Business hotel real-time reservation intelligent reminding method and related device

Abstract

The invention belongs to the field of travel hotel reservation reminding, and discloses a travel hotel real-time reservation intelligent reminding method and a related device, and the method comprises the steps: extracting travel features in user traffic order information, inputting a pre-trained user reservation time difference prediction model, and predicting the travel reservation time difference; and the optimal reminding time is calculated in combination with the countdown of the current travel, and hotel reservation reminding information is actively sent. The user reservation time difference prediction model is trained based on historical travel data of the user, and a reservation behavior mode of the user can be learned in a personalized manner; through dynamic matching of the predicted time difference and countdown, the system intelligently triggers reminding before the user possibly lags behind booking, it is ensured that the intervention opportunity is accurate, and the problem that an existing system cannot push reminding in time is solved. By adopting the method, a hotel reservation lag phenomenon is effectively reduced, plan interruption and extra cost caused by no room in a destination hotel in a business trip peak period are avoided, and the satisfaction degree of a user to a business travel platform and the overall service experience are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of artificial intelligence, specifically to the field of business travel hotel reservation reminders, and particularly relates to a real-time reservation intelligent reminder method and related device for business travel hotels. Background Art

[0002] In the process of modern enterprise digital management, the business travel management system has become an important tool to improve enterprise operation efficiency. When users are on business trips, attending meetings, training, etc., they can initiate an online business trip application through the system. After being approved by the leader, they can freely choose the transportation route, transportation mode, and hotel, and the whole process is convenient and efficient. However, the reservation of transportation and hotels often needs to be redirected to an external reservation platform to complete, and some problems have gradually emerged in this operation mode in practical applications.

[0003] Currently, when users use the business travel management system, there is a common phenomenon that they are active in transportation reservation but lag in hotel reservation. Since business trip applications are usually submitted in advance, users are used to choosing a hotel after arriving at the business trip destination. This behavior easily leads to the situation that there are no available rooms in the hotels near the business trip location during the peak business trip period. Especially for business travel users who are used to choosing hotels near the business trip destination, this way of lagging reservation not only affects the travel plan but also may increase the business travel cost due to urgently looking for alternative hotels, reducing the satisfaction of users with the business travel platform.

[0004] Therefore, the existing business travel management methods cannot timely push hotel reservation reminders before users arrive at the business trip destination, affecting user experience and service satisfaction. Summary of the Invention

[0005] The present invention provides a real-time reservation intelligent reminder method and related device for business travel hotels. By using this method, hotel reservation reminders can be timely pushed to users before they arrive at the business trip destination, thereby improving user experience and service satisfaction.

[0006] To achieve the above purpose, the present invention adopts the following technical solutions: In the first aspect, the present invention provides a real-time reservation intelligent reminder method for business travel hotels, including: Extracting the to-be-traveled feature information based on the user transportation order information of the user to be reminded; Inputting the to-be-traveled feature information into a pre-trained user reservation time difference prediction model to predict the travel reservation time difference; wherein, the user reservation time difference prediction model is trained based on the historical travel feature information of historical users; Confirming the hotel reservation reminder time based on the travel reservation time difference and the current travel countdown; Sending a business travel hotel reservation reminder message to the user to be reminded based on the hotel reservation reminder time.

[0007] A further improvement of the present invention is that before extracting the to-be-traveled characteristic information based on the user traffic order information of the to-be-reminded user, it further includes: Based on the offline data of the big data cluster, using the real-time data query method, screen the users who have passed the business trip application form approval and have not booked hotels, and use the screened users as the to-be-reminded users.

[0008] A further improvement of the present invention is that extracting the to-be-traveled characteristic information based on the user traffic order information of the to-be-reminded user includes: Obtain the user traffic order information of the to-be-reminded user; Analyze the user traffic order information of the to-be-reminded user; Extract the to-be-traveled characteristic information, where the to-be-traveled characteristic information includes the departure time, arrival time, destination hotel information, and the remaining number of hotel nights of the to-be-reminded user.

[0009] A further improvement of the present invention is that confirming the hotel reservation reminder time based on the travel reservation time difference and the current travel countdown includes: Obtain the travel reservation time difference and the current travel countdown; the travel reservation time difference is the difference between the hotel reservation time and the time when the business trip application form is approved; Judge the travel reservation time difference and the current travel countdown: If the current travel countdown is less than the travel reservation time difference, use the current moment as the hotel reservation reminder time and immediately send a business travel hotel reservation reminder message to the to-be-reminded user; otherwise, use the difference between the current travel countdown and the travel reservation time difference as the hotel reservation reminder time.

[0010] A further improvement of the present invention is that before sending a business travel hotel reservation reminder message to the to-be-reminded user based on the hotel reservation reminder time, it further includes: Before reaching the hotel reservation reminder time, if the to-be-reminded user cancels the user traffic order information or refunds the ticket, cancel sending the business travel hotel reservation reminder message to the to-be-reminded user.

[0011] A further improvement of the present invention is that after sending a business travel hotel reservation reminder message to the to-be-reminded user based on the hotel reservation reminder time, it further includes: If the to-be-reminded user makes a flight change, re-predict the new travel reservation time difference according to the flight-changed user traffic order information, so as to determine the new hotel reservation reminder time based on the new travel reservation time difference and the travel countdown, and send a business travel hotel reservation reminder message to the to-be-reminded user according to the new hotel reservation reminder time.

[0012] A further improvement of the present invention lies in that, before sending the business travel hotel reservation reminder information to the user to be reminded based on the hotel reservation reminder time, it further includes: Verifying the business travel hotel reservation reminder information to be sent; Sending the verified business travel hotel reservation reminder information to the user to be reminded via an in-site message.

[0013] A further improvement of the present invention lies in that the basic model of the user reservation time difference prediction model adopts a deep neural network model, a LightGBM model, an Xgboost model or a Catboost model; when the basic model adopts a deep neural network model, the specific structure includes: an input layer with multiple feature nodes; a first hidden layer with 128 nodes, using the ReLU activation function and a Dropout layer of 0.3; a second hidden layer with 64 nodes, using the ReLU activation function and a Dropout layer of 0.2; and an output layer with 1 node, using a linear activation function for outputting the predicted value.

[0014] In a second aspect, the present invention further provides a real-time business travel hotel reservation intelligent reminder system, including: A feature extraction module, configured to extract the to-be-traveled feature information based on the user's transportation order information of the user to be reminded; A prediction module, configured to input the to-be-traveled feature information into a pre-trained user reservation time difference prediction model to predict the travel reservation time difference; wherein, the user reservation time difference prediction model is trained based on the historical travel feature information of historical users; A time confirmation module, configured to confirm the hotel reservation reminder time based on the travel reservation time difference and the current travel countdown; A reminder module, configured to send the business travel hotel reservation reminder information to the user to be reminded based on the hotel reservation reminder time.

[0015] The present invention further provides a real-time hotel reservation intelligent reminder device, including: A memory, configured to store a computer program; A processor, configured to implement the steps of the above-mentioned real-time business travel hotel reservation intelligent reminder method when executing the computer program.

[0016] Compared with the prior art, the present invention has the following beneficial effects: The present invention provides a method for intelligent reminder of real-time reservation of business travel hotels. By extracting travel characteristics from the user's transportation order information, inputting them into a pre-trained user reservation time difference prediction model to predict the travel reservation time difference, and then combining the current travel countdown to calculate the optimal reminder time, and actively sending hotel reservation reminder information. The user reservation time difference prediction model is trained based on the user's historical travel data and can personalized learn the user's reservation behavior pattern; through the dynamic matching of the predicted time difference and the countdown, the system can intelligently trigger a reminder before the user may lag in making a reservation, ensuring the accuracy of the intervention timing and solving the problem that the existing system cannot push reminders in time. Using this method effectively reduces the phenomenon of lagging hotel reservations, avoids plan interruptions and additional costs caused by the lack of rooms in destination hotels during peak business travel periods, and improves the user's satisfaction with the business travel platform and the overall service experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 FIG. is a data flow diagram related to the method for intelligent reminder of real-time reservation of business travel hotels provided by an embodiment of the present invention; Figure 2 FIG. is an architecture diagram of the intelligent reminder system for real-time reservation of business travel hotels provided by an embodiment of the present invention; Figure 3 FIG. is an execution flow diagram of the method for intelligent reminder of real-time reservation of business travel hotels provided by an embodiment of the present invention; Figure 4 FIG. is a schematic diagram of the change of reminder status provided by an embodiment of the present invention; Figure 5 FIG. is a basic model structure diagram of the user reservation time difference prediction model provided by an embodiment of the present invention; Figure 6 FIG. is an application logic diagram of the user reservation time difference prediction model provided by an embodiment of the present invention; Figure 7 FIG. is a flow chart of a method for intelligent reminder of real-time reservation of business travel hotels provided by an embodiment of the present invention; Figure 8 FIG. is a schematic structural diagram of an intelligent reminder system for real-time reservation of business travel hotels provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] To further understand the content of the present invention, the following provides a detailed description of the present invention in combination with the accompanying drawings and specific embodiments. It should be understood that the embodiments are only for explaining the present invention and not for limiting it.

[0019] For the convenience of better understanding the technical solution, the technical terms related to the present invention are explained as follows: Big Data ETL Technology: A technology that processes data from the source end through extraction (Extract), transformation (Transform), and loading (Load) to the destination end, which is a key link in the big data processing process.

[0020] Prediction Algorithm: A class of mathematical models and technical methods used to predict future trends or results based on historical data and relevant information, which are widely applied in multiple fields. Commonly used ones include regression analysis, time series analysis, machine learning algorithms, and deep learning algorithms.

[0021] Data Cache TTL (Time-To-Live) Technology: A mechanism for managing the expiration period of cached data. It sets a specific survival time for each data item in the cache, and when this time expires, the data item will be automatically removed from the cache.

[0022] In-site Message Technology: Refers to the technical means within a website or application to enable information transmission and communication between users and between users and the system. Generally, it is divided into links such as message storage, message push, and message display. The main key technologies are network communication, information security, etc. Currently, common scenarios include e-commerce platforms, enterprise office platforms, social platforms, etc. In this embodiment, it is used to push intelligent reminder messages to business travel users.

[0023] Neural Network Model: A computational model that mimics the structure and function of the human nervous system, aiming to achieve the processing, analysis, and learning of complex data by simulating the connection and signal transmission mechanisms between neurons.

[0024] LightGBM (Light Gradient Boosting Machine): An efficient machine learning algorithm based on the gradient boosting framework, developed by Microsoft, aiming to process large-scale data at a faster speed and with lower memory consumption while maintaining or improving the model accuracy.

[0025] XGBoost: Full name eXtreme Gradient Boosting, is an efficient integrated learning model based on the gradient boosting decision tree algorithm.

[0026] CatBoost: Full name Categorical Boosting, is a machine learning algorithm based on gradient boosting decision trees.

[0027] Hive: A data warehouse tool based on Hadoop, used for storing, querying, and analyzing large-scale structured data stored in the Hadoop distributed file system; among them, Hadoop is an open-source framework for large-scale data processing.

[0028] Redis: Redis (Remote Dictionary Server) is an open-source, high-performance key-value storage system with rich features and a wide range of application scenarios.

[0029] The candidate question recommendation method provided in this embodiment will be described in detail below with reference to the accompanying drawings: This method is a technical application in a new scenario, targeting the scenario where users forget to book or do not book a hotel during business trips. In the original situation, the user initiates a business trip application form -> the application form is approved -> the user travels -> hotel order -> the user goes on a business trip -> the journey ends -> business trip expense settlement -> business trip expense reimbursement. On this basis, this method improves the hotel reservation process, optimizes the order placement link during the user's travel, and completes the reminder for hotel reservation, thus locking in the hotel order link in advance to prevent no rooms available upon arrival due to late reservation, especially for business travelers who habitually book hotels near their business trip destinations. In this embodiment, by combining the user's application form and traffic order details, through big data and time series prediction algorithms, the hotel reservation reminder time is predicted, and reminder messages are pushed in a timely manner according to the predicted hotel reservation reminder time to achieve hotel reservation reminder.

[0030] This embodiment provides a real-time hotel reservation intelligent reminder method for business travel, and the specific steps are as follows: S1. Data collection: By collecting the user's historical travel orders, historical hotel orders, user's real-time application form information, user's real-time traffic order information, the user's hotel reservation information and traffic information under the offline data of the big data cluster, the user's behavior characteristics and order situation are obtained. The real-time + offline data combination method is used to obtain the user's main characteristics, and then it is judged whether the user needs a hotel reservation reminder. The data collection process is as Figure 1 shown, Figure 1 In the figure, first, the user's real-time application form information is collected from MySQL. MySQL is an open-source relational database management system; through real-time data query of Hive, the historical application form and historical traffic order information are obtained; the ETL technology is used to process the user's real-time application form information, historical application form, user's real-time traffic order information and user's real-time traffic order information, and then real-time data analysis is carried out to screen out the users who have passed the business trip application form approval and have not booked a hotel, and transmit the data corresponding to the users who need hotel reservation reminder to the subsequent process. At the same time, the above data information that has passed data verification is transmitted back to MySQL for storage as the subsequent historical travel characteristic information.

[0031] S2. Feature construction: Obtain the basic information of the user, the basic information of the user's affiliated organization, the time interval between transportation and hotel bookings in the user's historical orders, etc., as well as the departure and arrival times of the user's train or flight tickets, the business trip location information in the user's application form, the hotel information at the business trip location, the remaining number of hotel nights at the user's location, the time to transportation tickets, and other basic information. Extract feature information based on the above information. Among them, the basic information of the user, the basic information of the user's affiliated organization, the departure and arrival times of the user's train or flight tickets, the business trip location information in the user's application form, the time to transportation tickets, and other above information can be directly extracted from the user's transportation order information; the time interval between transportation and hotel bookings in the user's historical orders is obtained based on the database; hotel-related information such as hotel information at the business trip location and the remaining number of hotel nights at the user's location is obtained through analysis of the destination information extracted from the user's transportation order information.

[0032] S3. Training of the user reservation time difference prediction model: In this embodiment, the user reservation time difference prediction model uses a time series algorithm to predict the hotel reservation reminder time; the time series algorithm preferably uses algorithms such as deep learning network / LightGBM / Xgboost / Catboost to complete the prediction of data indicators, and uses the classification prediction regression method for time prediction. The specific construction and training process of the user reservation time difference prediction model is as follows: Train the user reservation time difference prediction model using historical travel feature information, including: Obtain the historical travel feature information of historical users, where the historical travel feature information includes the departure time, arrival time, destination hotel information, remaining number of hotel nights, and historical information corresponding to the remaining time until departure of the user to be reminded, as well as the historical travel reservation time difference; Use a deep neural network, LightGBM model, Xgboost model, or Catboost model as the initial model of the user reservation time difference prediction model; among them, the basic model of this prediction model is a deep neural network, and the specific framework structure is as Figure 5 shown, and the specific structure and functions include: The input layer, as the data entry of the model, contains multiple feature nodes, denoted by n-features; the specific quantity is determined by the feature dimension of the input data and is responsible for receiving the original feature data. The first hidden layer has 128 nodes; it uses the ReLU (Rectified Linear Unit) activation function and is responsible for enhancing the model's ability to fit complex relationships and performing a preliminary non-linear transformation on the input data. The Dropout layer (0.3) randomly "drops" some neurons in the first hidden layer with a probability of 0.3, reducing the risk of overfitting, making the model more robust, and playing a regularization role during training. The second hidden layer contains 64 nodes and also uses the ReLU activation function to further perform a non-linear transformation on the data processed by Dropout and extract more abstract features. The Dropout layer (0.2) randomly drops the neurons in the second hidden layer with a probability of 0.2 to continue regularization and prevent the model from overly relying on specific neurons. The output layer has 1 node and uses a linear activation function to output the predicted value. In this embodiment, the predicted value is the learning objective of the user's booking time difference prediction model, that is, the travel booking time difference.

[0033] The initial model of the user's booking time difference prediction model is trained using historical travel feature information to obtain a trained user's booking time difference prediction model.

[0034] The application logic of the user's booking time difference prediction model specifically includes the following steps: Step 1: Determine the learning objective of the model.

[0035] In this embodiment, the learning objective of the model is the travel booking time difference, and the specific formula is Travel booking time difference = Hotel booking time - Completion time of the approval of the user's business trip application form; among them, the completion time of the approval of the user's business trip application form is the time when the business trip application form is approved.

[0036] Step 2: Determine the offline or off-line learning process of the model.

[0037] The user's booking time difference prediction model in this embodiment is a supervised learning. It is necessary to collect labeled training data offline, and the label is the travel booking time difference of the user. All the data generated during the period from when the user submits the application form to the final hotel booking can be obtained and processed into corresponding features, that is, historical travel feature information, through an offline process. Finally, a regression prediction task is performed using the supervised label to predict the time difference between the completion of the approval and the hotel booking of the user. Finally, RMSE / MAE / R 2The model is evaluated using indicators. Among them, RMSE (Root Mean Squared Error) is the root mean square error, which is the square root of the mean of the squared errors between the predicted values and the true values. MAE (Mean Absolute Error) is the mean absolute error, which is the average of the absolute errors between the predicted values and the true values. R 2 (R-squared), that is, the coefficient of determination, is used to measure the goodness of fit of the model to the data. Thus, a prediction model for the user's booking time difference is trained.

[0038] Step 3: Determine the online inference service of the model.

[0039] After the model training is completed, the trained model is loaded through the model server. After the user submits an application form, user data records are automatically collected online, formed into features and fed into the model, and the model gives the prediction result, that is, the travel booking time difference.

[0040] Step 4: Execute the model application logic.

[0041] Generally, the traditional method is to compare multiple times with each other, and then all users fix a threshold and send uniformly according to this threshold. In this embodiment, a dynamic threshold sending method is adopted, and the travel booking time difference for this trip is predicted according to the user's habits. For example Figure 7 As shown, the specific process is as follows: 1. Start. The user logs in to the APP, and the APP (Application, that is, the application program) embeds the intelligent real-time booking reminder method for business travel hotels provided in this embodiment.

[0042] 2. Check whether the application form has been approved. If not, the process ends; if the business trip application form has been approved, start the user booking time difference prediction model to predict the travel booking time difference for this trip.

[0043] 3. Obtain the user's habitual booking time as the trigger threshold T1, and detect the user's business trip countdown time (travel countdown) T2.

[0044] 4. Judge whether T2 is greater than T1. If not, immediately send an in-site message to remind the user to book a hotel, and the process ends; if so, enter the countdown service.

[0045] 5. Continuously judge whether T2 - T1 is equal to 0. If not, continue the countdown service loop; if so, immediately send an in-site message to send a business travel hotel booking reminder message to the user to be reminded, and the process ends.

[0046] Exemplarily, this embodiment also provides a business travel hotel real-time booking intelligent reminder system for implementing the steps of the above method. The architecture of this system is as Figure 2As shown in the figure, the system will calculate based on real-time user traffic order information as the trigger condition. After passing through the model, the real-time data enters the intelligent reminder system, and the reminder time is stored in the Redis cache to determine when the user needs to be reminded. Before the user is reminded, the system needs to confirm whether the user has been reminded, the number of times the user has been reminded, and whether the user has booked a hotel, etc. Therefore, the system needs to design the control of the user reminder status, and also needs to perform operations such as timed deletion of invalid user information in the TTL manner. This process is all handled during the intelligent reminder process.

[0047] To implement the above functions, the specific technical route (execution process) is as Figure 3 shown, specifically including: First, the real-time situation of the user traffic order is used as the trigger starting point for the entire system response; second, the traffic and hotel data storage module records the user's valid reservation and cancellation information, etc.; third, the offline data includes the user's historical orders, historical application form situations, historical hotel reservation information, etc., which are the T+1 precipitation data of the real-time data, where T refers to the current data generation date; fourth, the data ETL module is used to clean the user's valid information through the big data data warehouse and perform funnel analysis, etc.; fifth, the business trip application form and traffic order data are passed in as real-time data for real-time analysis and feature combination; sixth, the model module is obtained through training of historical data and can predict the time difference from the travel time in the traffic order to the time when the reminder is needed; seventh, the intelligent reminder system will record the reminder times, reminder times, whether a reminder is needed, data verification, etc. and send a reminder signal to the in-site message; eighth, the in-site message reminder module will send it to the user who needs to be reminded according to the user's account configuration information template.

[0048] Combined with Figure 3 shown, the specific implementation steps of the system include: Step 1: Offline data from the big data cluster, query real-time data through Hive, and screen users whose application forms have passed but have no hotel reservation orders through ETL.

[0049] Step 2: Obtain the user's subscription historical features from the offline data of the big data cluster and obtain the historical travel reservation time difference.

[0050] Step 3: After the user traffic order is created and placed, parse the user traffic order information, respond to the reservation reminder service, and promptly obtain the user traffic order information as the trigger point for the system execution steps.

[0051] Step 4: Combine the real-time and offline features of users (including feature data such as users' basic information, travel modes, journey lengths, time differences between users' previous transportation and hotel orders, time periods of users' travel, numbers of hotels in users' business trip locations, users' business trip standards, time differences between users and business trip dates, etc.), establish a prediction model for the time difference of users' booking, predict the time difference of the current travel booking, and then combine the current travel countdown to determine the hotel booking reminder time.

[0052] Step 5: As Figure 4 shown, establish a mechanism for sending reminder data and a state change mechanism for data disappearance. There will be many state changes between the issuance of transportation orders (ticket issuance) and users' actual hotel bookings, and the status of in-site messages needs to be continuously adjusted. The system will automatically determine whether to send business travel hotel booking reminder messages and whether to re-remind, etc.

[0053] Step 6: Verify the business travel hotel booking reminder information to be sent. For the information that passes the verification, send it in the form of in-site messages to achieve hotel booking reminders; among them, it is preferably to send in-site messages within the App. This process involves the design and switch of in-site message templates, as well as front-end display, etc.

[0054] Step 7: Create log data, automatically collect and summarize log data for hotel booking reminders, and these log data are used for subsequent model verification and evaluation.

[0055] As a preferred solution of this embodiment, regularly clean up expired data, and perform expiration processing through data caching to achieve real-time update.

[0056] Exemplarily, in this embodiment, during the above data extraction and processing process, some data needs to be prepared in advance, such as the hotel distribution in cities, the geographical locations of hotels themselves, city codes, etc.

[0057] It can be seen that this embodiment provides a method for intelligent reminder of real-time booking of business travel hotels, which has the following advantages: First, use big data technology to obtain users who potentially need to be reminded; Second, establish a data model based on historical transportation and hotel order information to predict and obtain the hotel booking reminder time; Third, use scheduled tasks and dynamic state management methods to perform intelligent reminders for users.

[0058] Exemplarily, as Figure 7 shown, this embodiment also provides a method for intelligent reminder of real-time booking of business travel hotels, including: Extract the to-be-traveled feature information based on the user transportation order information of the users to be reminded; Input the to-be-traveled feature information into a pre-trained user booking time difference prediction model to predict the travel booking time difference; wherein, the user booking time difference prediction model is trained based on the historical travel feature information of historical users; Based on the travel booking time difference and the current travel countdown, confirm the hotel booking reminder time; Based on the hotel booking reminder time, send a business hotel booking reminder message to the user to be reminded.

[0059] In this embodiment, before extracting the to-be-traveled feature information based on the user traffic order information of the user to be reminded, it further includes: based on the offline data of the big data cluster, using the real-time data query method, screening the users who have passed the business trip application form approval and have not booked hotels, and taking the screened users as the users to be reminded.

[0060] This method realizes the accurate positioning of target users through the offline data of the big data cluster and real-time query, systematically identifies the group of users to be reminded who truly have a hotel booking need, that is, those who have been approved for business trips, have their transportation arranged but have not booked hotels, avoids sending invalid reminders to irrelevant users, significantly improves the operation efficiency and resource utilization rate of the reminder system, and ensures that the reminder service focuses on the core needs.

[0061] In this embodiment, extracting the to-be-traveled feature information based on the user traffic order information of the user to be reminded includes: Obtain the user traffic order information of the user to be reminded; Analyze the user traffic order information of the user to be reminded; Extract the to-be-traveled feature information, where the to-be-traveled feature information includes the departure time, arrival time, destination hotel information, and remaining hotel room nights of the user to be reminded.

[0062] The extracted feature information, especially the destination hotel information and real-time room availability, can provide high-value inputs for the subsequent prediction model, making the prediction result (booking time difference) more in line with the actual preferences of users (such as the habit of staying in hotels in a certain area) and the current market conditions (such as hotel tightness), thereby significantly improving the accuracy of reminder timing judgment and the pertinence of reminder content. For example, it may prompt the user to pay attention to the room availability of the favorite hotel.

[0063] In this embodiment, confirming the hotel booking reminder time based on the travel booking time difference and the current travel countdown includes: Obtain the travel booking time difference and the current travel countdown; the travel booking time difference is the difference between the hotel booking time and the time when the business trip application form is approved; Judge the travel booking time difference and the current travel countdown: If the current travel countdown is less than the travel booking time difference, the current time will be used as the hotel booking reminder time, and the business hotel booking reminder information will be immediately sent to the user to be reminded; otherwise, the difference between the current travel countdown and the travel booking time difference will be used as the hotel booking reminder time.

[0064] This method provides a dynamic and adaptive reminder triggering mechanism. When it is detected that the user's trip is approaching and the customary booking window is not open (countdown < time difference), the system can intervene immediately to prevent the user from missing the best booking opportunity; otherwise, a reasonable advance reminder is set to ensure timeliness and avoid disturbing the user too early. This ensures that the reminder is delivered accurately at the critical moment, effectively dealing with the risk of urgent travel or user forgetfulness.

[0065] In this embodiment, before sending the business hotel reservation reminder information to the user to be reminded based on the hotel reservation reminder time, it also includes: before the hotel reservation reminder time arrives, if the user to be reminded cancels the user's transportation order information or refunds the ticket, then canceling the sending of the business hotel reservation reminder information to the user to be reminded.

[0066] In this way, the system's fault tolerance and user experience are enhanced. When it is detected that the user's trip has been cancelled, the system automatically cancels the corresponding hotel reminder, effectively avoiding sending invalid or even confusing interference information to users whose trips have expired, maintaining the accuracy and professionalism of the reminder service, and enhancing users' trust in the platform.

[0067] In this embodiment, after sending the business hotel reservation reminder information to the user to be reminded based on the hotel reservation reminder time, the method further includes: If the user to be reminded changes his or her ticket, the new travel booking time difference will be re-predicted based on the user's transportation order information after the change, so as to determine the new hotel booking reminder time based on the new travel booking time difference and the travel countdown, and the business hotel booking reminder information will be sent to the user to be reminded based on the new hotel booking reminder time.

[0068] This approach ensures the continuity and adaptability of the service. When the user's itinerary changes (rebooking), the system can respond dynamically and recalculate the forecast and reminder time based on the updated traffic information, ensuring that the hotel reminder is always synchronized with the user's latest itinerary, avoiding the problem of the original reminder being invalid or misplaced due to itinerary changes, and improving the robustness of the service and user satisfaction in dynamically changing scenarios.

[0069] In this embodiment, before sending the business hotel reservation reminder information to the user to be reminded based on the hotel reservation reminder time, the method further includes: Verify the business hotel reservation reminder information to be sent; The verified business hotel reservation reminder information will be sent to the users who need to be reminded via the in-site message.

[0070] By performing information verification before sending reminders and specifying the method of sending via in-site messages, the reliability and delivery rate of the reminder service are improved. The information verification process helps filter out incorrect or non-compliant reminder content, ensuring that the information received by users is accurate. Clearly using in-site messages as the sending channel utilizes the highly reliable and traceable message system built into the platform, improving the guarantee of successful information reaching users and facilitating convenient processing by users within the same platform, thus optimizing the overall interaction experience.

[0071] In this embodiment, the basic model of the user reservation time difference prediction model adopts a deep neural network model, a LightGBM model, an Xgboost model, or a Catboost model; when the basic model adopts a deep neural network model, the specific structure includes: an input layer with multiple feature nodes; a first hidden layer with 128 nodes, using the ReLU activation function and a Dropout layer of 0.3; a second hidden layer with 64 nodes, using the ReLU activation function and a Dropout layer of 0.2; and an output layer with 1 node, using linear activation for outputting the predicted value As Figure 8 shown, this embodiment also provides a business travel hotel real-time reservation intelligent reminder system, including: a feature extraction module for extracting to-be-traveled feature information based on the user's transportation order information of the user to be reminded; a prediction module for inputting the to-be-traveled feature information into a pre-trained user reservation time difference prediction model to predict the travel reservation time difference; wherein, the user reservation time difference prediction model is trained based on the historical travel feature information of historical users; a time confirmation module for confirming the hotel reservation reminder time based on the travel reservation time difference and the current travel countdown; a reminder module for sending a business travel hotel reservation reminder message to the user to be reminded based on the hotel reservation reminder time.

[0072] The present invention also provides a hotel real-time reservation intelligent reminder device, including: a memory for storing a computer program; a processor for implementing the steps of the business travel hotel real-time reservation intelligent reminder method when executing the computer program.

[0073] The present invention also provides a computer program product, including computer program / instructions, which implement the steps of the business travel hotel real-time reservation intelligent reminder method when executed by a processor.

[0074] When the processor executes the computer program, it implements the steps of the above-mentioned real-time reservation intelligent reminder for business travel hotels. For example: based on the user traffic order information of the user to be reminded, the to-be-traveled feature information is extracted; the to-be-traveled feature information is input into a pre-trained user reservation time difference prediction model to predict the travel reservation time difference; wherein, the user reservation time difference prediction model is trained based on the historical travel feature information of historical users; based on the travel reservation time difference and the current travel countdown, the hotel reservation reminder time is confirmed; based on the hotel reservation reminder time, a business travel hotel reservation reminder message is sent to the user to be reminded.

[0075] Exemplarily, the computer program can be divided into one or more modules / units, and the one or more modules / units are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units can be a series of computer program instruction segments capable of completing preset functions, and the instruction segments are used to describe the execution process of the computer program in the real-time reservation intelligent reminder device for business travel hotels. For example, the computer program can be divided into a feature extraction module, a prediction module, a time confirmation module, and a reminder module; the specific functions of each module are as follows: the feature extraction module is used to extract the to-be-traveled feature information based on the user traffic order information of the user to be reminded; the prediction module is used to input the to-be-traveled feature information into a pre-trained user reservation time difference prediction model to predict the travel reservation time difference; wherein, the user reservation time difference prediction model is trained based on the historical travel feature information of historical users; the time confirmation module is used to confirm the hotel reservation reminder time based on the travel reservation time difference and the current travel countdown; the reminder module is used to send a business travel hotel reservation reminder message to the user to be reminded based on the hotel reservation reminder time.

[0076] The real-time reservation intelligent reminder device for business travel hotels can be a computing device such as a desktop computer, a notebook, a palm computer, and a cloud server. The real-time reservation intelligent reminder device for business travel hotels may include, but is not limited to, a processor and a memory. Those skilled in the art can understand that the above are examples of the real-time reservation intelligent reminder device for business travel hotels, and do not constitute a limitation on the real-time reservation intelligent reminder device for business travel hotels. It may include more components than the above, or combine some components, or different components. For example, the real-time reservation intelligent reminder device for business travel hotels may further include an input / output device, a network access device, a bus, etc.

[0077] The so-called processor may be a Central Processing Unit (CPU), or may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The processor is the control center of the real-time reservation intelligent reminder for business and travel hotels, and connects various parts of the real-time reservation intelligent reminder device for business and travel hotels through various interfaces and lines.

[0078] The memory can be used to store the computer programs and / or modules. The processor realizes various functions of the real-time reservation intelligent reminder device for business and travel hotels by running or executing the computer programs and / or modules stored in the memory, and by calling the data stored in the memory.

[0079] The memory may mainly include a program storage area and a data storage area. Among them, the program storage area can store an operating system, application programs required for at least one function (such as a sound playback function, an image playback function, etc.); the data storage area can store data created according to the use of the mobile phone (such as audio data, phone book, etc.). In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, at least one magnetic disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0080] The present invention also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the real-time reservation intelligent reminder method for business and travel hotels are realized.

[0081] If the modules / units integrated in the real-time reservation intelligent reminder system for business and travel hotels are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium.

[0082] Based on such understanding, all or part of the processes in the above-mentioned intelligent reminder method for real-time booking of business travel hotels of the present invention can also be completed by instructing relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-mentioned intelligent reminder method for real-time booking of business travel hotels can be realized. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file or preset intermediate form, etc.

[0083] The computer-readable storage medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc.

[0084] It should be noted that the content included in the computer-readable storage medium can be appropriately increased or decreased according to the requirements of legislation and patent practice within the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable storage medium does not include electrical carrier signals and telecommunication signals.

[0085] The above-mentioned embodiments are only one of the implementation manners capable of realizing the technical solution of the present invention. The scope of protection required by the present invention is not only limited by this embodiment, but also includes any changes, substitutions and other implementation manners that are easily conceivable by those skilled in the art within the technical scope disclosed by the present invention.

[0086] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: it is still possible to modify the specific implementation manners of the present invention or make equivalent substitutions, and any modification or equivalent substitution that does not depart from the spirit and scope of the present invention should be covered within the scope of the protection of the claims of the present invention.

Claims

1. A real-time reservation intelligent reminder method for business and travel hotels, characterized in that, include: Based on the user transportation order information of the user to be reminded, extract the characteristic information of the trip to be made; Inputting the characteristic information of the trip to be traveled into a pre-trained user booking time difference prediction model to predict the travel booking time difference; wherein the user booking time difference prediction model is trained based on the historical travel characteristic information of historical users; Confirm the hotel reservation reminder time based on the travel reservation time difference and the current travel countdown; Based on the hotel reservation reminder time, send business hotel reservation reminder information to the users to be reminded.

2. The real-time reservation intelligent reminder method for business and travel hotels according to claim 1, wherein Before extracting the characteristic information of the trip to be made based on the user transportation order information of the user to be reminded, the method further includes: Based on the offline data of the big data cluster, real-time data query is used to screen users who have passed the business trip application form approval but have not booked a hotel, and the screened users are used as users to be reminded.

3. The real-time reservation intelligent reminder method for business travel hotels according to claim 1, characterized in that, The extracting of the to-be-reminded travel feature information based on the user's traffic order information includes: Get the user's transportation order information of the user to be reminded; Analyze the user's transportation order information of the user to be reminded; The characteristic information of the trip to be taken is extracted, and the characteristic information of the trip to be taken includes the departure time, arrival time, destination hotel information and remaining hotel room nights of the user to be reminded.

4. The real-time reservation intelligent reminder method for business travel hotels according to claim 1, characterized in that, The confirmation of the hotel reservation reminder time based on the travel reservation time difference and the current travel countdown includes: Obtain the travel booking time difference and the current travel countdown; the travel booking time difference is the difference between the hotel booking time and the time of approval of the business trip application form; Determine the travel booking time difference and the current travel countdown: If the current travel countdown is less than the travel booking time difference, the current time will be used as the hotel booking reminder time, and the business hotel booking reminder information will be immediately sent to the user to be reminded; otherwise, the difference between the current travel countdown and the travel booking time difference will be used as the hotel booking reminder time.

5. The real-time reservation intelligent reminder method for business travel hotels according to claim 1, wherein Before sending the business hotel reservation reminder information to the user to be reminded based on the hotel reservation reminder time, the method further includes: Before the hotel reservation reminder time arrives, if the user to be reminded cancels the user's transportation order information or refunds the ticket, the business hotel reservation reminder information sent to the user to be reminded will be cancelled.

6. The real-time reservation intelligent reminder method for business and travel hotels according to claim 1, characterized in that, After sending the business hotel reservation reminder information to the user to be reminded based on the hotel reservation reminder time, the method further includes: If the user to be reminded changes his or her ticket, the new travel booking time difference will be re-predicted based on the user's transportation order information after the change, so as to determine the new hotel booking reminder time based on the new travel booking time difference and the travel countdown, and the business hotel booking reminder information will be sent to the user to be reminded based on the new hotel booking reminder time.

7. The real-time reservation intelligent reminder method for business and travel hotels according to claim 1, wherein, Before sending the business hotel reservation reminder information to the user to be reminded based on the hotel reservation reminder time, the method further includes: Verify the business hotel reservation reminder information to be sent; The verified business hotel reservation reminder information will be sent to the users who need to be reminded via the in-site message.

8. The real-time reservation intelligent reminder method for business and travel hotels according to claim 1, characterized in that The base model of the user reservation time difference prediction model adopts a deep neural network model, a LightGBM model, an Xgboost model, or a Catboost model; when the base model adopts a deep neural network model, the specific structure includes: an input layer with multiple feature nodes; a first hidden layer with 128 nodes, using the ReLU activation function, and the Dropout layer is 0.3; a second hidden layer with 64 nodes, using the ReLU activation function, and the Dropout layer is 0.2; it also includes an output layer with 1 node, using linear activation, for outputting the predicted value.

9. A real-time booking intelligent reminder system for business and travel hotels, characterized in that, It includes: A feature extraction module, configured to extract to-be-traveled feature information based on the user traffic order information of the user to be reminded; A prediction module, configured to input the to-be-traveled feature information into a pre-trained user reservation time difference prediction model to predict the travel reservation time difference; wherein, the user reservation time difference prediction model is trained based on the historical travel feature information of historical users; A time confirmation module, configured to confirm the hotel reservation reminder time based on the travel reservation time difference and the current travel countdown; A reminder module, configured to send a business travel hotel reservation reminder message to the user to be reminded based on the hotel reservation reminder time.

10. An intelligent reminder device for real-time hotel reservation, characterized in that, It includes: A memory, configured to store a computer program; A processor, configured to implement the steps of the business travel hotel real-time reservation intelligent reminder method according to any one of claims 1-8 when executing the computer program.

Images