A Business Trip Recommendation Method and System Based on Collaborative Filtering and Recurrent Neural Network

Abstract

This application relates to the field of business travel management, and specifically relates to a business travel recommendation method and system based on collaborative filtering and recurrent neural network. The method includes: obtaining the approved business travel application form and extracting business travel information; generating an initial recommendation plan through user-user collaborative filtering and item-item collaborative filtering algorithms; retrieving the user's historical travel records, constructing a user travel profile, training a time series model of user behavior through a temporal neural network architecture, and generating a personalized travel plan through Monte Carlo simulation by the trained time series model; inputting the initial recommendation plan and the personalized travel plan into a dynamic filtering module to output a final business travel recommendation plan; pushing the final business travel recommendation plan to the user terminal, receiving the feedback information of the user, and optimizing the collaborative filtering module and the LSTM temporal analysis module based on the feedback information. It can effectively reduce itinerary delays in emergency situations and improve the business travel experience of employees.

Description

Technical Field

[0001] The present application relates to the field of business travel management, and particularly to a business travel recommendation method and system based on collaborative filtering and recurrent neural network. Background Art

[0002] In the field of business travel management, with the increasing frequency of corporate business travel activities and the diversified and personalized development of business travel needs, traditional business travel management methods have gradually exposed many problems and are difficult to meet the refined and intelligent needs of modern corporate business travel management. The specific problems are as follows:

[0003] In the traditional business travel process, data in key links such as OA approval, employee reservation, and financial reimbursement are stored independently of each other, forming information silos. According to statistics, most enterprises still adopt the mode of "employees booking by themselves + reimbursing with pasted tickets afterwards", which makes it impossible to effectively link data between departments such as administration, finance, and travel. Even if some enterprises introduce business travel platforms, the systems cannot be connected. For example, the business travel platform and the financial system are separated from each other. This situation directly leads to low efficiency in the implementation of business travel policies and serious lag in budget monitoring. Once compliance risks occur, it is very difficult to conduct effective traceability.

[0004] When employees make business travel reservations, they need to jump back and forth between multiple different platforms to compare prices, consuming a lot of time and energy. Moreover, the refund and rescheduling process is cumbersome and complex. In case of emergencies, employees are extremely likely to delay their trips due to untimely refund and rescheduling, seriously affecting work arrangements and having a poor experience. Summary of the Invention

[0005] In view of the above problems, a more advanced and intelligent business travel recommendation method is needed, which can fully consider the personalized needs and temporal characteristics of users, combine real-time information and user feedback, and realize the organic integration of multiple recommendation methods, so as to improve the accuracy, practicability and user satisfaction of business travel recommendations. Based on this, the present invention proposes a business travel recommendation method and system based on collaborative filtering and recurrent neural network.

[0006] In a first aspect, the technical solution of the present invention provides a business travel recommendation method based on collaborative filtering and recurrent neural network, including the following steps:

[0007] Obtain the approved business travel application form and extract business travel information, including destination, date, job level, and business travel standard;

[0008] Input the business travel information into the collaborative filtering module, and generate an initial recommendation plan through user-user collaborative filtering and item-item collaborative filtering algorithms;

[0009] Input the business trip information into the LSTM time series analysis module, retrieve the user's historical travel records in the recent M times, construct the user travel profile, train the time series model of the user behavior through the time series neural network architecture, and generate a personalized travel plan through Monte Carlo simulation for the trained time series model;

[0010] Input the initial recommendation plan and the personalized travel plan into the dynamic filtering module, perform secondary screening in combination with the real-time inventory of business trip suppliers, the budget constraint based on business trip standards, and the user travel profile tags, and output the final business trip recommendation plan;

[0011] Push the final business trip recommendation plan to the user side, receive the feedback information of the user, and optimize the collaborative filtering module and the LSTM time series analysis module based on the feedback information.

[0012] The combination of the collaborative filtering module and the LSTM time series analysis module can accurately analyze the business trip needs and preferences of users, improve the utilization rate of contracted hotels, reduce the situation of employees choosing high-price individual customer channels. It effectively reduces the business trip cost burden of enterprises. The automated recommendation and screening process greatly reduces the workload of manual review and reduces the financial labor cost. At the same time, the accurate recommendation reduces unnecessary refund and rescheduling, saving a large amount of funds for enterprises. The dynamic filtering module strictly screens according to business trip standards, and can effectively avoid problems such as over-standard reimbursement and non-compliant bills. Through real-time monitoring and early warning, potential violations are discovered and processed in a timely manner, reducing the economic losses caused by illegal expenditures for enterprises.

[0013] As a further limitation of the technical solution of the present invention, the steps for the collaborative filtering module to generate the initial recommendation plan include:

[0014] Calculate the user interest matching degree through cosine similarity or Pearson correlation coefficient, identify the neighbor group with similar preferences to the target user, and generate a group preference recommendation list based on the behavior data of the neighbor group;

[0015] Analyze the associated behavior sequence of the user's hotel reservation and transportation tool selection, introduce a time decay factor, and construct a user-item interaction matrix;

[0016] Adopt the modified cosine similarity algorithm to calculate the similarity between items and generate an associated item recommendation list;

[0017] Perform weighted fusion on the recommended items of the group preference recommendation list and the associated item recommendation list to form an initial recommendation plan.

[0018] As a further limitation of the technical solution of the present invention, the steps for performing weighted fusion on the recommended items of the group preference recommendation list and the associated item recommendation list to form an initial recommendation plan include:

[0019] Take the user interest matching degree as the initial weight of the recommended items in the group preference recommendation list, and take the corrected cosine similarity score as the initial weight of the recommended items in the associated item recommendation list;

[0020] Normalize the weights of the two types of recommendations to the same dimension;

[0021] Calculate the scores of group preference recommendation and associated item recommendation respectively;

[0022] Sort by the weighted total score and retain the top N items as the initial recommendation scheme.

[0023] As a further limitation of the technical solution of the present invention, the steps for the LSTM time series analysis module to generate a user personalized travel plan include:

[0024] Retrieve the user's recent M historical travel records to generate a historical behavior sequence, and the historical travel records include structured data and unstructured note information;

[0025] Parse the unstructured note information through natural language processing technology to generate user portrait tags;

[0026] Jointly input the user portrait tags and the historical behavior sequence into the trained LSTM model, and capture time series features through time decay weighted sum and attention mechanism;

[0027] Generate the probability distributions of destinations, hotels, and flights at the output layer of the LSTM model;

[0028] Generate a personalized travel plan based on the probability distribution through Monte Carlo simulation.

[0029] As a further limitation of the technical solution of the present invention, the steps for generating a personalized travel plan based on the probability distribution through Monte Carlo simulation include:

[0030] Determine the candidate solution set according to the probability distribution output by the LSTM;

[0031] Superimpose external influence factors on each candidate solution;

[0032] Conduct N random samplings. Each time of sampling, randomly select a candidate solution according to the weighted probability distribution, and calculate the comprehensive score of the probability weight + difference standard compliance + real-time inventory availability of the candidate solution;

[0033] Statistically analyze all sampling results, and output the top K solutions with the highest comprehensive scores among the selected candidate solutions, which are the personalized travel plans.

[0034] As a further limitation of the technical solution of the present invention, the steps of inputting the initial recommendation plan and the personalized travel plan into the dynamic filtering module, performing secondary screening in combination with the real-time inventory of travel suppliers, the budget constraint based on travel standards, and the user travel portrait tags, and outputting the final travel recommendation plan include:

[0035] Taking the union of the initial recommendation plan generated by the collaborative filtering module and the personalized travel plan generated by the LSTM time series analysis module;

[0036] Obtaining the real-time inventory through the supplier API interface and retrieving the budget constraint from the enterprise travel standard database;

[0037] Performing compliance filtering based on the real-time inventory, budget constraint, and user portrait tags, and excluding the items that exceed the standard;

[0038] Generating the final travel recommendation plan for the options that pass the screening according to the preset weights; where the weight A of the personalized plan is greater than the initial recommendation weight B.

[0039] As a further limitation of the technical solution of the present invention, the steps of optimizing the collaborative filtering module and the LSTM time series analysis module based on the feedback information include:

[0040] Adjusting the weights of the recommended items in the group preference recommendation list and the weights of the recommended items in the associated item recommendation list in the collaborative filtering module according to the user's click and booking behaviors;

[0041] Updating the user portrait tags according to the user's click and booking behaviors, and further updating the attention mechanism parameters of the LSTM model in the LSTM time series analysis module.

[0042] In a second aspect, the technical solution of the present invention also provides a travel recommendation system based on collaborative filtering and recurrent neural network, including a travel application form receiving module, a collaborative filtering module, an LSTM time series analysis module, a dynamic filtering module, and a recommendation optimization module;

[0043] The travel application form receiving module is used to obtain the approved travel application form and extract travel information, including the destination, date, job level, and travel standards;

[0044] The collaborative filtering module is used to generate an initial recommendation plan based on the travel information through user-user collaborative filtering and item-item collaborative filtering algorithms;

[0045] The LSTM time series analysis module is used to retrieve the user's recent M historical travel records according to the travel information, construct a user travel portrait, train a time series model of the user's behavior through a time series neural network architecture, and generate a personalized travel plan through Monte Carlo simulation for the trained time series model;

[0046] A dynamic filtering module, which is used to perform secondary screening based on the initial recommendation plan and the personalized travel plan, in combination with the real-time inventory of travel suppliers, the budget constraints based on travel standards, and the user travel portrait tags, and output the final travel recommendation plan;

[0047] A recommendation optimization module, which is used to push the final travel recommendation plan to the user side, receive the feedback information of the user, and optimize the collaborative filtering module and the LSTM time series analysis module based on the feedback information.

[0048] As a further limitation of the technical solution of the present invention, the steps for the collaborative filtering module to generate the initial recommendation plan include:

[0049] Calculate the user interest matching degree through cosine similarity or Pearson correlation coefficient, identify the neighbor group with similar preferences to the target user, and generate a group preference recommendation list based on the behavior data of the neighbor group;

[0050] Analyze the association behavior sequence of the user's hotel reservation and transportation selection, introduce a time decay factor, and construct a user-item interaction matrix;

[0051] Adopt the modified cosine similarity algorithm to calculate the similarity between items and generate an associated item recommendation list;

[0052] Perform weighted fusion on the recommended items of the group preference recommendation list and the associated item recommendation list to form an initial recommendation plan.

[0053] As a further limitation of the technical solution of the present invention, the collaborative filtering module forming the initial recommendation plan specifically includes:

[0054] Take the user interest matching degree as the initial weight of the recommended items in the group preference recommendation list, and take the modified cosine similarity score as the initial weight of the recommended items in the associated item recommendation list;

[0055] Normalize the weights of the two types of recommendations to the same dimension;

[0056] Calculate the scores of the group preference recommendation and the associated item recommendation respectively;

[0057] Sort by the weighted total score and retain the top N items as the initial recommendation plan.

[0058] As a further limitation of the technical solution of the present invention, the steps for the LSTM time series analysis module to generate the user's personalized travel plan include:

[0059] Retrieve the user's recent M historical travel records, including structured data and unstructured note information;

[0060] Parse the unstructured note information through natural language processing technology to generate user portrait tags;

[0061] Jointly input the user portrait tags and the historical behavior sequence into the trained LSTM model, and capture the time series features through the time decay weighted sum and the attention mechanism;

[0062] Generate the probability distributions of the destination, hotel, and flight at the output layer of the LSTM model;

[0063] Generate a personalized travel plan through Monte Carlo simulation based on the probability distribution.

[0064] As a further limitation of the technical solution of the present invention, the steps for the LSTM time series analysis module to generate a personalized travel plan through Monte Carlo simulation based on the probability distribution include:

[0065] Determine the candidate solution set according to the probability distribution output by the LSTM;

[0066] Superimpose external influence factors on each candidate solution;

[0067] Conduct N random samplings. Each time of sampling, randomly select a candidate solution according to the weighted probability distribution, and calculate the comprehensive score of the probability weight + difference standard compliance + real-time inventory availability of the candidate solution;

[0068] Statistically analyze all sampling results, and output the top K solutions with the highest comprehensive scores among the selected candidate solutions, that is, the personalized travel plan.

[0069] As a further limitation of the technical solution of the present invention, the dynamic filtering module is specifically used to take the union of the initial recommendation solutions generated by the collaborative filtering module and the personalized travel plans generated by the LSTM time series analysis module;

[0070] Obtain the real-time inventory through the supplier API interface, and retrieve the budget constraints from the corporate travel standard database;

[0071] Based on the real-time inventory, budget constraints, and user portrait tags, perform compliance filtering to exclude the excessive items;

[0072] Generate the final travel recommendation solutions for the selected options according to the preset weights; where the weight A of the personalized solution is greater than the initial recommendation weight B.

[0073] As a further limitation of the technical solution of the present invention, the recommendation optimization module is specifically used to adjust the weights of the recommended items in the group preference recommendation list and the weights of the recommended items in the associated item recommendation list in the collaborative filtering module according to the user's click and booking behaviors; update the user portrait tags according to the user's click and booking behaviors, and further update the attention mechanism parameters of the LSTM model in the LSTM time series analysis module.

[0074] As can be seen from the above technical solutions, the present application has the following advantages: The business trip recommendation method based on collaborative filtering and recurrent neural network of the present invention integrates the information of approved business trip application forms and inputs them into different modules for processing, realizing the centralized analysis and utilization of data. The data interaction between each module breaks the traditional information silos, enabling departments such as administration, finance, and travel to share data, thereby improving the execution efficiency of business trip policies, realizing real-time budget monitoring, and effectively tracing compliance risks.

[0075] This method provides users with a one-stop business trip recommendation service, avoiding the cumbersome process of employees jumping between multiple platforms to compare prices. At the same time, through accurate recommendations and reasonable itinerary planning, the need for itinerary changes and cancellations is greatly reduced. Even if itinerary changes and cancellations are required, the system can respond quickly, effectively reducing itinerary delays in emergency situations and enhancing the business trip experience of employees. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] In order to more clearly illustrate the technical solutions of the present application, the accompanying drawings required for description will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0077] Figure 1 It is a schematic flowchart of the method provided by the embodiment of the present invention.

[0078] Figure 2 It is a block diagram of the system provided by the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0079] In order to make the application purpose, features, and advantages of the present application more obvious and understandable, the technical solutions protected by the present application will be clearly and completely described below by using specific embodiments and the accompanying drawings. Obviously, the embodiments described below are only some embodiments of the present application, rather than all embodiments. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments in the present application belong to the scope of protection of the present application.

[0080] As Figure 1 shown, the embodiment of the present invention provides a business trip recommendation method based on collaborative filtering and recurrent neural network, including the following steps:

[0081] S1: Obtain the approved business trip application form and extract the business trip information, including the destination, date, job level, and business trip standard;

[0082] The system establishes an interface connection with the enterprise's OA approval system. When the business trip application form completes the approval process in the OA system, the OA system will send the application form data to this business trip recommendation system through the interface. After receiving the application form data, the system uses regular expressions or data parsing tools to extract information such as the destination, date, job level, and business trip standard from the application form. For example, for the date information, a date parsing library can be used to convert the text-formatted date into a date object recognizable by the system.

[0083] The parsed fields include: destination (city code), date (departure / return timestamp), job level (such as P7 / M3), and business trip standard (JSON structure, including air ticket / hotel budget limits).

[0084] S2: Input the business trip information into the collaborative filtering module, and generate an initial recommendation plan through user-user collaborative filtering and item-item collaborative filtering algorithms;

[0085] After receiving the business trip information, the collaborative filtering module uses it as a screening condition to screen out user behavior data related to the current business trip information from the user behavior database. Execute user-user collaborative filtering and item-item collaborative filtering algorithms respectively.

[0086] S3: Input the business trip information into the LSTM time series analysis module, retrieve the user's recent M historical travel records, construct a user travel profile, train a time series model of user behavior through a time series neural network architecture, and the trained time series model generates a personalized travel plan through Monte Carlo simulation;

[0087] After receiving the business trip information, the LSTM time series analysis module retrieves the user's recent M historical travel records from the historical travel record database according to the user identification. Preprocess the historical travel records, including operations such as data cleaning and feature extraction, to construct a user travel profile. Use a time series neural network architecture (such as LSTM) to train the time series of user behavior. After training is completed, use the Monte Carlo simulation method to generate a personalized travel plan.

[0088] In the embodiment of the present invention, query the user's recent M = 12 historical orders from a distributed database (such as HBase); parse the remarks through NLP (such as extract the "prefer quiet hotel" label by the BERT model); load the pre-trained LSTM model, input the time series features + profile labels, and output the probability distribution.

[0089] S4: Input the initial recommendation plan and the personalized travel plan into the dynamic filtering module, and perform secondary screening in combination with the real-time inventory of business trip suppliers, the budget constraint based on the business trip standard, and the user travel profile labels, and output the final business trip recommendation plan;

[0090] Merge the candidate sets of the initial recommendation and the personalized plan; call the real-time inventory API of the aviation information / hotel PMS system to verify the availability; exclude out-of-tolerance items (such as room rate > rank limit), and sort them according to the weight A:B = 7:3.

[0091] S5: Push the final business travel recommendation plan to the user side, receive the feedback information of the user, and optimize the collaborative filtering module and the LSTM time series analysis module based on the feedback information.

[0092] Send the final business travel recommendation plan to the user side by means of message push, email or APP notification, etc. The user operates on the recommendation plan on the user side, such as clicking, booking, etc., and the system records the feedback information of the user. Optimize the collaborative filtering module and the LSTM time series analysis module according to the feedback information of the user.

[0093] In this step, record the user's click / book behavior on the recommended item (MySQL log table); update the similarity matrix of collaborative filtering and the LSTM model parameters through the daily offline task.

[0094] In some embodiments, the steps for the collaborative filtering module to generate the initial recommendation plan include:

[0095] S21: Calculate the user interest matching degree through cosine similarity or Pearson correlation coefficient, identify the neighbor group with similar preferences to the target user, and generate a group preference recommendation list based on the behavior data of the neighbor group;

[0096] In this step, obtain the behavior data of all users from the user behavior database, including preference information for hotels, transportation means, etc. For the target user, calculate the interest matching degree between it and other users using cosine similarity or Pearson correlation coefficient. Sort according to the interest matching degree, and select the top N users with the highest interest matching degree with the target user as the neighbor group. Analyze the behavior data of the neighbor group, count their selection frequencies for different hotels, transportation means, etc., and generate a group preference recommendation list.

[0097] The embodiment of the present invention uses the Pearson correlation coefficient to calculate the matching degree, and the formula is as follows:

[0098]

[0099] In the formula, respectively refer to the target user and the user to be compared, is the user 's rating for the item ; the user 's rating for the item ; is the user The average rating for all items For the user The average rating for all items For the user The interest similarity with the user ranges from [-1, 1]. 1 indicates a perfect positive correlation, and -1 indicates a perfect negative correlation.

[0100] Construct a user-item rating matrix R. An example of the matrix is shown in Table 1, where the rows represent users, the columns represent items, and the values are ratings. For each user, calculate the average of the rated items. Only calculate the covariance and standard deviation for the co-rated items and output the similarity.

[0101] Table 1: Example of user-item rating matrix

[0102]

[0103] For each user, calculate the average of all their rated items. Example: The average score of user_123 =(4 + 5) / 2 = 4.5;

[0104] If the sim(i, j) between hotel_789 and flight_abc is 0.72, they are regarded as associated items.

[0105] S22: Analyze the associated behavior sequence of the user's hotel reservation and transportation choice, introduce a time decay factor, and construct a user-item interaction matrix;

[0106] Extract the associated behavior sequence of the user's hotel reservation and transportation choice from the user behavior database. For example, the user first reserves a certain hotel and then selects a certain means of transportation. Introduce a time decay factor to assign higher weights to recent behavior data and lower weights to long-term behavior data. Based on the associated behavior sequence and the time decay factor, construct a user-item interaction matrix, where the elements in the matrix represent the user's preference degree for items.

[0107] S23: Use the modified cosine similarity algorithm to calculate the similarity between items and generate a recommended list of associated items;

[0108] According to the user-item interaction matrix, use the modified cosine similarity algorithm to calculate the similarity between different items (such as different hotels, different means of transportation). For each item, select the top M items with the highest similarity to it to generate a recommended list of associated items.

[0109] The formula for the modified cosine similarity algorithm is as follows:

[0110]

[0111] In the formula, Items for which similarity is to be calculated, At the same time for the items and the items Set of users who rate, For user The rating for the item is, For user The rating for the item is, For user The average rating for all items, For item and item The similarity, with a value range of [-1, 1], where 1 represents a perfect positive correlation and -1 represents a perfect negative correlation.

[0112] S24: Weightedly fuse the recommended items in the group preference recommendation list and the associated item recommendation list to form an initial recommendation scheme. Specifically, it includes:

[0113] S241: For each recommended item in the group preference recommendation list, use its corresponding user interest matching degree as the initial weight. For each recommended item in the associated item recommendation list, use its corresponding modified cosine similarity score as the initial weight;

[0114] S242: Normalize the weights of the two types of recommendations to the same dimension;

[0115] Calculate the total weight of all recommended items in the group preference recommendation list and the associated item recommendation list respectively. For each recommended item, divide its initial weight by the corresponding total weight to obtain the normalized weight.

[0116] S243: Calculate the scores of the group preference recommendation and the associated item recommendation respectively;

[0117] For each recommended item in the group preference recommendation list, calculate the score according to its normalized weight and other relevant factors (such as the user's historical rating for this type of item).

[0118] For each recommended item in the associated item recommendation list, calculate the score according to its normalized weight and other relevant factors (such as the price and evaluation of the item).

[0119] S244: Sort by the weighted total score and retain the top N items as the initial recommendation scheme. Specifically, merge the recommended items in the group preference recommendation list and the associated item recommendation list, and calculate the weighted total score according to their scores. Sort from high to low according to the weighted total score and retain the top N items as the initial recommendation scheme.

[0120] In some embodiments, the steps for the LSTM time series analysis module to generate a personalized travel plan for the user include:

[0121] S31: Retrieve the user's last M historical travel records, including structured data and unstructured note information; In this step, query the user's last M historical travel records from the historical travel record database according to the user identifier, including structured data such as travel date, destination, hotel selection, transportation selection, etc., and unstructured data such as the note information left by the user during the travel.

[0122] S32: Parse the unstructured note information through natural language processing technology to generate user portrait tags;

[0123] Use natural language processing technology (such as word segmentation, part-of-speech tagging, named entity recognition, etc.) to process the unstructured note information. Extract the key information in the note information, such as the user's preferences, special needs, etc., to generate user portrait tags.

[0124] S33: Jointly input the user portrait tags and the historical behavior sequence into the trained LSTM model, and capture the time series features through the time decay weighted sum and attention mechanism;

[0125] In this step, integrate the user portrait tags and the historical behavior sequence to form input data. Introduce time decay weighting into the input data, assign higher weights to recent behavior data, and assign lower weights to long-term behavior data. Use the attention mechanism to make the LSTM model pay more attention to important time series features.

[0126] S34: Generate the probability distributions of destinations, hotels, and flights at the output layer of the LSTM model;

[0127] After the LSTM model processes the input data, generate the probability distributions of destinations, hotels, and flights at the output layer.

[0128] S35: Generate a personalized travel plan through Monte Carlo simulation based on the probability distribution. Specifically, it includes:

[0129] S351: Determine the candidate solution set according to the probability distribution output by the LSTM;

[0130] According to the probability distributions of destinations, hotels, and flights output by the LSTM model, select several combinations of destinations, hotels, and flights with higher probabilities to form the candidate solution set.

[0131] S352: Superimpose external influence factors on each candidate solution;

[0132] Consider the impact of external factors on candidate solutions, such as weather, holidays, etc., and superimpose the corresponding external impact factors on each candidate solution.

[0133] S353: Conduct N random samplings. Each time a sampling is performed, randomly select a candidate solution according to the weighted probability distribution, and calculate the comprehensive score of the probability weight + difference standard compliance + real-time inventory availability of the candidate solution;

[0134] Conduct N random samplings. Each time a sampling is performed, randomly select a candidate solution from the set of candidate solutions according to the weighted probability distribution. Calculate the probability weight, difference standard compliance, and real-time inventory availability of the candidate solution, and add the three to obtain the comprehensive score.

[0135] S354: Statistically analyze all sampling results, and output the top K solutions with the highest comprehensive scores among the selected candidate solutions, that is, the personalized travel plan.

[0136] It should be noted that the training of the LSTM model is as follows:

[0137] Collect time-series data related to business travel, such as user browsing / book reservation sequences, hotel / flight price fluctuations, user portraits (preferences, consumption capabilities), timestamps (seasons, holidays), etc.

[0138] Numeric features (prices, ratings) need to be standardized (Z-score or Min-Max) to eliminate the influence of dimensions. Categorical features (user types, destinations) need to be encoded (such as Embedding or One-Hot), and long sequences need to be segmented into input segments of a fixed length (such as 30-day behavior sequences). Divide the training set (70%), validation set (15%), and test set (15%) proportionally to ensure consistent distribution to locate model problems (overfitting or distribution shift).

[0139] The network structure is designed as follows:

[0140] Input layer: LSTM(units = 128, return_sequences = True) (process sequence output).

[0141] Hidden layer: Stack LSTM layers or combine with Dense layers to extract high-order features, and the number of parameters needs to be controlled to avoid overfitting (such as reducing the number of layers when the amount of data is small).

[0142] Output layer: Use linear activation for regression tasks (such as price prediction), and use Softmax for classification tasks (such as user behavior prediction).

[0143] The LSTM controls the information flow through the input gate, forget gate, and output gate, generates gating coefficients using the Sigmoid function, processes candidate memory cells using the Tanh function, optimizes long-term dependence learning, and uses Xavier or He initialization to avoid gradient explosion / vanishing.

[0144]

[0145] In the formula, is the weight matrix, is the bias term, is the Sigmoid function, with the output range [0, 1], where 0 means complete forgetting and 1 means complete retention.

[0146] For regression tasks, the mean squared error (MSE) or mean absolute error (MAE) is used. For classification tasks, the cross-entropy loss is used. The Adam optimizer (adaptive learning rate) is used, with the initial learning rate set to 0.001, combined with a learning rate decay strategy (such as cosine annealing). The regularization technique uses Dropout (e.g., adding Dropout(0.3) after the LSTM layer) to prevent overfitting. L2 regularization constrains the weights, and gradient clipping (limiting the gradient norm to 5 - 15) avoids gradient explosion.

[0147] Observe the Loss curve of the training / validation set: If the validation Loss increases while the training Loss decreases, it indicates overfitting, and regularization needs to be enhanced or the data needs to be augmented. Use early stopping. When the validation Loss does not decrease for multiple consecutive rounds, terminate the training.

[0148] In some embodiments, the steps of inputting the initial recommendation plan and the personalized travel plan into the dynamic filtering module, performing secondary screening in combination with the real-time inventory of the travel supplier, the budget constraint based on the travel standard, and the user travel profile tags, and outputting the final travel recommendation plan include:

[0149] S41: Take the union of the initial recommendation plan generated by the collaborative filtering module and the personalized travel plan generated by the LSTM time series analysis module;

[0150] Merge all the recommended items in the initial recommendation plan and the personalized travel plan, remove the duplicates, and obtain the union result.

[0151] S42: Obtain the real-time inventory through the supplier API interface and retrieve the budget constraint from the corporate travel standard database;

[0152] Establish an API interface connection with the travel supplier and obtain the real-time inventory information through interface calls. Query the budget constraint information based on the travel standard from the corporate travel standard database.

[0153] S43: Conduct compliance filtering based on real-time inventory, budget constraints, and user profile tags, and exclude the items that exceed the standards; in the business travel dynamic planning system, the real-time inventory refers to the real-time available reservation resource data obtained by docking with the systems of business travel service providers (such as airlines, hotels, car rental companies), including flights: remaining seat numbers, cabin classes (economy class / business class), price fluctuations. Hotels: available room types, negotiated prices, occupancy rates. Other services: pick-up and drop-off vehicles, meeting venues, etc.

[0154] For each recommended item in the union result, check whether it meets the requirements of real-time inventory, budget constraints, and user profile tags. Exclude the recommended items that do not meet the requirements to obtain the options that pass the screening.

[0155] S44: Generate the final business travel recommendation plan for the options that pass the screening according to the preset weights; among them, the weight A of the personalized plan is greater than the weight B of the initial recommendation.

[0156] Assign weights A and weight B (A > B) to the personalized plan and the initial recommendation plan among the options that pass the screening respectively. Calculate the scores of each option according to the weights, and sort them from high to low to generate the final business travel recommendation plan.

[0157] In some embodiments, the steps of optimizing the collaborative filtering module and the LSTM time series analysis module based on the feedback information include:

[0158] S51: Adjust the weights of the recommended items in the group preference recommendation list and the weights of the recommended items in the associated item recommendation list in the collaborative filtering module according to the user's click and reservation behaviors;

[0159] Record the user's click and reservation behaviors on the recommendation plan, and analyze the user's preference changes. According to the user's preference changes, adjust the weights of the recommended items in the group preference recommendation list and the weights of the recommended items in the associated item recommendation list. For example, if the user clicks or reserves a certain recommended item, increase the weight of this recommended item. User clicks on Flight A → Update the user's rating vector in User-CF; Re-calculate the item similarity matrix by SVD decomposition (weekly offline task).

[0160] S52: Update the user profile tags according to the user's click and reservation behaviors, and then update the attention mechanism parameters of the LSTM model in the LSTM time series analysis module.

[0161] According to the user's click and reservation behaviors, update the user profile tags to reflect the user's latest preferences and needs. Use the updated user profile tags to re-train the attention mechanism parameters of the LSTM model to make the model pay more attention to the user's current needs.

[0162] For example, when a user books Hotel B, this behavior is added to the training set, triggering incremental training. The attention mechanism parameters are adjusted by increasing the time decay weight of recent behaviors.

[0163] The collaborative operation of the time series analysis module and the collaborative filtering module constructs a dual guarantee system for intelligent business travel recommendations. The intervention of the dynamic filtering module ensures that the recommendation results precisely meet the enterprise management requirements. The time series analysis module generates personalized travel plans based on the user's historical travel data (such as destination preferences, time period selection rules, etc.) using time series prediction algorithms. The collaborative filtering module, through cross-user behavior similarity analysis, discovers high-quality choices in the collective wisdom. For example, it identifies colleague groups with travel patterns highly consistent with the current user and extracts the frequently chosen transportation means and hotels.

[0164] After the dual modules output the recommendation results, the dynamic filtering module immediately initiates the travel standard rule verification process. The system automatically retrieves the user's business travel standard database, which includes hard constraint conditions such as budget limits, cabin classes, hotel star ratings, as well as differential rules based on attributes such as position and department. For example, it provides flexible space for senior management users to "upgrade the flight ticket to first class", while strictly limiting the options for interns to "only economy class".

[0165] During the filtering process, the system adopts a hierarchical decision-making mechanism: first, it excludes obviously over-standard items, then conducts a cost-benefit analysis on the critical value plans, and finally generates a compliance score in combination with the user's historical behavior data (such as past business travel standard implementation). For recommendations that meet the travel standards, the system directly pushes them to the user side; if there is a risk of over-standard, it automatically triggers travel standard optimization suggestions, such as recommending flights with price lows during adjacent time periods or providing preferential packages for enterprise-agreement hotels, maximizing the user experience while ensuring compliance. This dynamic balance mechanism of "intelligent recommendation + rule constraint" realizes the refinement and humanization of business travel management.

[0166] For example, (1) the initial recommendation plan generated by the collaborative filtering module

[0167] Data sources: Based on group behavior (user-user collaborative filtering) and item association rules (item-item collaborative filtering).

[0168] Reflect common preferences (such as "economy class flights often chosen by users in the same industry");

[0169] Solve the cold start problem, but lack in-depth exploration of the user's individual time series behavior.

[0170] Example: Recommend the "early morning flight from Beijing to Shanghai" (because 80% of similar users choose this flight).

[0171] (2) The personalized travel plan generated by the LSTM time series analysis module

[0172] Data source: Based on the user's own historical behavior sequence (such as the last 12 trip records).

[0173] Capture individual dynamics (such as "this user has recently changed flights frequently and prefers flights with high flexibility");

[0174] Combine external events (such as weather, exhibitions) to predict future demand.

[0175] Example: Recommend a "Beijing - Shanghai midday flight" (because the user has chosen midday trips in the past 3 times).

[0176] Since the initial recommendation may ignore recent changes in the user's behavior (such as the tendency to switch from "economy class" to "business class"); it cannot respond to dynamic factors in real time (such as a temporary price increase for a flight). The personalized solution relies on the user's historical data, and the prediction for new users or low - frequency users is inaccurate; it may deviate from the corporate travel policy (such as recommending a five - star hotel beyond the travel allowance).

[0177] The core function of the dynamic filtering module is to exclude options that exceed the budget or are out of stock in the initial recommendation or personalized solution (such as first - class seats being sold out). Sort the cost - performance of critical value solutions (such as prices close to the upper limit of the travel allowance). Filter high - risk recommendations based on user portrait tags (such as "frequently cancels hotels").

[0178] Example:

[0179] Initial recommendation: Flight A (economy class, group preference, price compliant);

[0180] Personalized solution: Flight B (business class, individual preference, price exceeds the travel allowance by 10%);

[0181] Dynamic filtering result: Keep Flight A, but prompt "You can apply for travel allowance flexibility" (because the user has a recent tendency to upgrade).

[0182] As Figure 2 shown, the embodiment of the present invention also provides a business travel recommendation system based on collaborative filtering and recurrent neural network, including a business travel application form receiving module, a collaborative filtering module, an LSTM time - series analysis module, a dynamic filtering module, and a recommendation optimization module;

[0183] The business travel application form receiving module is used to obtain the approved business travel application form and extract business travel information, including destination, date, job level, and travel allowance;

[0184] The collaborative filtering module is used to generate an initial recommendation solution based on the business travel information through user - user collaborative filtering and item - item collaborative filtering algorithms;

[0185] The LSTM time series analysis module is used to retrieve the user's recent M historical travel records according to the business trip information, construct a user travel profile, train a time series model of user behavior through a time series neural network architecture, and generate a personalized travel plan through Monte Carlo simulation for the trained time series model;

[0186] The dynamic filtering module is used to perform secondary screening based on the initial recommendation plan and the personalized travel plan, combined with the real-time inventory of travel suppliers, the budget constraint based on travel standards, and the user travel profile tags, and output the final travel recommendation plan;

[0187] The recommendation optimization module is used to push the final travel recommendation plan to the user side, receive the feedback information of the user, and optimize the collaborative filtering module and the LSTM time series analysis module based on the feedback information.

[0188] In some embodiments, the steps for the collaborative filtering module to generate the initial recommendation plan include: calculating the user interest matching degree through cosine similarity or Pearson correlation coefficient, identifying the neighbor group with similar preferences to the target user, and generating a group preference recommendation list based on the behavior data of the neighbor group; analyzing the associated behavior sequence of the user's hotel reservation and transportation choice, introducing a time decay factor, and constructing a user-item interaction matrix; calculating the similarity between items using the modified cosine similarity algorithm to generate an associated item recommendation list; and performing weighted fusion on the recommendation items of the group preference recommendation list and the associated item recommendation list to form the initial recommendation plan.

[0189] In some embodiments, the specific steps for the collaborative filtering module to form the initial recommendation plan include: taking the user interest matching degree as the initial weight of the recommendation items in the group preference recommendation list, and taking the modified cosine similarity score as the initial weight of the recommendation items in the associated item recommendation list; normalizing the weights of the two types of recommendations to the same dimension; calculating the scores of the group preference recommendation and the associated item recommendation respectively; and sorting the top N items according to the weighted total score as the initial recommendation plan.

[0190] In some embodiments, the steps for the LSTM time series analysis module to generate the user's personalized travel plan include: retrieving the user's recent M historical travel records, including structured data and unstructured note information; parsing the unstructured note information through natural language processing technology to generate user profile tags; jointly inputting the user profile tags and the historical behavior sequence into the trained LSTM model, and capturing time series features through time decay weighted sum and attention mechanism; generating the probability distributions of destinations, hotels, and flights at the output layer of the LSTM model; and generating a personalized travel plan through Monte Carlo simulation based on the probability distributions.

[0191] In some embodiments, the steps of the LSTM time series analysis module generating a personalized travel plan based on a probability distribution through Monte Carlo simulation include: determining a set of candidate plans according to the probability distribution output by the LSTM; superimposing external influence factors on each candidate plan; performing N random samplings, and randomly selecting a candidate plan according to a weighted probability distribution each time of sampling, and calculating the comprehensive score of the probability weight + difference standard compliance + real-time inventory availability of the candidate plan; counting all sampling results, and outputting the top K plans with the highest comprehensive scores among the selected candidate plans, that is, the personalized travel plan.

[0192] In some embodiments, the dynamic filtering module is specifically configured to take the union of the initial recommendation plan generated by the collaborative filtering module and the personalized travel plan generated by the LSTM time series analysis module; obtain real-time inventory through the supplier API interface, and retrieve the budget constraint from the corporate travel standard database; perform compliance filtering based on the real-time inventory, budget constraint and user portrait tags, and exclude the items that exceed the standard;

[0193] Generating a final travel recommendation plan for the options passing the screening according to preset weights; wherein the weight A of the personalized plan is greater than the weight B of the initial recommendation.

[0194] In some embodiments, the recommendation optimization module is specifically configured to adjust the weights of the recommended items in the group preference recommendation list and the weights of the recommended items in the associated item recommendation list in the collaborative filtering module according to the user's click and booking behaviors; update the user portrait tags according to the user's click and booking behaviors, and further update the attention mechanism parameters of the LSTM model in the LSTM time series analysis module.

[0195] An embodiment of the present invention further provides an electronic device, which includes: a processor, a communication interface, a memory, and a communication bus. Among them, the processor, the communication interface, and the memory complete communication with each other through the communication bus. The communication bus can be used for information transmission between the electronic device and the sensor. The processor can call the logical instructions in the memory to execute the following method: S1: Obtain the approved travel application form and extract travel information, including the destination, date, rank, and travel standard; S2: Input the travel information into the collaborative filtering module, and generate an initial recommendation plan through user-user collaborative filtering and item-item collaborative filtering algorithms; S3: Input the travel information into the LSTM time series analysis module, retrieve the user's recent M historical travel records, construct a user travel profile, train a time series model of user behavior through a time series neural network architecture, and the trained time series model generates a personalized travel plan through Monte Carlo simulation; S4: Input the initial recommendation plan and the personalized travel plan into the dynamic filtering module, and perform secondary screening in combination with the real-time inventory of travel suppliers, budget constraints based on travel standards, and user travel profile tags, and output the final travel recommendation plan; S5: Push the final travel recommendation plan to the user terminal, receive the user's feedback information, and optimize the collaborative filtering module and the LSTM time series analysis module based on the feedback information.

[0196] In addition, when the logical instructions in the above-mentioned memory can be implemented in the form of software functional units and sold or used as an independent product, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0197] An embodiment of the present invention provides a non-transitory computer-readable storage medium that stores computer instructions, and the computer instructions cause a computer to execute the method provided in the above method embodiment. For example, it includes: S1: Obtain a travel application form that has been approved, and extract travel information, including the destination, date, job level, and travel standard; S2: Input the travel information into a collaborative filtering module, and generate an initial recommendation plan through user-user collaborative filtering and item-item collaborative filtering algorithms; S3: Input the travel information into an LSTM time series analysis module, retrieve the user's recent M historical travel records, construct a user travel profile, train a time series model of user behavior through a time series neural network architecture, and the trained time series model generates a personalized travel plan through Monte Carlo simulation; S4: Input the initial recommendation plan and the personalized travel plan into a dynamic filtering module, and perform secondary screening in combination with the real-time inventory of travel suppliers, budget constraints based on travel standards, and user travel profile tags, and output a final travel recommendation plan; S5: Push the final travel recommendation plan to the user terminal, receive the feedback information of the user, and optimize the collaborative filtering module and the LSTM time series analysis module based on the feedback information.

[0198] The foregoing description of the disclosed embodiments enables those skilled in the art to practice or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Thus, the invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A business trip recommendation method based on collaborative filtering and recurrent neural network, characterized in that, It includes the following steps: Obtain the approved business trip application form and extract business trip information, including destination, date, rank, and business trip standard; Input the business trip information into the collaborative filtering module, and generate an initial recommendation plan through user-user collaborative filtering and item-item collaborative filtering algorithms; Input the business trip information into the LSTM time series analysis module, retrieve the user's recent M historical travel records, construct a user travel profile, train a time series model of user behavior through a time series neural network architecture, and generate a personalized travel plan through Monte Carlo simulation for the trained time series model; Input the initial recommendation plan and the personalized travel plan into the dynamic filtering module, and perform secondary screening by combining the real-time inventory of business trip suppliers, budget constraints based on business trip standards, and user travel profile tags, and output the final business trip recommendation plan; Push the final business trip recommendation plan to the user terminal, receive the user's feedback information, and optimize the collaborative filtering module and the LSTM time series analysis module based on the feedback information.

2. The business trip recommendation method based on collaborative filtering and recurrent neural network according to claim 1, wherein The steps for the collaborative filtering module to generate an initial recommendation plan include: Calculate the user interest matching degree through cosine similarity or Pearson correlation coefficient, identify the neighbor group with similar preferences to the target user, and generate a group preference recommendation list based on the behavior data of the neighbor group; Analyze the associated behavior sequence of the user's hotel reservation and transportation choice, introduce a time decay factor, and construct a user-item interaction matrix; According to the user-item interaction matrix, use the modified cosine similarity algorithm to calculate the similarity between items, and generate an associated item recommendation list; Perform weighted fusion on the recommended items of the group preference recommendation list and the associated item recommendation list to form an initial recommendation plan.

3. The travel recommendation method based on collaborative filtering and recurrent neural network according to claim 2, wherein The steps for performing weighted fusion on the recommended items of the group preference recommendation list and the associated item recommendation list to form an initial recommendation plan include: Take the user interest matching degree as the initial weight of the recommended items in the group preference recommendation list, and take the modified cosine similarity score as the initial weight of the recommended items in the associated item recommendation list; Normalize the weights of the two types of recommendations to the same dimension; Calculate the scores of the group preference recommendation and the associated item recommendation respectively; Retain the top N items according to the weighted total score as the initial recommendation plan.

4. The travel recommendation method based on collaborative filtering and recurrent neural network according to claim 3, wherein The steps for the LSTM time series analysis module to generate a user personalized travel plan include: Retrieve the user's recent M historical travel records to generate a historical behavior sequence, and the historical travel records include structured data and unstructured note information; Parse the unstructured note information through natural language processing technology to generate user profile tags; Jointly input the user profile tags and the historical behavior sequence into the trained LSTM model, and capture time series features through time decay weighting and attention mechanism; Generate the probability distributions of the destination, hotel, and flight at the output layer of the LSTM model; Generate a personalized travel plan through Monte Carlo simulation based on the probability distribution.

5. The travel recommendation method based on collaborative filtering and recurrent neural network according to claim 4, wherein The steps for generating a personalized travel plan through Monte Carlo simulation based on the probability distribution include: Determine the candidate plan set according to the probability distribution output by the LSTM; Overlay external influence factors on each candidate plan; Perform N random samplings. Each time a sampling is conducted, randomly select a candidate solution according to the weighted probability distribution, and calculate the comprehensive score of the probability weight + difference standard compliance + real-time inventory availability of the candidate solution; Statistically analyze all sampling results, and output the top K solutions with the highest comprehensive scores among the selected candidate solutions, that is, the personalized travel solutions.

6. The travel recommendation method based on collaborative filtering and recurrent neural network according to claim 5, characterized in that, The steps of inputting the initial recommended solution and the personalized travel solution into the dynamic filtering module, and performing secondary screening in combination with the real-time inventory of travel suppliers, budget constraints based on travel standards, and user travel portrait tags to output the final travel recommendation solution include: Take the union of the initial recommended solution generated by the collaborative filtering module and the personalized travel solution generated by the LSTM time series analysis module; Obtain the real-time inventory through the supplier API interface, and retrieve the budget constraints from the enterprise travel standard database; Perform compliance filtering based on the real-time inventory, budget constraints, and user portrait tags, and exclude the items that exceed the standard; Generate the final travel recommendation solution for the options that pass the screening according to the preset weights; where the weight A of the personalized solution is greater than the initial recommended weight B.

7. The travel recommendation method based on collaborative filtering and recurrent neural network according to claim 6, characterized in that The steps of optimizing the collaborative filtering module and the LSTM time series analysis module based on the feedback information include: Adjust the weights of the recommended items in the group preference recommendation list and the weights of the recommended items in the associated item recommendation list in the collaborative filtering module according to the user's click and booking behaviors; Update the user portrait tags according to the user's click and booking behaviors, and then update the attention mechanism parameters of the LSTM model in the LSTM time series analysis module.

8. A business travel recommendation system based on collaborative filtering and recurrent neural network, characterized in that, It includes a travel application form receiving module, a collaborative filtering module, an LSTM time series analysis module, a dynamic filtering module, and a recommendation optimization module; The travel application form receiving module is used to obtain the approved travel application form and extract travel information, including the destination, date, job level, and travel standards; The collaborative filtering module is used to generate an initial recommended solution based on the travel information through user-user collaborative filtering and item-item collaborative filtering algorithms; The LSTM time series analysis module is used to retrieve the user's recent M historical travel records according to the travel information, construct a user travel portrait, train a time series model of user behavior through a time series neural network architecture, and the trained time series model generates a personalized travel solution through Monte Carlo simulation; The dynamic filtering module is used to perform secondary screening based on the initial recommended solution and the personalized travel solution, in combination with the real-time inventory of travel suppliers, budget constraints based on travel standards, and user travel portrait tags, and output the final travel recommendation solution; The recommendation optimization module is used to push the final travel recommendation solution to the user side, receive the user's feedback information, and optimize the collaborative filtering module and the LSTM time series analysis module based on the feedback information.

9. The travel recommendation system based on collaborative filtering and recurrent neural network according to claim 8, wherein The collaborative filtering module calculates the user interest matching degree through cosine similarity or Pearson correlation coefficient, identifies the neighbor group with similar preferences to the target user, and generates a group preference recommendation list based on the behavior data of the neighbor group; Analyze the associated behavior sequence of the user's hotel reservation and transportation tool selection, introduce a time decay factor, and construct a user-item interaction matrix; use the modified cosine similarity algorithm to calculate the similarity between items, and generate an associated item recommendation list; Perform weighted fusion on the recommended items in the group preference recommendation list and the associated item recommendation list to form an initial recommendation plan.

10. The travel recommendation system based on collaborative filtering and recurrent neural network according to claim 9, wherein, The LSTM time series analysis module retrieves the user's recent M historical travel records, including structured data and unstructured note information; parses the unstructured note information through natural language processing technology to generate user portrait tags; jointly inputs the user portrait tags and the historical behavior sequence into the trained LSTM model, and captures time series features through time decay weighted sum and attention mechanism; Generate the probability distributions of destinations, hotels, and flights at the output layer of the LSTM model; Generate a personalized travel plan based on the probability distribution through Monte Carlo simulation.

Images