Enterprise travel planning method and device based on artificial intelligence

Abstract

The invention discloses an enterprise business travel planning method and device based on artificial intelligence, and a server side method comprises the steps: receiving a business travel planning request sent by a client side, wherein business trip application form information carried by the business travel planning request comprises a target user identifier, a business trip destination and business trip time; obtaining a target travel standard corresponding to the target user identifier, and retrieving multiple types of consumption data conforming to a business trip destination and business trip time; acquiring behavior data in historical travel belonging to the target user identifier in a preset period, inputting a pre-trained personalized travel demand identification model, and outputting travel demand preference information corresponding to the target user identifier; and optimizing various consumption data according to the target travel standard and the travel demand preference information, generating travel planning information conforming to the target user identifier, and sending the travel planning information to the client for display. Therefore, by adopting the method and the device, the business travel which not only meets the personalized business travel demand of the user but also meets the business travel standard can be accurately generated, so that the satisfaction degree of the user is improved.

Description

Technical Field

[0001] The present application relates to the field of machine learning technology, and in particular to an artificial intelligence-based corporate travel planning method and device. Background Art

[0002] In the context of global business activities, corporate employee travel has become an indispensable part of corporate operations. Corporate management is faced with the problem of how to arrange employee travel efficiently and economically while ensuring the employee's travel experience. For example, an employee of a company needs to travel overseas to attend an important business meeting. The company needs to plan a travel plan for him in a short period of time that meets both travel standards and the employee's personal preferences.

[0003] In the related technology, the existing travel planning system usually performs travel planning based on a simple rule engine, lacking a comprehensive analysis of the user's historical travel records and the company's travel standards. This makes it impossible for the system to accurately generate a travel itinerary that meets both the user's personalized travel needs and the travel standards, resulting in low user satisfaction with travel itinerary planning. Summary of the invention

[0004] The embodiments of the present application provide a method and device for corporate travel planning based on artificial intelligence. In order to have a basic understanding of some aspects of the disclosed embodiments, a simple summary is given below. This summary is not a general review, nor is it intended to identify key / important components or describe the scope of protection of these embodiments. Its only purpose is to present some concepts in a simple form as a preface to the detailed description that follows.

[0005] In a first aspect, an embodiment of the present application provides an enterprise travel planning method based on artificial intelligence, which is applied to a server, and the method includes: Receive a travel planning request sent by a client, the travel planning request carries travel application form information, and the travel application form information includes a target user identifier, a travel destination, and a travel time; From the pre-established mapping relationship between user IDs and travel standards, obtain the target travel standards corresponding to the target user ID, and retrieve multiple types of consumption data that match the business trip destination and business trip time; Obtaining historical travel behavior data belonging to the target user ID within a preset period; The behavior data is input into a pre-trained personalized travel demand recognition model, and the travel demand preference information corresponding to the target user ID is output; the pre-trained personalized travel demand recognition model is trained based on the historical behavior data collected and pre-processed in the historical travel belonging to different user IDs; According to the target travel standards and travel demand preference information, various types of consumption data are optimized to generate travel planning information that meets the target user identification and send it to the client for display.

[0006] Optionally, the multiple types of consumption data include a transportation consumption information list, a hotel consumption information list, and a restaurant consumption information list; According to the target travel standards and travel demand preference information, various types of consumption data are optimized to generate travel planning information that meets the target user identification, including: Traverse the transportation consumption information, hotel consumption information and catering consumption information in the transportation consumption information list, hotel consumption information list and catering consumption information list respectively; The traversed transportation consumption information, hotel consumption information, and catering consumption information are combined to obtain joint consumption information, and the steps of traversing the transportation consumption information, hotel consumption information, and catering consumption information in the transportation consumption information list, the hotel consumption information list, and the catering consumption information list are continued until a plurality of joint consumption information are obtained after the traversal is completed; Calculate the total consumption amount of each joint consumption information; Comparing the total consumption amount of each joint consumption information with the target travel standard, so as to eliminate the joint consumption information with a consumption amount greater than the target travel standard from the multiple joint consumption information, and obtain multiple target joint consumption information; Based on multiple target joint consumption information and travel demand preference information, travel planning information that meets the target user identification is generated.

[0007] Optionally, based on multiple pieces of target joint consumption information and travel demand preference information, travel planning information that matches the target user identifier is generated, including: Performing semantic analysis on the travel demand preference information to extract the demand preference keyword sequence of the user corresponding to the target user ID; The demand text is marked with the demand preference keyword sequence to guide the preset large language model and obtain prompt information; Input the prompt information and multiple target joint consumption information into a preset large language model, and use the preset large language model to find and output a preset number of target joint consumption information that meets the travel demand preference information from the multiple target joint consumption information; the preset large language model is ChatGPT; A preset amount of target joint consumption information is used as travel planning information that matches the target user identifier.

[0008] Optionally, based on multiple pieces of target joint consumption information and travel demand preference information, travel planning information that matches the target user identifier is generated, including: According to the travel demand preference information, determine the importance that the user corresponding to the target user ID places on different consumption categories; According to the importance users place on different consumption categories, the default weights of each piece of consumption information in each target joint consumption information are dynamically adjusted to obtain the target weights of each piece of consumption information in each target joint consumption information. The default weights are used to represent the average preferences of users. Score each piece of consumption information in each target joint consumption information to obtain a preference score for each piece of consumption information in each target joint consumption information; The target weight of each piece of consumer information in each target joint consumption information and the preference score of each piece of consumer information in each target joint consumption information are weighted and summed to obtain the user personalized total score of each target joint consumption information; Sort multiple target joint consumption information based on the user personalized total score of each target joint consumption information; From the sorted multiple pieces of target joint consumption information, a preset number of target joint consumption information are obtained in order from large to small according to the total scores, as travel planning information that meets the target user identification.

[0009] Optionally, different consumption categories include transportation consumption, hotel consumption and catering consumption; travel demand preference information includes accommodation preference information, transportation preference information and catering preference information; Based on the travel demand preference information, determine the importance that the user corresponding to the target user ID places on different consumption categories, including: Analyze the first preference level of transportation consumption according to the airline information, flight type information, seat information and cabin class information included in the transportation preference information; Analyze the second preference level of hotel consumption based on the hotel star information, geographical location information, room type information and special accommodation demand information included in the accommodation preference information; Analyze the third preference level of catering consumption according to the catering type information, catering budget information and special dietary demand information included in the catering preference information; From the pre-established mapping relationship between preference levels and scores, the scores corresponding to the first preference level, the second preference level and the third preference level are respectively obtained as the degree of importance that the user corresponding to the target user identifier has on different consumption categories.

[0010] Optionally, according to the importance that users attach to different consumption categories, the default weight of each consumption information in each target joint consumption information is dynamically adjusted to obtain the target weight of each consumption information in each target joint consumption information, including: Calculate the target weight of each piece of consumer information in each piece of target joint consumption information according to the default weight of each piece of consumer information in each piece of target joint consumption information and the degree of importance that users place on different consumer categories; Among them, the calculation formula of the target weight of each consumption information in each target joint consumption information is:

[0011] in, For the target joint consumption information The target weight of consumption information, is the adjustment factor, which is used to control the sensitivity of weight adjustment. It is Consumption information in the corresponding consumption category The degree of attention paid to is an index variable, which is used to traverse all consumption data during the summation process. is 3.

[0012] Optionally, each piece of target joint consumption information includes transportation parameters, hotel parameters, and catering parameters; Score each piece of consumer information in each target joint consumer information to obtain the preference score of each piece of consumer information in each target joint consumer information, including: Get the transportation consumption items corresponding to the transportation parameters; get the hotel consumption items corresponding to the hotel parameters; get the catering consumption items corresponding to the catering parameters; The transportation consumption items, hotel consumption items and catering consumption items are numerically quantified by using a preset scoring quantification table to obtain multiple numerical values ​​corresponding to the transportation consumption items, multiple numerical values ​​corresponding to the hotel consumption items and multiple numerical values ​​corresponding to the catering consumption items; The multiple values ​​corresponding to the transportation consumption items, the multiple values ​​corresponding to the hotel consumption items, and the multiple values ​​corresponding to the catering consumption items are summed up respectively to obtain the preference score of each consumption information in each target joint consumption information.

[0013] Optionally, follow the steps below to generate a pre-trained personalized travel demand recognition model, including: Collect and pre-process historical behavior data from historical travels belonging to different user IDs, where the historical behavior data includes historical travel records and online operation information of users on the travel booking platform; Perform feature extraction on the preprocessed historical behavior data to obtain multiple initial features; Extract key features related to travel demand preferences from multiple initial features; The key features related to travel demand preferences are encoded using the One-Hot encoding method to obtain multiple numerical features; Associating the preprocessed historical behavior data with multiple corresponding numerical features to obtain model training samples; Based on the model training samples, a pre-trained personalized travel demand recognition model is generated.

[0014] Optionally, a pre-trained personalized travel demand recognition model is generated based on the model training samples, including: Use decision tree algorithm to create a personalized travel demand identification model; Input the model training samples into the personalized travel demand recognition model and output the model loss value; When the loss value reaches the minimum, a pre-trained personalized travel demand recognition model is generated; or, when the loss value does not reach the minimum, the step of inputting the model training sample into the personalized travel demand recognition model is continued until the model loss value reaches the minimum; wherein,

[0015] in, is the loss value of the model, is the number of model training samples, It is The travel demand preference information of multiple numerical features corresponding to the pre-processed historical behavior data, It is The predicted travel demand preference information corresponding to the preprocessed historical behavior data.

[0016] In a second aspect, an embodiment of the present application provides an enterprise travel planning device based on artificial intelligence, the device comprising: A travel planning request receiving module is used to receive a travel planning request sent by a client, the travel planning request carries the travel application form information, and the travel application form information includes the target user identifier, the travel destination and the travel time; A first data acquisition module is used to acquire a target travel standard corresponding to a target user identifier from a pre-established mapping relationship between user identifiers and travel standards, and retrieve multiple types of consumption data that match a business trip destination and a business trip time; The second data acquisition module is used to acquire the historical travel behavior data belonging to the target user identifier within a preset period; A model processing module is used to input the behavior data into a pre-trained personalized travel demand recognition model, and output the travel demand preference information corresponding to the target user ID; the pre-trained personalized travel demand recognition model is trained based on the collection and pre-processing of historical behavior data in historical travels belonging to different user IDs; The travel planning information output module is used to optimize various types of consumption data according to the target travel standards and travel demand preference information, generate travel planning information that meets the target user identification, and send it to the client for display.

[0017] The technical solution provided by the embodiments of the present application may have the following beneficial effects: In an embodiment of the present application, on the one hand, various types of consumption data are optimized by combining target travel standards and travel demand preference information. The target travel standards control the consumption scope of travel, and the travel demand preference information is identified based on a personalized travel demand identification model obtained based on machine learning, which can comprehensively characterize the personalized needs of users. The combination of the two for comprehensive analysis can generate a travel itinerary that meets both the personalized needs of users and the corporate travel standards; on the other hand, by considering the personalized needs of users, the generated travel plan is more likely to meet the expectations of users, thereby improving user satisfaction.

[0018] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0020] Figure 1 This is a method flow diagram of an artificial intelligence-based corporate travel planning method provided in an embodiment of the present application; Figure 2 is a schematic diagram of business trip application data submitted by different employees provided in an embodiment of the present application; Figure 3 It is a user interface diagram of a business trip application form provided by this application; Figure 4 This is a standard diagram of corporate travel for different employees provided by this application; Figure 5 It is a flowchart of a model training method provided by this application; Figure 6 This is a structural diagram of an enterprise travel planning device based on artificial intelligence provided by the present application; Figure 7 It is a structural schematic diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0021] The following description and the accompanying drawings sufficiently illustrate specific embodiments of the present application to enable those skilled in the art to practice them.

[0022] It should be clear that the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0023] When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. Instead, they are only examples of devices and methods consistent with some aspects of the present application as detailed in the attached claims.

[0024] In the description of the present application, it should be understood that the terms "first", "second", etc. are used for descriptive purposes only and should not be understood as indicating or implying relative importance. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to the specific circumstances. In addition, in the description of the present application, unless otherwise specified, "multiple" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the objects associated before and after are in an "or" relationship.

[0025] Currently, existing travel planning systems usually perform travel planning based on simple rule engines.

[0026] The inventors realized that the current method lacks a comprehensive analysis of the user's historical travel records and the company's travel standards, making it impossible for the system to accurately generate a travel itinerary that meets both the user's personalized travel needs and the travel standards, resulting in low user satisfaction with travel itinerary planning.

[0027] In order to solve the above problems, the present application provides an enterprise travel planning method and device based on artificial intelligence to solve the problems existing in the above related technical problems. In the embodiment of the present application, on the one hand, by combining the target travel standard and the travel demand preference information, various types of consumption data are optimized and processed. The target travel standard controls the consumption scope of the travel. The travel demand preference information is identified based on the personalized travel demand identification model obtained by machine learning, which can fully characterize the personalized needs of users. The combination of the two for comprehensive analysis can generate a travel itinerary that meets both the personalized needs of users and the corporate travel standards; on the other hand, by considering the personalized needs of users, the generated travel plan is more likely to meet the expectations of users, thereby improving user satisfaction. The following uses an exemplary embodiment for detailed description.

[0028] The following will be combined with the attached Figure 1 -Attached Figure 5 , the corporate travel planning method based on artificial intelligence provided by the embodiment of the present application is introduced in detail. The method can be implemented by relying on a computer program and can be run on a corporate travel planning device based on artificial intelligence based on the von Neumann system. The computer program can be integrated into an application or run as an independent tool application.

[0029] See also Figure 1 , provides a flow chart of an enterprise travel planning method based on artificial intelligence for an embodiment of the present application, which is applied to the server. Figure 1 As shown, the method of the embodiment of the present application may include the following steps: S101, receiving a travel planning request sent by a client, the travel planning request carrying travel application form information, the travel application form information including a target user identifier, a travel destination, and a travel time; Among them, the client refers to the user interface or the platform with which the user directly interacts, such as a web page, mobile application or desktop software, through which the user submits a travel planning request. A travel planning request is a request initiated by a user, asking the system to provide planning and arrangements for his or her business trip itinerary. The business trip application form information is a series of information that the user needs to provide when planning a business trip. This information usually includes the purpose, destination, time, etc. of the business trip, which is used for travel planning. The target user identifier refers to a unique identifier used to identify and distinguish different users, which can be a user name, employee ID, email address, etc. The business trip destination refers to the city or place that the user needs to reach on the business trip. The business trip time refers to the start and end dates of the user's planned business trip.

[0030] In some embodiments of the present application, for example, Xiao Zhang is a sales manager of a company, and he needs to plan a business trip. First, use the company's travel management mobile application to submit a business trip request. In the application, click "New Business Trip Application" and fill in the relevant information. Specifically enter the following business trip application form information: The target user identifier is Xiao Zhang's employee ID "SALES123456", which is his unique identifier in the company system. Business trip destination: He enters "Shanghai", which is the destination of his business trip. Business trip time: He plans to depart on June 1, 2024 and return on June 5. The application (client) sends Zhang Wei's business trip application form information to the company's server system. After receiving the request, the server system obtains the target user identifier, business trip destination, and business trip time included in the business trip application form information carried by the request.

[0031] Among them, the information filled in by different users on the business trip application form is as follows: Figure 2 Business trip application form, for example Figure 3 shown.

[0032] S102, obtaining a target travel standard corresponding to a target user identifier from a pre-established mapping relationship between user identifiers and travel standards, and retrieving multiple types of consumption data that match a business trip destination and a business trip time; Among them, travel standards are a series of regulations and restrictions set by the company for employees on business trips, including accommodation level, transportation mode, food and beverage expenses, etc. The mapping relationship refers to the correspondence between user IDs and travel standards, which is usually stored in a database and used to quickly find the corresponding travel standards based on user IDs. Multi-category consumption data refers to data on various consumption options related to business trips, such as prices and information of different flights, hotels, and catering services.

[0033] In some embodiments of the present application, when employee Zhang San (user ID: ZHANGSAN123) submits a business trip application to the system, the system obtains Zhang San's user ID. The system retrieves Zhang San's corresponding travel standards in the mapping relationship based on the user ID, such as accommodation not exceeding four stars, economy class flights, and meals not exceeding 300 yuan per day. Zhang San's business trip destination is Shanghai, and the business trip time is from June 1, 2024 to June 3, 2024. Based on Zhang San's business trip destination and time, the system retrieves all available flights, hotels, and catering services in the consumption data table.

[0034] For example Figure 4 As shown in the figure, different employees correspond to different travel standards. User ID is used to uniquely identify a user, usually a system-generated number. Name is the name of the user. Department is the department to which the user belongs. Accommodation Standard is the hotel grade standard that the user is allowed to stay in when on a business trip. Transportation Standard is the transportation grade standard that the user is allowed to choose when on a business trip. Meal Standard is the upper limit of the meal cost per meal when the user is on a business trip. Budget Upper Limit (Yuan) is the upper limit of the total budget for the user's business trip.

[0035] S103, obtaining historical travel behavior data belonging to the target user identifier within a preset period; The preset period refers to a period of time set for a specific purpose, such as a month, a quarter, or a year, for collecting and analyzing data. Historical travel refers to all business trips that employees have taken in the past, including completed business trips and related details. Behavioral data refers to the user's behavior records in the system, including operations such as searching, browsing, selecting, and booking.

[0036] In some embodiments of the present application, after obtaining the target travel standard corresponding to the target user identifier and retrieving multiple types of consumption data that match the business trip destination and time, the system analyzes the employee's business trip behavior data over the past year (January 1, 2023 to December 31, 2023). For example, the user identifier of the company employee Li Ming is "LM123". The system retrieves all business trip records related to "LM123" from the database, including flight reservations, hotel accommodation and catering consumption, and the user's online operation information on the travel booking platform.

[0037] S104, inputting the behavior data into a pre-trained personalized travel demand recognition model, and outputting travel demand preference information corresponding to the target user ID; the pre-trained personalized travel demand recognition model is trained based on the collection and pre-processing of historical behavior data in historical travels belonging to different user IDs; The travel demand preference information refers to the detailed information output by the model about the user's travel preferences and needs, such as the preferred hotel type, flight preferences, etc. The personalized travel demand identification model is an algorithm model used to predict the user's travel preferences based on the user's behavioral data.

[0038] In some embodiments of the present application, the system inputs the behavior data into a pre-trained personalized travel demand recognition model. After the model analyzes the behavior data, it outputs the employee's travel demand preference information, such as preference for business class, four-star hotels, and local specialty restaurants.

[0039] In some embodiments of the present application, the specific process of generating a pre-trained personalized travel demand identification model is: collecting and pre-processing historical behavior data in historical travels belonging to different user identifiers, the historical behavior data including historical travel records and online operation information of users on a travel booking platform; performing feature extraction on the pre-processed historical behavior data to obtain multiple initial features; extracting key features related to travel demand preferences from the multiple initial features; encoding the key features related to travel demand preferences using One-Hot encoding to obtain multiple numerical features; associating the pre-processed historical behavior data with the multiple numerical features corresponding thereto to obtain a model training sample; and generating a pre-trained personalized travel demand identification model based on the model training sample.

[0040] Among them, historical travel records refer to the records generated by users in past business trips, including specific information such as flights, hotels, and transportation. Online operation information is the operation records performed by users on the travel booking platform, such as searching for flights, browsing hotels, and selecting seats. Initial features are features extracted directly from historical behavior data without further processing. Key features are features that have a significant impact on travel demand preferences and are screened out from the initial features. One-Hot encoding is an encoding method for processing categorical variables, which converts categorical features into a series of binary (0 or 1) features. Numerical features are features that can be numerically operated, usually categorical features processed by encoding.

[0041] In an embodiment of the present application, by collecting and preprocessing historical travel records and online operation information belonging to different user identifiers, rich user behavior data can be obtained. Feature extraction is performed on these data to obtain multiple initial features. The key features related to travel demand preferences are further screened out, and these features are converted into numerical features using One-Hot encoding, so that the model can better understand and process these data. The preprocessed historical behavior data and the corresponding numerical features are associated to form a model training sample. Finally, a pre-trained personalized travel demand identification model is generated based on these samples. The beneficial effect of this process is that it enables the travel management system to more accurately predict and meet the personalized needs of users, thereby improving user experience and satisfaction, while also helping companies to more effectively control travel costs.

[0042] In some embodiments of the present application, the specific process of generating a pre-trained personalized travel demand identification model based on the model training samples includes: using a decision tree algorithm to create a personalized travel demand identification model; inputting the model training samples into the personalized travel demand identification model, and outputting the loss value of the model; when the loss value reaches the minimum, generating a pre-trained personalized travel demand identification model; or, when the loss value does not reach the minimum, continuing to execute the step of inputting the model training samples into the personalized travel demand identification model until the model loss value reaches the minimum; wherein,

[0043] in, is the loss value of the model, is the number of model training samples, It is The travel demand preference information of multiple numerical features corresponding to the pre-processed historical behavior data, It is The predicted travel demand preference information corresponding to the preprocessed historical behavior data.

[0044] S105, according to the target travel standard and travel demand preference information, various types of consumption data are optimized, travel planning information that meets the target user identification is generated, and sent to the client for display.

[0045] Among them, multiple types of consumption data include transportation consumption information list, hotel consumption information list and catering consumption information list.

[0046] In some embodiments of the present application, various types of consumption data are optimized according to target travel standards and travel demand preference information, and the specific process of generating travel planning information that meets the target user identification includes: traversing the transportation consumption information, hotel consumption information and catering consumption information in the transportation consumption information list, hotel consumption information and catering consumption information list respectively; combining the traversed transportation consumption information, hotel consumption information and catering consumption information to obtain joint consumption information, and continuing to execute the step of traversing the transportation consumption information, hotel consumption information and catering consumption information in the transportation consumption information list, hotel consumption information and catering consumption information list until a plurality of joint consumption information is obtained after the traversal; calculating the total consumption amount of each joint consumption information; comparing the total consumption amount of each joint consumption information with the target travel standard to eliminate the joint consumption information that is greater than the target travel standard from the plurality of joint consumption information to obtain a plurality of target joint consumption information; generating travel planning information that meets the target user identification according to the plurality of target joint consumption information and the travel demand preference information.

[0047] Among them, the transportation consumption information list is a list containing various transportation options and their related information (such as price, time, comfort, etc.), which is used for travel planning. The hotel consumption information list contains detailed information of different hotels, including price, star rating, location, etc., for selection when planning a business trip. The catering consumption information list is a list containing various catering options and their related information (such as price, taste, location, etc.), which is used for catering arrangements when traveling on business. Transportation consumption information refers to a specific transportation option information in the list, such as a specific flight or train itinerary. Hotel consumption information refers to the detailed information of a specific hotel in the list. Catering consumption information refers to the information of a specific catering option in the list. Joint consumption information is a comprehensive travel consumption plan formed by combining transportation, hotel and catering consumption information. The total consumption amount refers to the total cost of all consumption items in a joint consumption information. Target joint consumption information refers to joint consumption information that meets the target travel standards, that is, its total consumption amount does not exceed the travel budget set by the company. Travel planning information is the final travel arrangement and suggestions generated based on user needs and preferences, as well as budget standards.

[0048] In a possible implementation, a specific process of generating travel planning information that meets the target user identifier based on multiple target joint consumption information and travel demand preference information includes: performing semantic analysis on the travel demand preference information to extract the user demand preference keyword sequence corresponding to the target user identifier; marking the demand text used to guide the preset large language model for the demand preference keyword sequence to obtain prompt information; inputting the prompt information and the multiple target joint consumption information into the preset large language model, and searching and outputting a preset number of target joint consumption information that meets the travel demand preference information from the multiple target joint consumption information through the preset large language model; the preset large language model is ChatGPT; and using the preset number of target joint consumption information as the travel planning information that meets the target user identifier.

[0049] The demand preference keyword sequence is a series of keywords extracted from the travel demand preference information, which can represent the user's specific preferences. The demand text is a text describing the user's needs, which is used to guide or prompt the model to understand the user's specific needs. The prompt information is a text prompt generated based on the demand preference keyword sequence, which is used to guide the preset large language model to perform specific tasks. The preset large language model is a large language model that is set in advance and used for specific tasks, such as ChatGPT.

[0050] Among them, by conducting in-depth semantic analysis of travel demand preference information, the system can accurately extract keyword sequences that represent user demand preferences. These keyword sequences are then used to generate demand text prompt information to provide clear guidance for preset large language models such as ChatGPT. When these prompt information is input into ChatGPT together with the target joint consumption information, the model can effectively filter and output a preset number of target joint consumption information that meets the user's travel demand preferences from multiple possible travel plans. The beneficial effect of this method is that it can use advanced natural language processing technology to provide more personalized and accurate travel planning services, thereby improving user satisfaction, optimizing the travel experience, and ensuring that travel arrangements meet both budget standards and user personal preferences.

[0051] In another possible implementation, the specific process of generating travel planning information that meets the target user identifier based on multiple target joint consumption information and travel demand preference information includes: determining the importance of different consumption categories to the user corresponding to the target user identifier based on the travel demand preference information; dynamically adjusting the default weight of each consumption information in each target joint consumption information based on the importance of the user to different consumption categories, and obtaining the target weight of each consumption information in each target joint consumption information, and the default weight is used to represent the average preference of the user; scoring each consumption information in each target joint consumption information to obtain the preference score of each consumption information in each target joint consumption information; weighted summing the target weight of each consumption information in each target joint consumption information and the preference score of each consumption information in each target joint consumption information to obtain the user personalized total score of each target joint consumption information; sorting the multiple target joint consumption information based on the user personalized total score of each target joint consumption information; and obtaining a preset number of target joint consumption information in order from the sorted multiple target joint consumption information in descending order according to the total score as the travel planning information that meets the target user identifier.

[0052] Among them, different consumption categories include transportation consumption, hotel consumption and catering consumption; travel demand preference information includes accommodation preference information, transportation preference information and catering preference information.

[0053] Specifically, the specific process of determining the degree of importance that the user corresponding to the target user identifier attaches to different consumption categories based on the travel demand preference information includes: analyzing the first preference level of transportation consumption based on the airline information, flight type information, seat information and cabin class information included in the transportation preference information; analyzing the second preference level of hotel consumption based on the hotel star information, geographical location information, room type information and special accommodation demand information included in the accommodation preference information; analyzing the third preference level of catering consumption based on the catering type information, catering budget information and special dietary demand information included in the catering preference information; and obtaining the scores corresponding to the first preference level, the second preference level and the third preference level from the pre-established mapping relationship between preference levels and scores, as the degree of importance that the user corresponding to the target user identifier attaches to different consumption categories.

[0054] Specifically, according to the importance that users attach to different consumption categories, the default weight of each consumption information in each target joint consumption information is dynamically adjusted, and the specific process of obtaining the target weight of each consumption information in each target joint consumption information includes: according to the default weight of each consumption information in each target joint consumption information and the importance that users attach to different consumption categories, the target weight of each consumption information in each target joint consumption information is calculated; wherein, the calculation formula of the target weight of each consumption information in each target joint consumption information is:

[0055] in, For the target joint consumption information The target weight of consumption information, is the adjustment factor, which is used to control the sensitivity of weight adjustment. It is Consumption information in the corresponding consumption category The degree of attention paid to is an index variable, which is used to traverse all consumption data during the summation process. is 3.

[0056] Specifically, each piece of target joint consumption information includes transportation parameters, hotel parameters, and catering parameters.

[0057] In some embodiments, scoring each piece of consumption information in each piece of target joint consumption information to obtain a specific process of preference scores for each piece of consumption information in each piece of target joint consumption information includes: obtaining transportation consumption items corresponding to transportation parameters; obtaining hotel consumption items corresponding to hotel parameters; obtaining catering consumption items corresponding to catering parameters; numerically quantifying transportation consumption items, hotel consumption items and catering consumption items through a preset scoring quantization table to obtain multiple numerical values ​​corresponding to transportation consumption items, multiple numerical values ​​corresponding to hotel consumption items and multiple numerical values ​​corresponding to catering consumption items; summing up the multiple numerical values ​​corresponding to transportation consumption items, the multiple numerical values ​​corresponding to hotel consumption items and the multiple numerical values ​​corresponding to catering consumption items, respectively, to obtain the preference scores for each piece of consumption information in each piece of target joint consumption information.

[0058] In an embodiment of the present application, on the one hand, various types of consumption data are optimized by combining target travel standards and travel demand preference information. The target travel standards control the consumption scope of travel, and the travel demand preference information is identified based on a personalized travel demand identification model obtained based on machine learning, which can comprehensively characterize the personalized needs of users. The combination of the two for comprehensive analysis can generate a travel itinerary that meets both the personalized needs of users and the corporate travel standards; on the other hand, by considering the personalized needs of users, the generated travel plan is more likely to meet the expectations of users, thereby improving user satisfaction.

[0059] See also Figure 5 , which is a flowchart of a personalized travel demand identification model training method provided in an embodiment of the present application. Figure 5 As shown, the method of the embodiment of the present application may include the following steps: S201, collecting and preprocessing historical behavior data in historical travel belonging to different user identifiers, where the historical behavior data includes historical travel records and online operation information of users on a travel booking platform; S202, performing feature extraction on the pre-processed historical behavior data to obtain a plurality of initial features; S203, extracting key features related to travel demand preferences from multiple initial features; S204, encoding key features related to travel demand preference using a One-Hot encoding method to obtain multiple numerical features; S205, associating the pre-processed historical behavior data with a plurality of numerical features corresponding thereto to obtain a model training sample; S206: Generate a pre-trained personalized travel demand recognition model based on the model training samples.

[0060] In an embodiment of the present application, on the one hand, various types of consumption data are optimized by combining target travel standards and travel demand preference information. The target travel standards control the consumption scope of travel, and the travel demand preference information is identified based on a personalized travel demand identification model obtained based on machine learning, which can comprehensively characterize the personalized needs of users. The combination of the two for comprehensive analysis can generate a travel itinerary that meets both the personalized needs of users and the corporate travel standards; on the other hand, by considering the personalized needs of users, the generated travel plan is more likely to meet the expectations of users, thereby improving user satisfaction.

[0061] The following is an embodiment of the device of the present application, which can be used to execute the embodiment of the method of the present application. For details not disclosed in the embodiment of the device of the present application, please refer to the embodiment of the method of the present application.

[0062] See also Figure 6 , which shows a schematic diagram of the structure of an enterprise travel planning device based on artificial intelligence provided by an exemplary embodiment of the present application. The enterprise travel planning device based on artificial intelligence can be implemented as all or part of an electronic device through software, hardware or a combination of both. The device 1 includes a travel planning request receiving module 10, a first data acquisition module 20, a second data acquisition module 30, a model processing module 40, and a travel planning information output module 50.

[0063] The travel planning request receiving module 10 is used to receive the travel planning request sent by the client, the travel planning request carries the business trip application form information, and the business trip application form information includes the target user identification, the business trip destination and the business trip time; The first data acquisition module 20 is used to acquire the target travel standard corresponding to the target user identifier from the pre-established mapping relationship between the user identifier and the travel standard, and retrieve multiple types of consumption data that meet the business trip destination and business trip time; The second data acquisition module 30 is used to acquire the historical travel behavior data belonging to the target user identifier within a preset period; The model processing module 40 is used to input the behavior data into a pre-trained personalized travel demand recognition model, and output the travel demand preference information corresponding to the target user identifier; the pre-trained personalized travel demand recognition model is trained based on the historical behavior data collected and pre-processed in the historical travel belonging to different user identifiers; The travel planning information output module 50 is used to optimize various consumption data according to the target travel standard and travel demand preference information, generate travel planning information that meets the target user identification, and send it to the client for display.

[0064] It should be noted that the AI-based corporate travel planning device provided in the above embodiment only uses the division of the above functional modules as an example when executing the AI-based corporate travel planning method. In actual applications, the above functional distribution can be completed by different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the AI-based corporate travel planning device provided in the above embodiment and the AI-based corporate travel planning method embodiment belong to the same concept. The implementation process is detailed in the method embodiment, which will not be repeated here.

[0065] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0066] In an embodiment of the present application, on the one hand, various types of consumption data are optimized by combining target travel standards and travel demand preference information. The target travel standards control the consumption scope of travel, and the travel demand preference information is identified based on a personalized travel demand identification model obtained based on machine learning, which can comprehensively characterize the personalized needs of users. The combination of the two for comprehensive analysis can generate a travel itinerary that meets both the personalized needs of users and the corporate travel standards; on the other hand, by considering the personalized needs of users, the generated travel plan is more likely to meet the expectations of users, thereby improving user satisfaction.

[0067] The present application also provides a computer-readable medium having program instructions stored thereon, which, when executed by a processor, implement the artificial intelligence-based corporate travel planning method provided by the above-mentioned various method embodiments.

[0068] The present application also provides a computer program product comprising instructions, which, when executed on a computer, enables the computer to execute the corporate travel planning method based on artificial intelligence of each of the above method embodiments.

[0069] See also Figure 7 , is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Figure 7As shown, the electronic device 1000 may include: at least one processor 1001 , at least one network interface 1004 , a user interface 1003 , a memory 1005 , and at least one communication bus 1002 .

[0070] The communication bus 1002 is used to realize the connection and communication between these components.

[0071] The user interface 1003 may include a display screen (Display) and a camera (Camera). Optionally, the user interface 1003 may also include a standard wired interface and a wireless interface.

[0072] The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0073] Among them, the processor 1001 may include one or more processing cores. The processor 1001 uses various interfaces and lines to connect various parts of the entire electronic device 1000, and executes various functions and processes data of the electronic device 1000 by running or executing instructions, programs, code sets or instruction sets stored in the memory 1005, and calling data stored in the memory 1005. Optionally, the processor 1001 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 1001 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 1001, and it can be implemented separately through a chip.

[0074] Among them, the memory 1005 may include a random access memory (Random Access Memory, RAM) and may also include a read-only memory (Read-Only Memory). Optionally, the memory 1005 includes a non-transitory computer-readable storage medium. The memory 1005 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 1005 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 1005 may also be optionally at least one storage system located away from the aforementioned processor 1001. As Figure 7 As shown, the memory 1005 as a computer storage medium may include an operating system, a network communication module, a user interface module, and an enterprise travel planning application based on artificial intelligence.

[0075] exist Figure 7 In the electronic device 1000 shown, the user interface 1003 is mainly used to provide an input interface for the user and obtain data input by the user; and the processor 1001 can be used to call the artificial intelligence-based corporate travel planning application stored in the memory 1005, and specifically perform the following operations: Receive a travel planning request sent by a client, the travel planning request carries travel application form information, and the travel application form information includes a target user identifier, a travel destination, and a travel time; From the pre-established mapping relationship between user IDs and travel standards, obtain the target travel standards corresponding to the target user ID, and retrieve multiple types of consumption data that match the business trip destination and business trip time; Obtaining historical travel behavior data belonging to the target user ID within a preset period; The behavior data is input into a pre-trained personalized travel demand recognition model, and the travel demand preference information corresponding to the target user ID is output; the pre-trained personalized travel demand recognition model is trained based on the historical behavior data collected and pre-processed in the historical travel belonging to different user IDs; According to the target travel standards and travel demand preference information, various types of consumption data are optimized to generate travel planning information that meets the target user identification and send it to the client for display.

[0076] In one embodiment, when the processor 1001 optimizes various types of consumption data according to the target travel standard and the travel demand preference information to generate travel planning information that meets the target user identifier, the processor 1001 specifically performs the following operations: Traverse the transportation consumption information, hotel consumption information and catering consumption information in the transportation consumption information list, hotel consumption information list and catering consumption information list respectively; The traversed transportation consumption information, hotel consumption information, and catering consumption information are combined to obtain joint consumption information, and the steps of traversing the transportation consumption information, hotel consumption information, and catering consumption information in the transportation consumption information list, the hotel consumption information list, and the catering consumption information list are continued until a plurality of joint consumption information are obtained after the traversal is completed; Calculate the total consumption amount of each joint consumption information; Comparing the total consumption amount of each joint consumption information with the target travel standard, so as to eliminate the joint consumption information with a consumption amount greater than the target travel standard from the multiple joint consumption information, and obtain multiple target joint consumption information; Based on multiple target joint consumption information and travel demand preference information, travel planning information that meets the target user identification is generated.

[0077] In one embodiment, when the processor 1001 generates travel planning information that matches the target user identifier according to multiple pieces of target joint consumption information and travel demand preference information, the processor 1001 specifically performs the following operations: Performing semantic analysis on the travel demand preference information to extract the demand preference keyword sequence of the user corresponding to the target user ID; The demand text is marked with the demand preference keyword sequence to guide the preset large language model and obtain prompt information; Input the prompt information and multiple target joint consumption information into a preset large language model, and use the preset large language model to find and output a preset number of target joint consumption information that meets the travel demand preference information from the multiple target joint consumption information; the preset large language model is ChatGPT; A preset amount of target joint consumption information is used as travel planning information that matches the target user identifier.

[0078] In one embodiment, when the processor 1001 generates travel planning information that matches the target user identifier according to multiple pieces of target joint consumption information and travel demand preference information, the processor 1001 specifically performs the following operations: According to the travel demand preference information, determine the importance that the user corresponding to the target user ID places on different consumption categories; According to the importance users place on different consumption categories, the default weights of each piece of consumption information in each target joint consumption information are dynamically adjusted to obtain the target weights of each piece of consumption information in each target joint consumption information. The default weights are used to represent the average preferences of users. Score each piece of consumption information in each target joint consumption information to obtain a preference score for each piece of consumption information in each target joint consumption information; The target weight of each piece of consumer information in each target joint consumption information and the preference score of each piece of consumer information in each target joint consumption information are weighted and summed to obtain the user personalized total score of each target joint consumption information; Sort multiple target joint consumption information based on the user personalized total score of each target joint consumption information; From the sorted multiple pieces of target joint consumption information, a preset number of target joint consumption information are obtained in order from large to small according to the total scores, as travel planning information that meets the target user identification.

[0079] In one embodiment, when the processor 1001 determines the importance of different consumption categories of the user corresponding to the target user identifier according to the travel demand preference information, the processor 1001 specifically performs the following operations: Analyze the first preference level of transportation consumption according to the airline information, flight type information, seat information and cabin class information included in the transportation preference information; Analyze the second preference level of hotel consumption based on the hotel star information, geographical location information, room type information and special accommodation demand information included in the accommodation preference information; Analyze the third preference level of catering consumption according to the catering type information, catering budget information and special dietary demand information included in the catering preference information; From the pre-established mapping relationship between preference levels and scores, the scores corresponding to the first preference level, the second preference level and the third preference level are respectively obtained as the degree of importance that the user corresponding to the target user identifier has on different consumption categories.

[0080] In one embodiment, when the processor 1001 dynamically adjusts the default weight of each piece of consumption information in each piece of target joint consumption information according to the user's emphasis on different consumption categories and obtains the target weight of each piece of consumption information in each piece of target joint consumption information, the processor 1001 specifically performs the following operations: Calculate the target weight of each piece of consumer information in each piece of target joint consumption information according to the default weight of each piece of consumer information in each piece of target joint consumption information and the degree of importance that users place on different consumer categories; Among them, the calculation formula of the target weight of each consumption information in each target joint consumption information is:

[0081] in, For the target joint consumption information The target weight of consumption information, is the adjustment factor, which is used to control the sensitivity of weight adjustment. It is Consumption information in the corresponding consumption category The degree of attention paid to is an index variable, which is used to traverse all consumption data during the summation process. is 3.

[0082] In one embodiment, when the processor 1001 scores each piece of consumption information in each piece of target joint consumption information to obtain a preference score for each piece of consumption information in each piece of target joint consumption information, the processor 1001 specifically performs the following operations: Get the transportation consumption items corresponding to the transportation parameters; get the hotel consumption items corresponding to the hotel parameters; get the catering consumption items corresponding to the catering parameters; The transportation consumption items, hotel consumption items and catering consumption items are numerically quantified by using a preset scoring quantification table to obtain multiple numerical values ​​corresponding to the transportation consumption items, multiple numerical values ​​corresponding to the hotel consumption items and multiple numerical values ​​corresponding to the catering consumption items; The multiple values ​​corresponding to the transportation consumption items, the multiple values ​​corresponding to the hotel consumption items, and the multiple values ​​corresponding to the catering consumption items are summed up respectively to obtain the preference score of each consumption information in each target joint consumption information.

[0083] In one embodiment, when the processor 1001 generates a pre-trained personalized travel demand recognition model, the processor 1001 specifically performs the following operations: Collect and pre-process historical behavior data from historical travels belonging to different user IDs, where the historical behavior data includes historical travel records and online operation information of users on the travel booking platform; Perform feature extraction on the preprocessed historical behavior data to obtain multiple initial features; Extract key features related to travel demand preferences from multiple initial features; The key features related to travel demand preferences are encoded using the One-Hot encoding method to obtain multiple numerical features; Associating the preprocessed historical behavior data with multiple corresponding numerical features to obtain model training samples; Based on the model training samples, a pre-trained personalized travel demand recognition model is generated.

[0084] In one embodiment, when the processor 1001 generates a pre-trained personalized travel demand recognition model according to the model training sample, the processor 1001 specifically performs the following operations: Use decision tree algorithm to create a personalized travel demand identification model; Input the model training samples into the personalized travel demand recognition model and output the model loss value; When the loss value reaches the minimum, a pre-trained personalized travel demand recognition model is generated; or, when the loss value does not reach the minimum, the step of inputting the model training sample into the personalized travel demand recognition model is continued until the model loss value reaches the minimum; wherein,

[0085] in, is the loss value of the model, is the number of model training samples, It is The travel demand preference information of multiple numerical features corresponding to the pre-processed historical behavior data, It is The predicted travel demand preference information corresponding to the preprocessed historical behavior data.

[0086] In an embodiment of the present application, on the one hand, various types of consumption data are optimized by combining target travel standards and travel demand preference information. The target travel standards control the consumption scope of travel, and the travel demand preference information is identified based on a personalized travel demand identification model obtained based on machine learning, which can comprehensively characterize the personalized needs of users. The combination of the two for comprehensive analysis can generate a travel itinerary that meets both the personalized needs of users and the corporate travel standards; on the other hand, by considering the personalized needs of users, the generated travel plan is more likely to meet the expectations of users, thereby improving user satisfaction.

[0087] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program for corporate travel planning based on artificial intelligence can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the storage medium of the program for corporate travel planning based on artificial intelligence can be a disk, an optical disk, a read-only storage memory, or a random access memory, etc.

[0088] The above disclosure is only the preferred embodiment of the present application, which certainly cannot be used to limit the scope of rights of the present application. Therefore, equivalent changes made according to the claims of the present application are still within the scope covered by the present application.

Claims

1. A corporate travel planning method based on artificial intelligence, characterized in that: Applied to the server, the method includes: Receive a travel planning request sent by a client, the travel planning request carrying business trip application form information, the business trip application form information including a target user identifier, a business trip destination, and a business trip time; From the pre-established mapping relationship between the user identifier and the travel standard, the target travel standard corresponding to the target user identifier is obtained, and multiple types of consumption data that match the business trip destination and the business trip time are retrieved; Acquire historical travel behavior data belonging to the target user identifier within a preset period; The behavior data is input into a pre-trained personalized travel demand recognition model, and the travel demand preference information corresponding to the target user identifier is output; the pre-trained personalized travel demand recognition model is trained based on the collection and pre-processing of historical behavior data in historical travels belonging to different user identifiers; According to the target travel standards and travel demand preference information, various types of consumption data are optimized to generate travel planning information that meets the target user identifier and send it to the client for display.

2. The method according to claim 1, characterized in that The multiple types of consumption data include a transportation consumption information list, a hotel consumption information list, and a catering consumption information list; The optimizing various consumption data according to the target travel standard and travel demand preference information to generate travel planning information that meets the target user identifier includes: Traversing the transportation consumption information, hotel consumption information and catering consumption information in the transportation consumption information list, hotel consumption information list and catering consumption information list respectively; The traversed transportation consumption information, hotel consumption information and catering consumption information are combined to obtain joint consumption information, and the step of traversing the transportation consumption information, hotel consumption information and catering consumption information in the transportation consumption information list, hotel consumption information list and catering consumption information list is continued until a plurality of joint consumption information are obtained after the traversal is completed; Calculate the total consumption amount of each joint consumption information; Comparing the total consumption amount of each joint consumption information with the target travel standard, so as to remove the joint consumption information with a consumption amount greater than the target travel standard from the multiple joint consumption information, and obtain multiple target joint consumption information; Travel planning information that matches the target user identifier is generated according to the multiple pieces of target joint consumption information and the travel demand preference information.

3. The method according to claim 2, characterized in that The generating of travel planning information that matches the target user identifier according to the plurality of target joint consumption information and the travel demand preference information includes: Performing semantic analysis on the travel demand preference information to extract a demand preference keyword sequence of the user corresponding to the target user identifier; Tagging the demand text for guiding the preset large language model according to the demand preference keyword sequence to obtain prompt information; Input the prompt information and the multiple target joint consumption information into a preset large language model, and search and output a preset number of target joint consumption information that meets the travel demand preference information from the multiple target joint consumption information through the preset large language model; the preset large language model is ChatGPT; The preset amount of target joint consumption information is used as travel planning information that matches the target user identifier.

4. The method according to claim 2, characterized in that: The generating of travel planning information that matches the target user identifier according to the plurality of target joint consumption information and the travel demand preference information includes: Determining, based on the travel demand preference information, the degree of importance that the user corresponding to the target user identifier places on different consumption categories; According to the importance that the user places on different consumption categories, the default weight of each consumption information in each target joint consumption information is dynamically adjusted to obtain the target weight of each consumption information in each target joint consumption information, and the default weight is used to represent the average preference of the user; Scoring each piece of consumption information in each piece of target joint consumption information to obtain a preference score for each piece of consumption information in each piece of target joint consumption information; The target weight of each piece of consumer information in each piece of target joint consumption information and the preference score of each piece of consumer information in each piece of target joint consumption information are weighted and summed to obtain a user personalized total score of each piece of target joint consumption information; sorting the plurality of target joint consumption information based on the user personalized total score of each target joint consumption information; From the sorted multiple pieces of target joint consumption information, a preset number of target joint consumption information are obtained in order from large to small according to the total scores, as the travel planning information that meets the target user identifier.

5. The method according to claim 4, characterized in that The different consumption categories include transportation consumption, hotel consumption and catering consumption; the travel demand preference information includes accommodation preference information, transportation preference information and catering preference information; Determining, based on the travel demand preference information, the degree of importance that the user corresponding to the target user identifier places on different consumption categories includes: Analyzing the first preference level of the transportation consumption according to the airline information, flight type information, seat information, and cabin class information included in the transportation preference information; Analyze the second preference level of the hotel consumption according to the hotel star information, geographical location information, room type information and special accommodation demand information included in the accommodation preference information; Analyzing the third preference level of the catering consumption according to the catering type information, catering budget information, and special dietary demand information included in the catering preference information; From the pre-established mapping relationship between preference levels and scores, the scores corresponding to the first preference level, the second preference level and the third preference level are respectively obtained as the degree of importance that the user corresponding to the target user identifier has on different consumption categories.

6. The method according to claim 4, characterized in that According to the importance that the user attaches to different consumption categories, dynamically adjusting the default weight of each consumption information in each target joint consumption information to obtain the target weight of each consumption information in each target joint consumption information includes: Calculating the target weight of each piece of consumption information in each piece of target joint consumption information according to the default weight of each piece of consumption information in each piece of target joint consumption information and the degree of importance the user places on different consumption categories; The calculation formula of the target weight of each consumption information in each target joint consumption information is: in, For the target joint consumption information The target weight of consumption information, is the adjustment factor, which is used to control the sensitivity of weight adjustment. It is Consumption information in the corresponding consumption category The degree of attention paid to is an index variable, which is used to traverse all consumption data during the summation process. is 3.

7. The method according to claim 4, characterized in that Each piece of target joint consumption information includes transportation parameters, hotel parameters and catering parameters; Scoring each piece of the consumer information in each piece of the target joint consumer information to obtain a preference score for each piece of the consumer information in each piece of the target joint consumer information includes: Obtain the transportation consumption items corresponding to the transportation parameters; obtain the hotel consumption items corresponding to the hotel parameters; obtain the dining consumption items corresponding to the dining parameters; The transportation consumption items, hotel consumption items and catering consumption items are numerically quantified by using a preset scoring quantification table to obtain multiple numerical values ​​corresponding to the transportation consumption items, multiple numerical values ​​corresponding to the hotel consumption items and multiple numerical values ​​corresponding to the catering consumption items; The multiple values ​​corresponding to the transportation consumption items, the multiple values ​​corresponding to the hotel consumption items, and the multiple values ​​corresponding to the dining consumption items are summed up respectively to obtain the preference score of each consumption information in each target joint consumption information.

8. The method according to claim 1, characterized in that Follow these steps to generate a pre-trained personalized travel demand recognition model, including: Collect and pre-process historical behavior data in historical travel belonging to different user identifiers, wherein the historical behavior data includes historical travel records and online operation information of users on the travel booking platform; Perform feature extraction on the preprocessed historical behavior data to obtain multiple initial features; Extracting key features related to travel demand preferences from the multiple initial features; The key features related to the travel demand preference are encoded using a One-Hot encoding method to obtain multiple numerical features; Associating the preprocessed historical behavior data with multiple corresponding numerical features to obtain model training samples; A pre-trained personalized travel demand recognition model is generated based on the model training samples.

9. The method according to claim 8, characterized in that The step of generating a pre-trained personalized travel demand recognition model based on the model training sample includes: Use decision tree algorithm to create a personalized travel demand identification model; Inputting the model training sample into the personalized travel demand identification model, and outputting the loss value of the model; When the loss value reaches the minimum, a pre-trained personalized travel demand identification model is generated; or, when the loss value does not reach the minimum, the step of inputting the model training sample into the personalized travel demand identification model is continued until the model loss value reaches the minimum; wherein, in, is the loss value of the model, is the number of model training samples, It is The travel demand preference information of multiple numerical features corresponding to the pre-processed historical behavior data, It is The predicted travel demand preference information corresponding to the preprocessed historical behavior data.

10. An enterprise travel planning device based on artificial intelligence, characterized in that: The device comprises: A travel planning request receiving module is used to receive a travel planning request sent by a client, wherein the travel planning request carries business trip application form information, and the business trip application form information includes a target user identifier, a business trip destination, and a business trip time; A first data acquisition module is used to acquire the target travel standard corresponding to the target user identifier from the pre-established mapping relationship between the user identifier and the travel standard, and retrieve multiple types of consumption data that meet the business trip destination and business trip time; A second data acquisition module is used to acquire the historical travel behavior data belonging to the target user identifier within a preset period; A model processing module, used for inputting the behavior data into a pre-trained personalized travel demand recognition model, and outputting travel demand preference information corresponding to the target user identifier; the pre-trained personalized travel demand recognition model is trained based on the collection and pre-processing of historical behavior data in historical travels belonging to different user identifiers; The travel planning information output module is used to optimize various types of consumption data according to the target travel standards and travel demand preference information, generate travel planning information that meets the target user identification, and send it to the client for display.

Images