshold of the travel service according to the business travel standards of the organization, the travel preferences of the traveler, and the market average price fluctuation; when the prediction result shows that the price of the travel service in a certain period is lower than the price threshold, it is considered that this period is a good value-for-money period, and it is suitable for booking the recommended booking opportunity; at this time, the application embodiments can automatically push the booking reminder to the traveler or the administrator, or perform the price locking processing (i.e., lock the current price of the travel service), so as to quickly enter the booking execution phase.

[0142] In further optional implementations, if the recommendation strategy bound by the traveler contains the cost-saving recommendation strategy, the application embodiments can also perform a guiding process in the recommendation phase to guide the traveler to select the travel scheme corresponding to the cost-saving recommendation strategy. The guiding process aims to guide the traveler to adopt the travel scheme with better cost, so as to achieve controllable optimization of travel cost.

[0143] Specifically, the traveler may bind one type of recommendation strategy, or may bind multiple types of recommendation strategies at the same time (for example, bind the cost-saving recommendation strategy and the balanced recommendation strategy at the same time). The application embodiments are not limited in terms of the specific configuration of the administrator when selecting the bound employees for each type of recommendation strategy; if the traveler binds multiple types of recommendation strategies at the same time, the traveler may receive travel schemes under multiple types of recommendation strategies at the same time. When the recommendation strategy bound by the traveler contains the cost-saving recommendation strategy, the travel scheme recommended for the traveler contains the travel scheme recommended under the cost-saving recommendation strategy, and at this time the application embodiments can perform a guiding process to promote the traveler to adopt the travel scheme recommended under the cost-saving recommendation strategy. In optional implementations, the above guiding process includes but is not limited to:

[0144] In the travel plan recommendation page, the travel plan corresponding to the saving recommendation strategy is highlighted, and the cost saving information relative to other travel plans is displayed; for example, in the travel plan recommendation page (the travel plan recommendation page displays the travel plan recommended for the traveler), a specific color label, saving prompt content, etc. are used to strengthen the visual effect of the saving travel plan (i.e. the travel plan recommended by the saving recommendation strategy); for example, Figure 7A An example of a travel plan recommendation page is shown, which can be referred to in combination with Figure 7A As shown, the recommended outbound flight is an early flight, and is highlighted, and the saving prompt content describing saving one day of accommodation cost is described in the recommended reason;

[0145] The saving incentive information is used to prompt the traveler that if the travel plan corresponding to the saving recommendation strategy is selected, an incentive reward is obtained; for example, if the saving travel plan is selected, the prompt of the incentive such as the point reward and the title reward can be displayed in the travel plan recommendation page; for example, in combination with Figure 7A As shown, the prompt of the reward is displayed under the item of the early departure of the recommended outbound flight, and the prompt of the first class reward is described in the recommended reason of the recommended outbound flight, that is, if the early outbound flight is adopted by the employee, one day of accommodation cost is saved for the enterprise, and the first class seat is given as an incentive;

[0146] When the travel plan corresponding to the saving recommendation strategy and the travel preference of the traveler have differences, but the saving amount exceeds the preset threshold, the corresponding guide information is displayed to guide the traveler to select the travel plan corresponding to the saving recommendation strategy; for example, when the travel preference of the traveler and the recommended saving travel plan have conflicts, but the saving amount of the saving travel plan exceeds the preset threshold (for example, 300 yuan), the prompt that the saving amount of the saving travel plan is large and the adoption is suggested is actively prompted.

[0147] When the traveler does not select the travel plan corresponding to the saving recommendation strategy, the intervention mechanism is triggered to prompt the traveler to fill in the reason for not adopting the saving recommendation strategy and / or initiate the approval process; that is, if the traveler finally does not adopt the saving travel plan, the reason needs to be filled in or the approval process is entered.

[0148] In an optional implementation, to further enhance the diversity and adaptability of the saving recommendation strategy, the application embodiment can also construct a recommendation plan library containing multiple cost optimization scenarios, for example: recommending a round trip on the same day to save accommodation costs, recommending a night train to save daytime working hours, recommending employees to carpool on the same train to reduce ground transportation costs, and recommending the use of public transportation to replace taxi travel, etc., so as to dynamically construct and recommend the saving travel plan to maximize the enterprise cost optimization goal.

[0149] In a further optional implementation, the embodiments of the present application can configure the objective function to guide the AI model to learn the recommendation strategy, so as to achieve the balance goal of cost saving and employee adoption rate set by the enterprise. As an optional implementation, the recommendation system used by the travel service platform to recommend the travel plan for the traveler can not be a hard-coded implementation based on fixed rules, but by configuring the strategy target of the AI model, so that the AI model learns from the historical data to form the recommendation logic of the recommendation strategy, that is, the recommendation logic of the recommendation strategy is learned by the AI model, so that the AI model recommends the corresponding travel plan based on the recommendation logic of the recommendation strategy.

[0150] Specifically, the AI model is trained based on the configured strategy target, and the strategy target is used to construct the objective function in the AI model training process. The strategy target can be regarded as a quantitative expectation index of the enterprise or other organization unit on the recommendation result of the travel plan, such as the travel cost reduction rate associated with the price factor (for example, hoping to save 15% cost each time compared with the average value), and the recommendation adoption rate of the travel plan associated with the preference factor (such as hoping that the employee adoption rate of the recommended plan reaches 80%). Therefore, the strategy target is converted into part of the training loss function of the AI model to guide the convergence direction of the model parameters of the AI model. In an optional implementation, the embodiments of the present application can provide a parameter configuration interface of the strategy target to allow the administrator to define the strategy target based on the specific needs of the enterprise or other organization unit, such as defining a 15% travel cost reduction and an 80% adoption rate. Further, in the specific training process of the AI model, the AI model can collect real-time selection feedback (such as accepting or rejecting the recommended travel plan, modifying the travel service in the travel plan, etc.) of the traveler for the recommended travel plan, and adjust the model parameters according to the selection feedback through reinforcement learning or regression strategy, to realize dynamic optimization of the recommendation logic, that is, the AI model can adjust the recommendation logic of the recommendation strategy in combination with the historical selection data of the travel plan.

[0151] For ease of understanding, Figure 7B An exemplary layered architecture diagram of the recommendation system of the travel service platform is shown as follows: Figure 7B As shown, the recommendation system is mainly divided into three levels: the bottom layer, the data layer, and the application layer.

[0152] Among them, the bottom layer is configured with multiple model modules, including: a basic large model for building basic capabilities, such as language understanding and general recommendation; a travel general model focusing on general recommendation strategies in the travel scenario; an enterprise management model mainly used for travel plan recommendation based on the travel cost control needs of the enterprise or other organization unit, such as travel plan recommendation under the saving type recommendation strategy, etc.; a user preference model and a colleague preference model mainly used for travel plan recommendation based on the personal preference or group preference of the traveler, such as travel plan recommendation under the comfort type recommendation strategy, etc.

[0153] The data layer provides data support for model training and execution, including: travel knowledge base, business trip knowledge base, used to build background knowledge for recommendation; user behavior, interaction data, used to train user preference model; enterprise management data, enterprise difference standard policy, used for enterprise and other organizational units to make recommendations for control based on business trip cost control requirements;

[0154] The application layer corresponds to the interaction mode with the user, such as the form of presenting the recommended business trip scheme to the user, including: multi-round dialogue interaction such as voice and text, dialogue box type interaction components, personalized entertainment content display, intelligent customer service, etc.

[0155] It should be noted that the underlying basic model and the business trip general model correspond to the default recommendation mechanism when there is no enterprise control or user preference, and the recommendation logic is directly generated based on the pre-trained large model. The user preference model and the colleague preference model correspond to the personalized business trip scheme recommendation that meets the preferences based on the user historical selection data and colleague booking behavior. The enterprise control model, enterprise control data, and enterprise difference standard policy correspond to the business trip scheme recommendation that prioritizes meeting the enterprise demand control when the enterprise sets clear control requirements (such as business trip cost control requirements). At this time, the administrator can dynamically adjust the strategy target and the weight of each recommendation factor, or even adjust the weight of the preference factor to 0, so as to completely recommend the business trip scheme according to the business trip cost control requirements of the enterprise.

[0156] In a further optional implementation, the embodiment of the present application can introduce a business trip rationality audit mechanism in the post-trip evaluation stage, focusing on high-spending preference behavior and frequent budget-exceeding behavior groups, to identify potential non-rational business trip behavior. Specifically, the embodiment of the present application can at least perform business trip rationality audit on the completed business trips of the concerned groups, where the concerned groups include high-spending preference behavior travelers and / or travelers who frequently submit budget-exceeding business trip applications, and the high-spending preference behavior travelers include: travelers who frequently choose travel services with higher prices than recommended travel services.

[0157] In an optional implementation, the embodiment of the present application can classify traveler behavior into groups and mark the concerned groups, including two types of groups: high-spending preference behavior travelers, such as travelers who still frequently choose travel services with higher prices in the case of saving options in the recommended business trip scheme; and travelers who frequently submit budget-exceeding business trip applications, such as personnel with poor budget compliance and high approval burden.

[0158] Therefore, at least the travel rationality of the concerned group can be audited, and when the travel rationality is audited, analysis and processing can be performed in multiple audit dimensions. Optionally, the multiple audit dimensions can at least include abnormal travel behavior identification. The abnormal travel behavior identification is to identify and analyze whether there is abnormal travel behavior. The abnormal travel behavior includes at least one of the following: repeated travel behavior; multiple selection of non-agreement suppliers or travel services exceeding the travel standard without explanation; selection of detours or non-direct transportation ticket services; multiple changes of travel services; multiple adjacent booking of travel services; travel service overage without use; offline change of transportation ticket services. Correspondingly, the analysis of abnormal travel behavior can include at least one of the following: repeated travel behavior identification, such as detecting multiple travel applications of a user to the same destination within a short period of time, and judging whether there is a behavior of combining trips; over-standard preference identification, such as identifying whether the user has multiple behaviors of bypassing enterprise agreement suppliers or selecting travel service resources higher than the standard (for example, hotels with higher prices, non-agreement flights, etc.) without explanation; path efficiency verification, such as analyzing whether the user has the behavior of selecting detours or non-direct flights, and judging whether there is a situation of greatly increasing travel time and cost; change frequency analysis, such as counting the frequent ticket changes and changes of the employee, to analyze the stability of the employee's trip, and to determine whether there is a situation of scheduling confusion or resource waste.

[0159] Further, the multiple audit dimensions can further include checking the integrity of the travel itinerary, including: verifying whether the actual travel route is consistent with the travel application by comparing the travel application information, the actual order record, and the location positioning information collected during the trip; if the deviation distance of the travel route exceeds the set distance, it is marked as a travel integrity anomaly. Specifically, the travel itinerary integrity verification is a key means to judge the authenticity of the travel behavior and the consistency of the travel application. The actual booked travel services (such as flights and hotel accommodations) in the travel itinerary are compared with the location trajectory (such as GPS positioning data, check-in information, etc.) of the employee during the trip. When the actual trajectory of the employee deviates from the destination by more than a set distance (such as 50 kilometers, the specific value can be configured), a travel integrity anomaly mark is automatically triggered. This mechanism can effectively prevent irregular behaviors such as fictitious travel and empty reimbursement.

[0160] Further, the plurality of auditing dimensions can further include analyzing whether the actual booking price of the travel service is abnormal, including: comparing the actual booking price of the travel service with the historical average price of the same type of travel service to determine whether there is a travel service order with a risk of cashing in, and marking the travel service order with the risk of cashing in. For example, compare each travel service order (such as a hotel accommodation order, an air ticket order, etc.) with the historical average price of the same type of travel service; when the price of a certain travel service order is higher than the average price by a certain percentage (such as more than 30%), or offline ticket refund / cancellation behavior occurs, it is marked as abnormal price behavior or has a risk of cashing in for further review by the financial or audit department.

[0161] Further, the embodiments of the present application can also support attribution analysis of travel cost abnormal travel behavior, determine the reason for travel cost waste, and develop corresponding control strategies based on the reason. For example, after the trip is completed, by analyzing historical order data, employee behavior patterns, booking timing, supplier selection, etc., the travel cost abnormal travel behavior (such as the price of the booked travel service being significantly higher than the average price of the same type of travel service, unnecessary upgrade of travel service, frequent ticket change, etc.) is attributed to analyze the specific reasons for cost waste, for example, late booking leading to high travel service price, not using the agreed travel service resources, repeatedly changing the itinerary, etc., and combined with the attribution results, develop appropriate travel control strategies, such as limiting the booking time of specific groups, improving the hit rate of agreed travel service resources, etc., thereby achieving controllable optimization of travel costs.

[0162] In further optional implementations, the embodiments of the present application can introduce a procurement optimization suggestion mechanism in the post-trip evaluation stage, thereby performing data mining and prediction analysis based on historical travel data to generate travel procurement optimization suggestions. As an optional implementation, the embodiments of the present application can generate a high-frequency supplier list based on high-frequency travel routes and / or high-frequency stay hotels; predict the bargaining potential of travel services according to the historical price data of travel service suppliers for travel services and the market price data of travel services; analyze the input-output results in combination with the savings of travel service suppliers after signing the agreement price and the fulfillment rate of agreement cooperation.

[0163] Specifically, by high-frequency route clustering and supplier identification, the cooperation object for which negotiation is recommended is determined, such as identifying high-frequency routes and counting the ticketing proportion of the corresponding airlines based on the start and end cities of the travel person's frequent round trips in the historical order records; analyzing the geographical location (such as the specific office building surrounding area), brand and price segment of the employee's high-frequency hotel stay, identifying the high-frequency hotel list; and thus generating a high-frequency supplier list for the above high-frequency routes and high-frequency hotels. Further, in combination with the historical transaction data and market price trend of the suppliers in the high-frequency supplier list, the future negotiation potential of the suppliers in the high-frequency supplier list is predicted; for example, analyzing the fluctuation range and average negotiation range of the historical discount rate of the suppliers in the historical years, and distinguishing the price strategy in the off-season and peak season; then, grabbing market price data to build a benchmark price curve of the same type of service as a comparison baseline; using a price prediction model to calculate the expected negotiation space of the supplier in a specific period, for example, a hotel has more than 15% negotiation space in the off-season.

[0164] Finally, the embodiments of the present application can track the effect of protocol cooperation, analyze the input-output results after establishing a protocol purchase with a specific supplier; for example, evaluating the proportion of travel service resources actually booked at the protocol price after signing a protocol with the supplier, that is, the protocol execution rate; comparing the protocol price with the average price under historical non-protocol procurement to calculate the cost saving amount; analyzing the cumulative saving proportion during the overall protocol execution period to output the ROI of protocol procurement. Further, in combination with the service quality score of the travel service resources at the protocol price, it is assisted to judge whether to continue cooperation with the travel service supplier.

[0165] In further optional implementations, the embodiments of the present application can also show the travel person behavior board to the administrator, and the content displayed by the travel person behavior board includes at least one of the following: the proportion of travel persons with high expenditure preference behavior and the corresponding excess expenditure proportion, the cost saving proportion of travel services and the corresponding cost saving method. For example, Figure 8A An example diagram of a board interface is shown, which is a strategic decision view in the travel management of an enterprise or other organization unit, mainly facing administrators, and is used to show travel cost control targets, actual saving effects, travel strategy scheme configuration states, and employee abnormal behavior warning information, etc.

[0166] Specifically, the board interface can include a travel behavior board, which refers to a management tool for visualizing information such as behavior patterns, preference choices, and savings contributions of employees during business trips. The display content includes but is not limited to: the proportion of employees with high spending preferences, such as the proportion of personnel who frequently choose flights, hotel accommodations, and other personnel that deviate from the recommended savings scheme; the proportion of excess spending corresponding to employees with high spending preferences, that is, the proportion of excess spending caused by such behavior, such as 20% of high-cost employees contributing 30% of excess spending; the cost saving ratio and the way, such as saving costs by combining trips, choosing protocol hotels, and using public transportation, and attributing the saving ratio, such as reducing 15% of accommodation costs by combining trips.

[0167] In combination Figure 8A as shown, Figure 8A The "part of the employees have overbooking situations" of the example can be regarded as corresponding to high spending preference behavior, and the saving target and the current saving rate reflect the contribution effect of the overall employee behavior on the saving target. By Figure 8A The board interface described above allows the administrator to view a comparison chart of the saving target and the actual saving rate, the saving amount of the strategy scheme, and the high spending preference behavior prompt. Further, with the help of the board, the administrator can identify potential cost waste behavior in a timely manner and improve the cost control level by enabling reasonable strategies such as same-day round-trip priority, chain hotel priority, and public transportation priority.

[0168] In further optional implementations, the embodiments of the present application can also support displaying a travel management interface to the administrator, wherein the travel management interface at least includes a strategy center page and a recommendation achievement page, the strategy center page is used for the administrator to configure various recommendation strategies and bind the recommendation strategies for employees;

[0169] The content displayed on the recommendation achievement page includes at least one of the following:

[0170] Core recommendation indicators, including at least one of the following: recommendation usage rate, indicating the proportion of travel personnel using recommended travel schemes to place orders; adjustment rate, indicating the proportion of travel personnel adjusting travel services in the recommended travel scheme; recommendation adoption rate, indicating the proportion of travel personnel using travel services and adjusted travel services in the travel scheme to place orders after saving the recommended travel scheme; and saving amount, indicating the total amount of travel costs saved by all recommended travel schemes;

[0171] Trend analysis chart, including at least one of the following: recommendation effect trend chart, showing the trend of recommendation usage rate, adjustment rate, and recommendation adoption rate at different times; and recommendation saving trend chart, showing the saving amount trend of recommended travel services and all travel services at different times;

[0172] The summary list records summary data of the ordered travel plan, and the summary data of an ordered travel plan includes at least one of the following: plan identification information, application order identification information, user information, itinerary information, budget information, plan saving information, and incentive information; and the summary data of an ordered travel plan is associated with an interactive option for entering a corresponding detail page.

[0173] The detail page displays content including at least one of the following: application order information, travel standard information, user itinerary demand information, actual order information, saving amount information, recommended travel plan information, and user adjusted travel plan information.

[0174] An example of a recommended result page is shown in FIG. 6, which shows Figure 8B An example of a recommended result page is shown in FIG. 6, which shows Figure 8B As shown, the recommended result page displays four key recommendation indicators, which are: recommended use rate, corresponding value 78%, indicating that 78% of users complete ordering based on recommended travel plans; adjustment rate, corresponding value 32%, indicating that 32% of users adjust the recommended travel service after ordering based on the recommended travel plan; recommended adoption rate, corresponding value 82%, indicating that the proportion of users who complete ordering based on recommended travel services or adjusted travel services after saving recommended travel plans is 82%; and recommended plan saving, indicating the cumulative travel cost saved due to the use of recommended travel plans. These indicators provide a perspective for the enterprise to quantitatively analyze the effect of the recommendation strategy, so as to support the enterprise to adjust the recommendation strategy in a timely manner according to the changing trend.

[0175] The recommended result page can also display trend analysis charts, which are: a recommended effect trend chart, showing the trend of recommended use rate, adjustment rate, and recommended adoption rate in each month, so as to help the administrator identify whether employee behavior has changed after strategy optimization; and a recommended saving trend chart, showing the growth of cumulative saving amount of different saving sources in each month, reflecting the trend of the contribution of the recommendation strategy to cost control over time.

[0176] The recommended result page can also display a summary list, which records summary data of ordered travel plans, including: recommended plan ID, application order number, user, itinerary, recommended budget, adjusted budget, budget difference (adjusted budget minus recommended budget), and incentive distribution information; and each record of the summary list supports detail viewing, providing an entry for viewing detailed content of each record.

[0177] The scheme provided by the embodiments of the present application can guarantee the basic travel needs of employees, effectively guide employees to preferentially select low-cost but acceptable travel schemes. In addition, the enterprise can set a saving target and the weight of each recommendation factor in the background, and dynamically adjust the recommendation strategy in combination with historical behavior data of employees and enterprise travel policies. In addition, when the recommendation system displays the recommended travel scheme to the traveler in the front end, it can guide the user to adopt the saving travel scheme through the saving prompt, the incentive mechanism and the differentiated visual reinforcement. It can be seen that the embodiments of the present application can realize the closed-loop optimization from the enterprise travel management demand configuration, the recommendation strategy generation, the front-end behavior guidance to the subsequent saving analysis, which helps to improve the recommendation adoption rate and reduce the average travel cost, and helps the enterprise to achieve the dual goals of travel compliance and cost control.

[0178] In a further optional implementation, the embodiments of the present application also provide a computer device for a travel service platform, which comprises a memory and a processor, the memory stores computer execution instructions, and the processor invokes the computer execution instructions to execute the travel management method provided by the embodiments of the present application. In a further optional implementation, the embodiments of the present application also provide a storage medium for storing computer execution instructions, which are executed by the processor to realize the travel management method provided by the embodiments of the present application. In a further optional implementation, the embodiments of the present application also provide a computer program comprising computer execution instructions, which are executed by the processor to realize the travel management method provided by the embodiments of the present application.

[0179] The above describes the multiple embodiment schemes provided by the embodiments of the present application. The optional modes introduced by each embodiment scheme can be combined, cross-referenced in the case of no conflict, thereby extending multiple possible embodiment schemes, which can be considered as the embodiments disclosed and disclosed by the embodiments of the present application. Although the embodiments of the present application are disclosed as above, the present application is not limited thereto. Any person skilled in the art can make various changes and modifications without departing from the spirit and scope of the present application, therefore the protection scope of the present application should be subject to the scope defined by the claims.

Claims

1. A travel management method characterized by, The method is applied to a travel service platform, the travel service platform comprising a travel service client and a travel background server, and the method comprises: obtaining travel application information of a trip person applying for business travel; when the travel application information is approved, and in the case of meeting a travel standard, recommending a business travel scheme for the trip person based on a recommendation strategy bound by the trip person; wherein the recommendation strategy bound by the trip person is preselected from multiple types of recommendation strategies, the recommendation strategy is associated with multiple recommendation factors influencing the recommended business travel scheme and the weight of each recommendation factor, the weight of at least one recommendation factor of different types of recommendation strategies is different, and the recommendation factor at least comprises a price factor, and the weight of the price factor is used to control the cost control degree of the recommended business travel scheme on the business travel cost.

2. The method of claim 1, wherein, The multiple types of recommendation strategies comprise at least two types as follows: a balanced recommendation strategy, which is used to recommend a business travel scheme by comprehensively considering the business travel cost, travel efficiency and travel preference of the trip person at least in the case of meeting the travel standard; a saving recommendation strategy, which is used to recommend a business travel scheme by giving priority to the business travel cost at least in the case of meeting the travel standard; an efficiency recommendation strategy, which is used to recommend a business travel scheme by giving priority to the travel efficiency at least in the case of meeting the travel standard; a comfort recommendation strategy, which is used to recommend a business travel scheme by giving priority to the travel preference and travel time consumption of the trip person at least in the case of meeting the travel standard.

3. The method of claim 2, wherein, The recommendation factor further comprises a preference factor, and the weight of the preference factor is used to control the preference consideration degree of the recommended business travel scheme on the travel preference of the trip person; wherein the price factor has multiple price weight levels corresponding to multiple cost control degrees of the business travel cost; and the preference factor has multiple preference weight levels corresponding to multiple preference consideration degrees of the travel preference. For the balanced recommendation strategy, the price weight level of the price factor corresponds to an intermediate degree of cost control degree, and the preference weight level of the preference factor corresponds to an intermediate degree of preference consideration degree. For the saving recommendation strategy, the price weight level of the price factor corresponds to the highest degree of cost control degree, and the preference weight level of the preference factor corresponds to the lowest degree of preference consideration degree. For the comfort recommendation strategy, the price weight level of the price factor corresponds to the lowest degree of cost control degree, and the preference weight level of the preference factor corresponds to the highest degree of preference consideration degree.

4. The method according to any one of claims 1 to 3, characterized in that, The travel standard comprises a travel standard upper limit and a recommended travel standard, and the recommended travel standard is used to control the business travel cost; wherein when the recommended travel standard meets a use condition, the recommended travel standard is used as the travel standard for the recommended business travel scheme, and the use condition comprises that the market average price of the travel service is lower than or equal to the recommended travel standard, and the recommended travel standard is lower than the travel standard upper limit; when the recommended travel standard does not meet the use condition, the travel standard upper limit is used as the travel standard for the recommended business travel scheme.

5. The method of claim 4, wherein, The business travel scheme comprises at least one travel service, and each travel service has a corresponding travel standard; the travel service comprises a hotel accommodation service, and the recommendation process of the hotel accommodation service comprises: determining a recall price range of the hotel accommodation service according to the recommended travel standard when the recommended travel standard meets a use condition; determining hotel accommodation services in the recall price range within a set geographical range of the expected accommodation location, and filtering the hotel accommodation services based on a filtering rule of the hotel accommodation service and a hotel filtering rule to obtain recalled hotel accommodation services; ranking the recalled hotel accommodation services according to the prices of the hotel accommodation services, and determining a hotel accommodation service recommended for the traveler based on a ranking result; wherein the score of any hotel accommodation service is obtained based on a current price of the hotel accommodation service, a highest price among the recalled hotel accommodation services, and a reference price intermediate value, and the closeness between the current price of the hotel accommodation service and the reference price intermediate value is positively correlated with the score of the hotel accommodation service; the reference price intermediate value is an intermediate value corresponding to an average price of the recalled hotel accommodation services and the highest price, and if the reference price intermediate value is higher than the recommended travel standard, the reference price intermediate value is adjusted to the recommended travel standard minus a preset value.

6. The method according to any one of claims 1 to 3, characterized in that, The travel standard is divided into multi-dimensional travel standards according to a plurality of personnel grouping dimensions, and the method further comprises: if the traveler is bound to at least two-dimensional travel standards, selecting a travel standard used when selecting a recommended travel scheme from the at least two-dimensional travel standards according to a priority of the at least two-dimensional travel standards, wherein the priority of the multi-dimensional travel standards is related to the priority of the plurality of personnel grouping dimensions; if the priorities of the at least two-dimensional travel standards are the same, selecting a highest travel standard or a lowest travel standard from the at least two-dimensional travel standards according to a pre-configured difference standard processing option, the difference standard processing option being divided into a high processing option and a low processing option, the high processing option corresponding to the selection of the highest travel standard, and the low processing option corresponding to the selection of the lowest travel standard; and / or, if there are a plurality of travelers in the travel application information, selecting a highest travel standard or a lowest travel standard from the travel standards bound by the plurality of travelers based on a pre-configured difference standard processing option as a travel standard used when selecting a recommended travel scheme; wherein the transportation ticket service and the hotel accommodation service each have a corresponding difference standard processing option, and the difference standard processing option is divided into a high processing option and a low processing option.

7. The method of claim 1, wherein, The method further comprises: approving the travel application information, and the approval at least includes a travel necessity verification; The process of the travel necessity verification comprises: analyzing the content of the travel application information; matching the analyzed content with a preset travel necessity rule library to obtain a rule matching result, the travel necessity rule library including a rule set of travel necessity conditions and / or a rule set of travel non-necessity conditions; and, performing a rationality analysis of the travel application information based on historical travel applications related to the analyzed content and the travel application information to obtain a rationality analysis result; combining the rule matching result and the rationality analysis result to generate a necessity risk score of the travel application information; If the necessity risk score exceeds a preset risk score threshold, the travel application information is marked as an abnormal travel application and risk reminder information is output, the risk reminder information being accompanied by a risk level corresponding to the necessity risk score; If the necessity risk score does not exceed the preset risk score threshold, it is determined that the travel necessity check of the travel application information is passed.

8. The method of claim 1, wherein, The method further comprises: recommending a recommended travel period for the traveler, including: obtaining real-time price data of a travel service related to the travel, and combining dynamic influencing factors affecting the price of the travel service, predicting price fluctuation trends corresponding to a plurality of candidate travel periods; and judging travel compliance of each candidate travel period according to travel compliance rules to obtain travel compliance results of each candidate travel period; recommending a recommended travel period according to the price fluctuation trends and the travel compliance results of each candidate travel period; wherein when detecting emergency information related to the travel, the recommended travel period is re-recommended; and the predicted travel cost and the travel compliance result of each candidate travel period are also displayed in a time sequence on a visual interface; and / or, when recommending a travel plan for the traveler, predicting a booking price and recommending a booking opportunity for the travel service, including: using a hybrid prediction model to predict a future price trend of the travel service, and when the price of the travel service in a certain period is lower than a price threshold, regarding the period as a recommended booking opportunity and automatically triggering a booking reminder or a price locking process; wherein the hybrid prediction model includes an LSTM model and a random forest model, the LSTM model is used to learn the change law of the price of the travel service over time, and the random forest model is used to mine the influence of discrete factors in historical booking data on the price of the travel service.

9. The method of claim 4, wherein, The travel plan includes at least one travel service, and each travel service has a corresponding travel standard, wherein the price of the recommended travel service does not exceed the upper limit of the corresponding travel standard; The method further comprises: marking a travel service in a travel service resource library that exceeds the upper limit of the corresponding travel standard as an over-standard state; and / or, if the traveler adjusts the travel service in the travel plan and the price of the adjusted travel service exceeds the corresponding recommended travel standard, triggering an over-standard reporting logic to prompt the traveler to fill in the over-standard reason; and / or, if the recommended strategy bound by the traveler includes a saving-type recommended strategy, performing a guide process to guide the traveler to select a travel plan corresponding to the saving-type recommended strategy, the guide process including at least one of the following: highlighting the travel plan corresponding to the saving-type recommended strategy on a travel plan recommendation page and displaying cost saving information relative to other travel plans; prompting saving incentive information to prompt the traveler to obtain an incentive reward if the traveler selects the travel plan corresponding to the saving-type recommended strategy; when the travel plan corresponding to the saving-type recommended strategy and the travel preference of the traveler have differences, but the saving amount exceeds a preset threshold, displaying corresponding guide information to guide the traveler to select the travel plan corresponding to the saving-type recommended strategy. When the trip person does not select the travel scheme corresponding to the saving type recommended strategy, an intervention mechanism is triggered to prompt the trip person to fill in the reason for not adopting the saving type recommended strategy and / or initiate an approval process.

10. The method of claim 1, wherein, The recommendation logic of the recommended strategy is learned by an AI model, so that the AI model recommends the corresponding travel scheme based on the recommendation logic of the recommended strategy; wherein the AI model is trained based on a configured strategy target, the strategy target is used to construct a target function in the AI model training process, and the strategy target includes: a travel cost reduction ratio associated with a price factor; a recommended adoption rate of the travel scheme associated with the preference factor; The AI model also adjusts the recommendation logic of the recommended strategy in combination with historical selection data of the travel scheme.

11. The method of claim 1, wherein, Further comprising: at least for the completed travel of the concerned population, travel rationality audit is performed, the concerned population includes trip persons with high expenditure preference behavior and / or trip persons who frequently submit travel applications exceeding the budget, and the trip persons with high expenditure preference behavior include: trip persons who frequently select other travel services with higher prices than recommended travel services; The travel rationality audit includes at least one of the following: analyze whether there is abnormal travel behavior, the abnormal travel behavior includes at least one of the following: repeated travel behavior, multiple selection of non-agreement suppliers or travel services exceeding the travel standard without explanation, selection of detour or non-direct transportation ticket service, multiple travel service changes, multiple travel service booking near the date, travel service overuse, offline change of transportation ticket service; verify the integrity of the travel, including: by comparing the travel application information, the actual order record and the location positioning information collected during the trip, verifying whether the actual travel route is consistent with the travel application; if the deviation distance of the travel route exceeds the set distance, it is marked as travel integrity abnormality; analyze whether the actual booking price of the travel service is abnormal, including: comparing the actual booking price of the travel service with the historical average price of the same type of travel service to determine whether there is a travel service order with the risk of money laundering, and marking the travel service order with the risk of money laundering.

12. The method of claim 1, wherein, Further comprising: show the trip person behavior board to the administrator, the content displayed on the trip person behavior board includes at least one of the following: the proportion of trip persons with high expenditure preference behavior and the corresponding excess expenditure proportion, the cost saving proportion of travel services and the corresponding cost saving method; and / or, attribute analysis is performed on the travel cost abnormal travel behavior to determine the reason for the waste of travel cost, and corresponding control strategies are developed based on the reason; and / or, data mining and prediction analysis are performed on the historical travel data to form travel procurement suggestions, including at least one of the following: based on high-frequency travel routes and / or high-frequency stay hotels, a high-frequency supplier list is generated; according to the historical price data of the travel service suppliers for the travel service and the market price data of the travel service, the negotiation potential of the travel service is predicted; combined with the saved cost of the travel service suppliers after signing the agreement price and the fulfillment rate of the agreement cooperation, the input-output result is analyzed; and / or, The travel management interface is displayed to the administrator, and the travel management interface includes at least a policy center page and a recommendation result page, the policy center page is used for the administrator to configure various recommendation policies and bind the recommendation policies for employees; The recommendation result page displays at least one of the following: Core recommendation indicators, including at least one of the following: recommendation usage rate, indicating the proportion of travelers using recommended travel solutions to place orders; adjustment rate, indicating the proportion of travelers adjusting travel services in the recommended travel solution; recommendation adoption rate, indicating the proportion of travelers using travel services and adjusted travel services in the recommended travel solution to place orders after saving the recommended travel solution; and saved amount, indicating the total travel cost saved by all recommended travel solutions; Trend analysis charts, including at least one of the following: a recommendation effect trend chart showing the change trend of the recommendation usage rate, the adjustment rate, and the recommendation adoption rate at different times; and a recommendation saving trend chart showing the change trend of the saved amount of each recommended travel service and all travel services at different times; A summary list recording summary data of ordered travel solutions, and the summary data of an ordered travel solution includes at least one of the following: solution identification information, application order identification information, user information, itinerary information, budget information, solution saving information, and reward incentive information; and the summary data of an ordered travel solution is associated with an interactive option to enter a corresponding detail page; The detail page displays at least one of the following: application order information, travel standard information, user itinerary demand information, actual order information, saved amount information, recommended travel solution information, and user adjusted travel solution information.

13. A computer program, characterized in that, Computer instructions are executed by a processor to implement the travel management method of any one of claims 1-12.

Images