A hotel red envelope distribution method, system, electronic device and storage medium

Abstract

A hotel red envelope distribution method, system, electronic device and storage medium, relating to the field of e-commerce. In the method, the trajectory characteristics of the target user are obtained, the trajectory characteristics include the hotel type, price range and geographical location browsed by the target user, and the departure place, destination and time of the air ticket, train ticket or bus ticket searched by the target user; the order probability is obtained according to the trajectory characteristics using a preset prediction model, and the order probability includes the natural conversion rate and the marketing conversion rate; multiple red envelope thresholds are generated according to the historical data of red envelope distribution; the red envelope sending strategy of the target user is determined according to the order probability and the multiple red envelope thresholds, and the amount of the red envelope is determined according to the red envelope sending strategy. The implementation of the technical solution provided by the present application improves the accuracy of red envelope delivery, thereby improving the return on investment.

Description

Technical Field

[0001] The present application relates to the technical field of e-commerce, and in particular to a method, system, electronic device and storage medium for distributing hotel red envelopes. Background Art

[0002] In recent years, with the rapid development of Internet technology, major e-commerce platforms have launched various preferential measures to attract users, improve user stickiness and order conversion rate. Red envelopes, as a common promotional method, have been widely used in online shopping, travel booking and other fields. Through accurate red envelope distribution strategies, the user experience can be effectively improved, and the user's trust and satisfaction with the platform can be increased, thereby promoting the growth of transaction volume.

[0003] Existing red envelope distribution methods have certain limitations. Although it is possible to predict the user's order probability through data analysis and machine learning models, in actual applications, these methods often find it difficult to accurately capture the user's personalized needs, resulting in inaccurate red envelope distribution strategies.

[0004] Therefore, a method is needed to customize the distribution of hotel red envelopes based on user trajectories. Summary of the invention

[0005] The present application provides a hotel red envelope distribution method, system, electronic device and storage medium. By acquiring the trajectory characteristics of the target user, it is possible to more accurately understand the user's travel needs and preferences, which helps to formulate personalized red envelope strategies for different users and improve the pertinence and effectiveness of marketing.

[0006] In a first aspect of the present application, a method for distributing hotel red envelopes is provided, which is applied to a travel platform, and the method comprises: Acquire trajectory features of the target user, the trajectory features including the type, price range, and geographic location of hotels browsed by the target user, and the departure place, destination, and time of the air ticket, train ticket, or bus ticket searched by the target user; Using a preset prediction model to obtain an order probability according to the trajectory characteristics, the order probability includes a natural conversion rate and a marketing conversion rate; Generate multiple red envelope thresholds based on historical data of red envelope issuance; A red envelope sending strategy for the target user is determined based on the order probability and the multiple red envelope thresholds, and the amount of the red envelope is determined based on the red envelope sending strategy.

[0007] Optionally, the method comprises: Obtaining the original trajectory data of all users before the current moment, and converting the original trajectory data into original features, normalizing the numerical features, and label encoding the discrete features; Use the target feature to split, calculate the difference between the information entropy of the node before the split and the sum of the information entropy of the two child nodes after the split to obtain the information gain, wherein the target feature is any original feature; Accumulating the information gains of all splitting points of the target feature in all decision trees to obtain the importance of the target feature, and screening out a feature set that meets the preset requirements according to the importance; The initial prediction model is trained according to the feature set to obtain the preset prediction model.

[0008] Optionally, generating multiple red envelope thresholds according to historical data of red envelope issuance includes: Determine a first user group to which red envelopes have been issued, obtain user data of the first user group, the user data including a first natural conversion rate and a first marketing conversion rate of each user, and calculate a first marketing gain of each user according to the first natural conversion rate and the first marketing conversion rate; Dividing the first user group into a plurality of standard red envelope groups according to the first natural conversion rate, the first marketing conversion rate, and the first marketing gain, wherein the plurality of standard red envelope groups include a non-red envelope group, a small number of red envelope groups, a large number of red envelope groups, and a normal red envelope group; The range of natural conversion rate, marketing conversion rate and marketing gain within the standard red envelope group is counted, and the red envelope threshold is determined based on the range.

[0009] Optionally, dividing the first user group into a plurality of standard red envelope groups according to the first natural conversion rate, the first marketing conversion rate, and the first marketing gain includes: Sort all users in the first user group from large to small according to their natural conversion rates, and obtain a first number of users ranked at the top to form a first user group; Sort the users in the first user group in ascending order of marketing gain, obtain a second number of users at the front to form a second user group, and determine the second user group as a group that does not send red envelopes; Remove users corresponding to the second user group from the first user group to obtain a second user group, sort all users in the second user group from large to small according to the natural conversion rate, obtain the third number of users ranked at the front to form a third user group, and determine the third user group as a group that sends less red envelopes; Removing users corresponding to the third user group from the second user group to obtain a third user group, sorting all users in the third user group from large to small according to marketing gain, obtaining a fourth number of users at the front to form a fourth user group, and determining the fourth user group as a multiple red envelope sending group; The fourth user group is obtained by removing users corresponding to the fourth user group from the third user group, and the fourth user group is determined as a normal red envelope sending group.

[0010] Optionally, counting the range of the natural conversion rate, the marketing conversion rate, and the marketing gain in the standard red envelope group, and determining the red envelope threshold according to the range includes: Determine the first natural range of the second natural conversion rate and the first gain range of the second marketing gain in the group that does not send red envelopes, determine the second natural range of the third natural conversion rate in the group that sends fewer red envelopes, and determine the second gain range of the third marketing gain in the group that sends more red envelopes; The first minimum value in the first natural range is used as the first threshold, and the first maximum value in the first gain range is used as the second threshold, so that the payment amount is zero when the natural conversion rate is greater than or equal to the first threshold and the marketing gain is less than the second threshold; The second minimum value in the second natural range is used as the third threshold, so that when the natural conversion rate is less than the first threshold or the marketing gain is greater than or equal to the second threshold, and the natural conversion rate is greater than or equal to the third threshold, the payment amount is greater than zero and less than or equal to the first value; The third minimum value in the second gain range is used as the fourth threshold value, so that when the marketing gain is greater than or equal to the fourth threshold value, the payment amount is greater than or equal to a second value, and the second value is greater than the first value.

[0011] Optionally, determining the red envelope sending strategy for the target user according to the order probability and the multiple red envelope thresholds further includes: Analyze the historical behavior data of the target user, wherein the historical behavior data includes browsing history, purchase history, evaluation history, collection history, sharing history, and hotel check-in history; Building a user profile based on the historical behavior data, the user profile includes the user's preference characteristics, spending power, active time periods, travel habits, and price sensitivity; Based on the user portrait, hotel red envelopes of target type and target amount are recommended to the target user.

[0012] Optionally, the recommending a hotel red packet of a target type and a target amount to the target user according to the user portrait includes: Identify target features in the user portrait, the target features including the type of hotel where the target user stays most often, the geographical location where the target user travels most frequently, and the average price range of the user's historical consumption; According to the target feature, a red envelope type matching the target user is selected from a predefined red envelope type library, wherein the red envelope type library includes but is not limited to economy hotel discount coupons, luxury hotel discount coupons, and specific brand hotel exclusive coupons; Combining the natural conversion rate and marketing conversion rate of the target user and the historical conversion rate data of each red envelope type in the red envelope type library, the target amount is obtained through a prediction algorithm; A hotel red envelope is distributed to the target user according to the red envelope type and the target amount.

[0013] In a second aspect of the present application, a hotel red envelope distribution system is provided, including a collection module, a prediction module, a threshold module and a sending module, wherein: a collection module configured to obtain trajectory features of a target user, the trajectory features including the type, price range, and geographic location of hotels browsed by the target user, and the departure place, destination, and time of air tickets, train tickets, or bus tickets searched by the target user; A prediction module, configured to obtain an order probability according to the trajectory characteristics using a preset prediction model, wherein the order probability includes a natural conversion rate and a marketing conversion rate; A threshold module, configured to generate multiple red envelope thresholds based on historical data of red envelope distribution; The sending module is configured to determine the red envelope sending strategy of the target user according to the order probability and the multiple red envelope thresholds, and determine the amount of the red envelope according to the red envelope sending strategy.

[0014] In the third aspect of the present application, an electronic device is provided, including a processor, a memory, a user interface and a network interface, the memory is used to store instructions, the user interface and the network interface are both used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes any one of the methods described above.

[0015] In a fourth aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores instructions, and when the instructions are executed, any of the methods described above is executed.

[0016] In summary, one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. By obtaining the target user's trajectory characteristics, such as the type of hotel they browse, price range, geographic location, and transportation information they search for (such as the departure, destination, and time of air tickets, train tickets, or bus tickets), we can more accurately understand the user's travel needs and preferences, which helps to formulate personalized red envelope strategies for different users and improve the targeting and effectiveness of marketing; 2. Using the preset prediction model, the order probability is calculated based on the user's trajectory characteristics, including the natural conversion rate and marketing conversion rate. This prediction capability enables the platform to more effectively identify users who are more likely to make hotel reservations and adjust the red envelope strategy accordingly, thereby improving the overall conversion rate; 3. By generating multiple red envelope thresholds based on historical data of red envelope issuance, and determining the red envelope sending strategy and amount based on the user's order probability and these thresholds, it is possible to ensure that the red envelope issuance is more reasonable and efficient, which can not only reduce unnecessary marketing costs, but also improve user experience, because users are more likely to complete the booking because they receive a red envelope of the right amount; 4. Personalized red envelope strategies not only help attract new users, but also enhance the stickiness of existing users by providing preferential measures that meet user expectations. When users feel that the platform pays attention to and respects them, they are more likely to become loyal users of the platform, thereby increasing the platform's user base and market share; 5. By collecting and analyzing users' trajectory characteristics, the platform can continuously optimize its prediction model and red envelope strategy to achieve more accurate and efficient marketing. This data-driven decision-making method helps to improve the platform's competitiveness and profitability. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a flowchart of the hotel red envelope distribution method disclosed in the embodiment of the present application; Figure 2 It is a schematic diagram of the architecture of the travel platform disclosed in the embodiment of the present application; Figure 3 It is a schematic diagram of the decision tree growth strategy for leaf growth disclosed in the embodiment of the present application; Figure 4 It is a schematic diagram of the technical process of the preset prediction model disclosed in the embodiment of the present application; Figure 5 is a schematic diagram of the DeepFM model structure disclosed in the embodiments of the present application; Figure 6 is a configuration threshold generation flow chart disclosed in an embodiment of the present application; Figure 7 It is a module diagram of the hotel red envelope distribution system disclosed in the embodiment of the present application; Figure 8 It is a structural schematic diagram of an electronic device disclosed in an embodiment of the present application.

[0018] Explanation of the reference numerals: 701, acquisition module; 702, prediction module; 703, threshold module; 704, sending module; 801, processor; 802, communication bus; 803, user interface; 804, network interface; 805, memory. DETAILED DESCRIPTION

[0019] In order to enable technicians in this field to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments.

[0020] In the description of the embodiments of the present application, words such as "for example" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "for example" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "for example" or "for example" is intended to present related concepts in a specific way.

[0021] In the description of the embodiments of the present application, the meaning of the term "multiple" refers to two or more. For example, multiple systems refer to two or more systems, and multiple screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. The terms "include", "comprise", "have" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.

[0022] This embodiment discloses a method for distributing hotel red envelopes, which is applied to travel platforms. Figure 1 is a flow chart of the hotel red envelope distribution method disclosed in the embodiment of the present application, such as Figure 1 As shown, the method comprises the following steps: S110, obtaining trajectory features of the target user, the trajectory features including the type, price range, and geographic location of the hotels browsed by the target user, and the departure place, destination, and time of the air ticket, train ticket, or bus ticket searched by the target user; In the operation of travel platforms, understanding and analyzing the trajectory characteristics of target users is a key step in formulating precise marketing strategies and improving user experience. These trajectory characteristics not only reflect the travel needs and preferences of users, but also provide valuable market insight data for the platform. The platform can understand the user's accommodation preferences by recording the types of hotels browsed by users (such as economy, luxury, business, etc.). This is crucial for recommending hotel red envelopes or discounts that meet their needs to users in the future. When users browse hotels, they often show a strong interest in hotels in a specific price range. By capturing this information, the platform can push hotel red envelopes that are more in line with their budget to users, thereby increasing the user's booking conversion rate. The geographical location of the user's browsing hotel is also one of the important trajectory characteristics. By analyzing the hotel browsing situation around the cities, regions or attractions that users often visit, the platform can recommend hotel red envelopes that are closer to their travel plans to users. In addition to hotel information, the platform also records relevant information about users' searches for air tickets, train tickets or bus tickets, including departure place, destination and travel time. This information helps the platform determine the user's travel plan and itinerary, thereby providing users with more intimate hotel recommendations and red envelope strategies. For example, if a user is about to go to a certain city to attend a meeting, the platform can recommend hotel red envelopes near the meeting place; if the user plans to stay near a certain tourist attraction, the platform can recommend hotel red envelopes with high cost-effectiveness in the surrounding area.

[0023] Figure 2 is a schematic diagram of the architecture of the travel platform disclosed in the embodiment of this application, such as Figure 2 As shown in the figure, the travel platform includes data and algorithms. The data part is processed in advance every day, including hotel, train, air ticket, and bus ticket trajectory features, normalization of numerical features, label encoding of discrete features and transfer to redis tables; the model part obtains trajectory features in real time and estimates the probability of placing an order before and after marketing. Finally, the red envelopes issued are adjusted according to the estimated order probability and the configuration file estimated based on historical data: more, less, normal, and no denomination.

[0024] Optionally, the method comprises: Obtaining the original trajectory data of all users before the current moment, and converting the original trajectory data into original features, normalizing the numerical features, and label encoding the discrete features; Use the target feature to split, calculate the difference between the information entropy of the node before the split and the sum of the information entropy of the two child nodes after the split to obtain the information gain, wherein the target feature is any original feature; Accumulating the information gains of all splitting points of the target feature in all decision trees to obtain the importance of the target feature, and screening out a feature set that meets the preset requirements according to the importance; The initial prediction model is trained according to the feature set to obtain the preset prediction model.

[0025] The original trajectory data of all users before the current moment is obtained from the system. These data may include the user's browsing history, click behavior, purchase history, etc., covering various behaviors of users on the platform. The obtained original trajectory data is converted into original features. This step may involve data cleaning, formatting, and feature extraction processes to ensure that the data is suitable for subsequent machine learning processing. The numerical features are normalized to eliminate the differences in dimensions between different features and improve the training efficiency and prediction accuracy of the model. At the same time, the discrete features are label encoded to convert non-numeric data into numerical data so that the model can process them. In the process of building a decision tree, any original feature is selected as the target feature for splitting. The purpose of splitting is to divide the data into two parts according to the value of the target feature, so that the distribution of the two parts of the data on the target variable is purer (that is, similar samples are more aggregated). The embodiment of the present application adopts a more efficient leaf-wise growth strategy with depth restriction. Leaf-wise can be split according to the leaf with the largest split gain among all current leaves, and this cycle is repeated, so that when the number of splits is the same, Leaf-wise can reduce the error and obtain higher accuracy. In order to avoid overfitting caused by the model generating a deeper decision tree, the embodiment of the present application adds a maximum depth limit on this basis, which not only ensures low error and high efficiency, but also prevents overfitting. Figure 3 This is a diagram of the leaf-based decision tree growth strategy. The difference between the information entropy of the node before the split and the sum of the information entropy of the two child nodes after the split is calculated. This difference is the information gain. Information gain reflects the degree of improvement in the purity of data distribution before and after the split, and is an important indicator for evaluating the split effect. Accumulate information gain to get feature importance: For each target feature, accumulate the information gain of all split points in all decision trees to get the importance of the feature. The importance of a feature reflects the contribution of the feature to the model's prediction ability. According to the importance of the feature, filter out feature sets that meet the preset requirements. The preset requirements may include feature importance thresholds, feature quantity limits, etc. Through screening, redundant and irrelevant features can be removed, reducing the complexity and computational cost of the model. Use feature sets to train the initial prediction model: Use the filtered feature set to train the initial prediction model. During the training process, the model will continuously adjust parameters to optimize the prediction performance. After a certain number of iterative training, the model will converge to a stable state, and the model obtained at this time is the preset prediction model. The model can predict the value of the target variable based on the input feature data, such as the user's purchase intention, behavioral preference, etc.

[0026] By converting the original trajectory data into original features, normalizing the numerical features, and label encoding the discrete features, the consistency and comparability of the data are ensured, providing a reliable basis for the subsequent feature importance calculation and model training. Using information gain as an indicator to evaluate the effect of feature splitting can accurately reflect the contribution of features to the model's prediction ability. By accumulating the information gain of all split points of the target feature in all decision trees, the importance of the feature can be obtained, providing a scientific basis for subsequent feature screening. According to the importance of the feature, the feature set that meets the preset requirements can be selected, which can remove redundant and irrelevant features and reduce the complexity and computational cost of the model. At the same time, the screened feature set focuses more on the features that have an important impact on the prediction results, improving the prediction accuracy and generalization ability of the model. Using the screened feature set to train the initial prediction model can reduce noise interference in the model training process and improve the training efficiency and convergence speed of the model. At the same time, since the feature set is more refined and effective, the preset prediction model obtained by training will also be improved in prediction performance.

[0027] S120, using a preset prediction model to obtain an order probability according to the trajectory characteristics, where the order probability includes a natural conversion rate and a marketing conversion rate; The "trajectory features" here refer to a series of data points that can reflect user behavior patterns, preferences, and potential needs. They usually come from the user's historical behavior records, such as browsing history, click behavior, purchase history, etc. The order probability is a comprehensive indicator that measures the probability that a user will complete a purchase at a certain point in the future. This probability value consists of two parts: the natural conversion rate and the marketing conversion rate. The natural conversion rate refers to the probability that a user will spontaneously complete a purchase based on their own needs and preferences without additional marketing activities or incentives. This probability reflects the user's real demand for the product and willingness to buy. When predicting the natural conversion rate, the preset prediction model will focus on the user's purchase history, browsing habits, price sensitivity and other characteristics, as well as the correlation between these characteristics and the user's future purchase behavior. The marketing conversion rate refers to the probability that a user is motivated and completes a purchase after a specific marketing activity (such as coupons, promotions, etc.) is implemented. This probability reflects the degree of influence of the marketing activity on the user's purchase behavior. When predicting marketing conversion rates, the preset prediction model not only considers the natural behavioral characteristics of users, but also pays special attention to the characteristics of the marketing campaign itself, such as the strength of the discount, the type of campaign, the duration of the campaign, etc., as well as the correlation between these characteristics and user purchasing behavior.

[0028] Figure 4 It is a schematic diagram of the technical process of the preset prediction model disclosed in the embodiment of this application, such as Figure 4 As shown, the model part of the prediction algorithm is divided into the model estimation and profile parts.

[0029] Figure 5 is a schematic diagram of the DeepFM model structure disclosed in the embodiment of this application, such as Figure 5 As shown, the DeepFM model consists of two parts, FM and DNN, in which the FM model mainly extracts low-order (ie, linear) features, while DNN mainly processes high-order (ie, more complex and highly abstract) features. This design enables DeepFM to balance the processing of low-order and high-order features without the need for complex artificial feature engineering. The biggest advantage is that the DeepFM model can be trained directly using the original features, and because FM and DNN share input vector features, the model is also quite good in training speed. The embodiment of the present application constructs an S-learner model based on Deep FM to predict the user's order probability. In the initial stage of the experiment, the accumulated two-week user trajectory data was used for model training. As the experiment progresses, incremental training is performed daily through AB experimental data. During the incremental training process, adaptability increases the learning rate to ensure that the model prioritizes the acquisition and learning of the latest knowledge.

[0030] S130, generating multiple red envelope thresholds according to historical data of red envelope issuance; The red envelope threshold refers to the threshold or standard at which users can receive red envelope rewards under certain conditions. These conditions may include the user's purchase amount, purchase frequency, membership level, type of activity participated in, etc. Different thresholds correspond to different red envelope reward strengths, aiming to meet the incentive needs of different users.

[0031] Optionally, generating multiple red envelope thresholds according to historical data of red envelope issuance includes: Determine a first user group to which red envelopes have been issued, obtain user data of the first user group, the user data including a first natural conversion rate and a first marketing conversion rate of each user, and calculate a first marketing gain of each user according to the first natural conversion rate and the first marketing conversion rate; Dividing the first user group into a plurality of standard red envelope groups according to the first natural conversion rate, the first marketing conversion rate, and the first marketing gain, wherein the plurality of standard red envelope groups include a non-red envelope group, a small number of red envelope groups, a large number of red envelope groups, and a normal red envelope group; The range of natural conversion rate, marketing conversion rate and marketing gain within the standard red envelope group is counted, and the red envelope threshold is determined based on the range.

[0032] Identify the user groups that have issued red envelopes in historical data, referred to here as the "first user group". Next, obtain detailed user data for the first user group, including each user's first natural conversion rate and first marketing conversion rate. The natural conversion rate refers to the probability that a user completes a purchase without additional incentives (such as red envelopes), while the marketing conversion rate refers to the probability that a user completes a purchase with red envelope incentives. In addition, it is necessary to calculate the first marketing gain of each user, that is, the additional conversion rate increase brought about by red envelope incentives. This can be calculated by comparing the conversion rates of users with and without red envelopes. Based on the user's first natural conversion rate, first marketing conversion rate, and first marketing gain, the first user group is divided into multiple standard red envelope groups. These groups include: Group without red envelopes: Both the natural conversion rate and marketing conversion rate of users are high, and the marketing gain is low, which means that these users will buy even without red envelopes, so no additional incentives are needed.

[0033] The group that sends fewer red envelopes: The natural conversion rate of users is high, but the improvement in marketing conversion rate is limited, or the marketing gain is low, indicating that these users are not very sensitive to red envelopes and the issuance of red envelopes can be appropriately reduced.

[0034] The group that sends out more red envelopes: The natural conversion rate of users is low, but the marketing conversion rate is significantly improved, or the marketing gain is high, which means that these users have a strong response to red envelope incentives and it is worth increasing the distribution of red envelopes.

[0035] Normal red envelope group: The user's natural conversion rate and marketing conversion rate are at a medium level, and the marketing gain is moderate, so red envelopes can be distributed according to conventional strategies.

[0036] For each standard red envelope group, calculate the range of its natural conversion rate, marketing conversion rate, and marketing gain. These ranges can help companies more clearly understand the behavioral characteristics and red envelope sensitivity of different user groups. Based on these ranges, combined with the company's marketing goals and cost budget, determine the red envelope thresholds corresponding to different user groups. These thresholds can be specific amounts, percentages, or quantities to guide subsequent red envelope distribution strategies.

[0037] By identifying the first user group that has issued red envelopes and obtaining their detailed user data (including the first natural conversion rate and the first marketing conversion rate), enterprises can divide users more carefully and accurately. This segmentation helps enterprises better understand the needs and preferences of different user groups, so as to formulate more targeted marketing strategies. The introduction of the first marketing gain as an evaluation indicator enables enterprises to quantitatively evaluate the actual impact of red envelope strategies on user purchasing behavior. This helps enterprises to more objectively weigh costs and benefits when formulating red envelope strategies and ensure the sustainability and effectiveness of marketing strategies. According to the first natural conversion rate, the first marketing conversion rate and the first marketing gain, the user groups are divided into multiple standard red envelope groups (including the non-red envelope group, the less red envelope group, the more red envelope group and the normal red envelope group), realizing the diversification of red envelope strategies. This diversified strategy can better meet the needs of different user groups and improve user satisfaction and loyalty. By counting the range of natural conversion rate, marketing conversion rate and marketing gain within the standard red envelope group and determining the red envelope threshold based on these ranges, enterprises can ensure the fairness and rationality of the red envelope strategy. At the same time, this scientific setting can also avoid excessive or insufficient red envelope issuance, thereby optimizing marketing costs and improving overall marketing effectiveness. Accurate red envelope strategies and scientific threshold settings can enhance users' shopping experience and satisfaction, thereby promoting users' purchasing behavior. By providing appropriate red envelope incentives for different user groups, companies can stimulate users' desire to buy, improve conversion rates and increase sales. The entire method is based on historical data on red envelope issuance for analysis and decision-making, reflecting the concept of data-driven marketing. By deeply mining and analyzing user data, companies can more accurately grasp market dynamics and user trends, providing strong support for future marketing strategy formulation.

[0038] Optionally, dividing the first user group into a plurality of standard red envelope groups according to the first natural conversion rate, the first marketing conversion rate, and the first marketing gain includes: Sort all users in the first user group from large to small according to their natural conversion rates, and obtain a first number of users ranked at the top to form a first user group; Sort the users in the first user group in ascending order of marketing gain, obtain a second number of users at the front to form a second user group, and determine the second user group as a group that does not send red envelopes; Remove users corresponding to the second user group from the first user group to obtain a second user group, sort all users in the second user group from large to small according to the natural conversion rate, obtain the third number of users ranked at the front to form a third user group, and determine the third user group as a group that sends less red envelopes; Removing users corresponding to the third user group from the second user group to obtain a third user group, sorting all users in the third user group from large to small according to marketing gain, obtaining a fourth number of users at the front to form a fourth user group, and determining the fourth user group as a multiple red envelope sending group; The fourth user group is obtained by removing users corresponding to the fourth user group from the third user group, and the fourth user group is determined as a normal red envelope sending group.

[0039] Figure 6 is a flow chart of generating a configuration threshold value disclosed in the embodiment of the present application, such as Figure 6 As shown, all users are sorted from large to small according to their natural conversion rate, and the first number of users at the top are taken to form the first user group. In the first user group, they are sorted from small to large according to their marketing gain, and the second number of users at the top are intercepted to form the second user group, which is determined as the non-red envelope group (no group). The logic here is that users with high natural conversion rates but low marketing gains may not need additional red envelope incentives because they have a high willingness to buy.

[0040] After removing the users who do not send red envelopes, the remaining users constitute the second user group. All users in the second user group are sorted again according to the natural conversion rate from large to small, and the third number of users at the top are intercepted to form the third user group, which is determined as the group that sends less red envelopes (down group). The logic here is that among the remaining users with higher natural conversion rates, a part of the users are selected as the group that sends less red envelopes, probably because although they have a certain willingness to buy, they need to be further improved through smaller red envelope incentives.

[0041] After removing the users in the group that sends less red envelopes, the remaining users constitute the third user group. All users in the third user group are sorted from large to small according to marketing gain, and the fourth number of users at the top are intercepted to form the fourth user group, which is determined to be the group that sends more red envelopes (up group). The logic here is that users with high marketing gain are more sensitive to red envelopes, so it is necessary to stimulate their purchasing behavior through larger red envelope incentives.

[0042] Normal red envelope group (equivalent to the remaining users who are not clearly divided, that is, the expansion or refinement of the equal group): After removing the users in the multiple red envelope group, the remaining users constitute the fourth user group, which is determined to be the normal red envelope group. The logic here is that the remaining users neither show a particularly high natural conversion rate nor a particularly high marketing gain, so they are classified as the normal red envelope group and may need a moderate red envelope incentive.

[0043] Through the comprehensive consideration of natural conversion rate and marketing gain, users can be segmented more accurately. This not only takes into account the user's own willingness to buy (natural conversion rate), but also the actual impact of the red envelope strategy on the user's purchasing behavior (marketing gain). User groups in different red envelope groups have different sensitivities and needs for red envelopes. Through precise segmentation, companies can provide more personalized red envelope strategies for different user groups, thereby optimizing the efficiency of red envelope use and improving user satisfaction and loyalty. By segmenting user groups and formulating corresponding red envelope strategies, companies can avoid excessive or insufficient red envelope issuance. This helps companies better control marketing costs and ensure the sustainability and effectiveness of red envelope strategies. Accurate red envelope strategies can stimulate users' desire to buy and improve conversion rates. Especially for users in the multi-red envelope group, they may be more sensitive to red envelope incentives and are therefore more likely to be converted into actual purchasing behavior. Parameters such as "first quantity", "second quantity", "third quantity" and "fourth quantity" can be flexibly adjusted according to actual conditions. This means that companies can quickly adjust red envelope strategies to adapt to new market environments based on market changes and changes in user behavior. Through personalized red envelope strategies, companies can better meet the needs of different user groups and improve user experience. This not only helps to improve user satisfaction and loyalty, but also attracts more new users.

[0044] Optionally, counting the range of the natural conversion rate, the marketing conversion rate, and the marketing gain in the standard red envelope group, and determining the red envelope threshold according to the range includes: Determine the first natural range of the second natural conversion rate and the first gain range of the second marketing gain in the group that does not send red envelopes, determine the second natural range of the third natural conversion rate in the group that sends fewer red envelopes, and determine the second gain range of the third marketing gain in the group that sends more red envelopes; The first minimum value in the first natural range is used as the first threshold, and the first maximum value in the first gain range is used as the second threshold, so that the payment amount is zero when the natural conversion rate is greater than or equal to the first threshold and the marketing gain is less than the second threshold; The second minimum value in the second natural range is used as the third threshold, so that when the natural conversion rate is less than the first threshold or the marketing gain is greater than or equal to the second threshold, and the natural conversion rate is greater than or equal to the third threshold, the payment amount is greater than zero and less than or equal to the first value; The third minimum value in the second gain range is used as the fourth threshold value, so that when the marketing gain is greater than or equal to the fourth threshold value, the payment amount is greater than or equal to a second value, and the second value is greater than the first value.

[0045] Group that does not send red envelopes: These users usually have a high natural conversion rate, but are not sensitive to red envelopes, so the marketing gain is low. The first natural range of the second natural conversion rate (assuming the average natural conversion rate of users in the group): [80%, 90%]. The first gain range of the second marketing gain (assuming the average marketing gain of users in the group): [0.1, 0.3]. Group that sends fewer red envelopes: These users have a slightly lower natural conversion rate, but still have a certain willingness to buy and have a certain response to red envelopes. The second natural range of the third natural conversion rate (assuming the average natural conversion rate of users in the group): [60%, 75%]. Group that sends more red envelopes: These users may not have a high natural conversion rate, but are very sensitive to red envelopes, so the marketing gain is high. The second gain range of the third marketing gain (assuming the average marketing gain of users in the group): [1.5, 2.5]. First threshold (lower limit of natural conversion rate): According to the first natural range of the second natural conversion rate of the group that does not send red envelopes, select the minimum value of the range, 80%, as the first threshold. This means that if the user's natural conversion rate is greater than or equal to 80%, they are likely to belong to the group that does not send red envelopes, so the amount of red envelopes issued is zero. Second threshold (upper limit of marketing gain): According to the first gain range of the second marketing gain of the group that does not send red envelopes, the maximum value of the range, 0.3, is selected as the second threshold. This means that if the user's marketing gain is less than 0.3, they are likely to belong to the group that does not send red envelopes (when the natural conversion rate condition is met at the same time), so the amount of red envelopes issued is also zero. Third threshold (lower limit of natural conversion rate, for the group that sends less red envelopes): According to the second natural range of the third natural conversion rate of the group that sends less red envelopes, the minimum value of the range, 60%, is selected as the third threshold. This means that if the user's natural conversion rate is less than 80% (i.e., does not meet the condition of the group that does not send red envelopes), and is greater than or equal to 60%, and the marketing gain is greater than or equal to 0.3 (or regardless of this condition, only based on the natural conversion rate), they may belong to the group that sends less red envelopes, so the amount of red envelopes issued is greater than zero and less than or equal to a first value (for example, 5 yuan). Fourth threshold (marketing gain lower limit, for the group that sends many red envelopes): Based on the second gain range of the third marketing gain of the group that sends many red envelopes, the minimum value of the range, 1.5, is selected as the fourth threshold. This means that if a user's marketing gain is greater than or equal to 1.5, they are likely to belong to the group that sends many red envelopes, and therefore the amount distributed is greater than or equal to a second value (for example, 10 yuan), and this second value is greater than the first value. For example, user A has a natural conversion rate of 85% and a marketing gain of 0.2. According to the threshold, user A's natural conversion rate is greater than the first threshold of 80%, and the marketing gain is less than the second threshold of 0.3, so user A belongs to the group that does not send red envelopes, and the amount of red envelopes distributed is zero. User B has a natural conversion rate of 70% and a marketing gain of 0.4.User B's natural conversion rate is less than the first threshold of 80%, and the marketing gain is greater than the second threshold of 0.3 (although the judgment here is mainly based on the natural conversion rate, because the condition of marketing gain is optional and used for further segmentation), and the natural conversion rate of user B is greater than or equal to the third threshold of 60%, so user B may belong to the group that sends less red envelopes, and the amount of distribution is greater than zero and less than or equal to 5 yuan. User C's natural conversion rate is 55%, and the marketing gain is 2.0. User C's natural conversion rate is less than the third threshold of 60% (but this does not affect the judgment, because it is already known that user C's marketing gain is very high), and the marketing gain is greater than or equal to the fourth threshold of 1.5, so user C belongs to the group that sends more red envelopes, and the amount of distribution is greater than or equal to 10 yuan.

[0046] By determining the range of natural conversion rates and marketing gains for different red envelope groups (no red envelope group, few red envelope groups, and many red envelope groups), companies can formulate red envelope strategies more finely. This helps ensure that red envelopes are issued more accurately and avoid waste of resources. Setting clear red envelope thresholds (first threshold, second threshold, third threshold, and fourth threshold) makes the logic of red envelope issuance clearer. Based on the user's natural conversion rate and marketing gain, companies can automatically determine how many red envelopes should be issued, or whether to issue red envelopes, thereby improving the efficiency and accuracy of red envelope issuance. By setting different red envelope amounts (zero, greater than zero and less than or equal to the first value, greater than or equal to the second value), companies can better control costs while ensuring marketing results. For users with high natural conversion rates but low marketing gains, red envelopes may not need to be issued; for users with moderate natural conversion rates but high marketing gains, smaller red envelopes can be issued; and for users with particularly high marketing gains, larger red envelopes can be issued to further stimulate their desire to buy. Accurate red envelope strategies can enhance users' shopping experience. For users who do not need red envelopes, unnecessary disturbances are avoided; for users who need red envelope incentives, they can receive the appropriate red envelope amount, thereby enhancing user satisfaction and loyalty. The setting of the red envelope threshold can be flexibly adjusted according to actual conditions. As the market environment and user behavior change, companies can quickly adjust red envelope strategies to adapt to the new market environment. Accurate red envelope strategies can stimulate users' desire to buy and improve conversion rates. Especially for users in the multi-red envelope group, they may be more sensitive to red envelope incentives, so they are more likely to be converted into actual purchasing behavior, thereby increasing sales.

[0047] S140. Determine a red envelope sending strategy for the target user based on the order probability and the multiple red envelope thresholds, and determine the amount of the red envelope based on the red envelope sending strategy.

[0048] When determining the red envelope sending strategy for target users, the company will comprehensively consider the user's order probability and multiple red envelope thresholds. Specifically, the company will determine the user group corresponding to the user (such as high probability purchase group, medium probability purchase group, low probability purchase group, etc.) based on the user's order probability, and determine the user's red envelope sending strategy based on the red envelope threshold corresponding to the user group. After determining the red envelope sending strategy, the company also needs to determine the specific red envelope amount based on the strategy. The size of the red envelope amount should match the user's purchase intention and the company's marketing goals. For example, for users who need strong incentives, the company may choose to issue larger red envelopes; for users who already have a strong purchase intention, the company may choose to issue smaller red envelopes or not issue red envelopes at all.

[0049] Optionally, determining the red envelope sending strategy for the target user according to the order probability and the multiple red envelope thresholds further includes: Analyze the historical behavior data of the target user, wherein the historical behavior data includes browsing history, purchase history, evaluation history, collection history, sharing history, and hotel check-in history; Building a user profile based on the historical behavior data, the user profile includes the user's preference characteristics, spending power, active time periods, travel habits, and price sensitivity; Based on the user portrait, hotel red envelopes of target type and target amount are recommended to the target user.

[0050] Historical behavior data refers to the user's various behavior records on the network platform, including but not limited to: Browsing history: Hotel pages, room types, prices, and other information that users have browsed, reflecting the user’s initial interests and concerns.

[0051] Purchase records: The user’s past hotel reservation records, including the booked room type, price, check-in date, etc., can reflect the user’s consumption habits and preferences.

[0052] Evaluation records: Users' evaluation of hotel services, including text evaluation and ratings, reflects users' satisfaction and expectations of hotel services.

[0053] Favorite records: Hotels or room types that users have favorited, indicating that the user has potential interest in these options.

[0054] Sharing history: Hotel information shared by users, which may involve content that users want to recommend to others.

[0055] Hotel check-in records: The user's actual check-in records, including the hotel they checked into, length of stay, amount of money spent, etc., provide direct evidence of the user's accommodation behavior.

[0056] By collecting and analyzing the historical behavior data of target users, a user profile can be constructed, which contains the following key features: Preference characteristics: hotel type, price range, geographical location, etc. preferred by the user.

[0057] Spending capacity: the economic status and willingness to pay reflected by the user’s past spending records.

[0058] Active time: The time pattern when users book hotels, such as holidays, weekends or specific months.

[0059] Travel habits: users’ travel frequency, destination preferences, travel methods, etc.

[0060] Price sensitivity: whether users tend to choose low-priced hotels or are willing to pay extra for specific services.

[0061] After building a user portrait, companies can more accurately recommend hotel red envelopes that meet the needs of target users: Target type: Based on the user's preference characteristics and spending power, recommend hotel red envelope types suitable for the user, such as luxury hotel red envelopes, economy hotel red envelopes, consecutive stay discount red envelopes, etc.

[0062] Target amount: Based on the user's price sensitivity and historical consumption records, set a reasonable red envelope amount to attract users while maintaining the company's profit level.

[0063] By analyzing the historical behavior data of target users, such as browsing history and purchase history, we can gain an in-depth understanding of users' preferences and consumption habits. This helps recommend hotel red envelopes that are more in line with users' needs and preferences, and improves the attractiveness and usage rate of red envelopes. The construction of user portraits provides multi-dimensional information about user preference characteristics, consumption capacity, active time periods, etc. This information enables companies to more accurately target target users and formulate personalized marketing strategies based on their characteristics and needs, including the timing, type, and amount of red envelopes. Personalized red envelope recommendations and precise marketing strategies can enhance users' shopping experience. Users feel not only material benefits, but also the attention and respect of companies for their needs, thereby enhancing user loyalty and satisfaction. Through precise red envelope distribution strategies, companies can avoid the waste of red envelope resources. For users with strong purchasing intentions and low price sensitivity, the issuance of red envelopes can be appropriately reduced; while for users with weak purchasing intentions and high price sensitivity, red envelopes of larger amounts can be issued to stimulate their desire to buy. This can not only ensure marketing effectiveness, but also effectively control costs.

[0064] Optionally, the recommending a hotel red packet of a target type and a target amount to the target user according to the user portrait includes: Identify target features in the user portrait, the target features including the type of hotel where the target user stays most often, the geographical location where the target user travels most frequently, and the average price range of the user's historical consumption; According to the target feature, a red envelope type matching the target user is selected from a predefined red envelope type library, wherein the red envelope type library includes but is not limited to economy hotel discount coupons, luxury hotel discount coupons, and specific brand hotel exclusive coupons; Combining the natural conversion rate and marketing conversion rate of the target user and the historical conversion rate data of each red envelope type in the red envelope type library, the target amount is obtained through a prediction algorithm; A hotel red envelope is distributed to the target user according to the red envelope type and the target amount.

[0065] By analyzing the user's historical check-in records, we can know the type of hotel the user prefers, such as economy hotels, luxury hotels, or mid-range hotels. This information is crucial for recommending the right type of red envelope for the user. The frequency of a user's travel in different geographical locations can reflect their possible work, living, or leisure areas. This information helps recommend hotel red envelopes near their frequently visited locations to users, making the red envelopes more practical and attractive. By analyzing the user's historical consumption records, we can know the average price range of the user's hotel consumption. This information is important for determining the amount range of the red envelope to ensure that the red envelope meets the user's spending capacity and expectations. After identifying the target features in the user portrait, it is necessary to select the red envelope type that matches the user from the predefined red envelope type library. The red envelope type library may include multiple types such as economy hotel discount coupons, luxury hotel full discount coupons, and specific brand hotel exclusive coupons. For example, for users who prefer economy hotels and have a lower historical consumption price, economy hotel discount coupons can be recommended to them; while for users who prefer luxury hotels and have a higher historical consumption price, luxury hotel full discount coupons or specific brand hotel exclusive coupons can be recommended to them. After determining the type of red envelope, it is necessary to combine the user's natural conversion rate and marketing conversion rate, as well as the historical conversion rate data of each red envelope type in the red envelope type library, to obtain the target amount through the prediction algorithm. At the same time, it is also necessary to refer to the historical conversion rate data of each red envelope type in the red envelope type library to understand the red envelope amount that is more easily accepted and used by users. By comprehensively considering these factors, the prediction algorithm can be used to obtain the optimal red envelope amount for the user to ensure that the red envelope can maximize the user's desire to buy. After determining the red envelope type and target amount, hotel red envelopes can be issued to target users based on this information. The issuance of red envelopes can be carried out through channels such as the user's mobile phone APP, email or text message to ensure that users can receive and use the red envelopes in a timely manner.

[0066] This embodiment also discloses a hotel red envelope distribution system. Figure 7 Schematic diagram of the module of the hotel red envelope distribution system disclosed in the embodiment of the present application, such as Figure 7 As shown, the system includes a collection module 701, a prediction module 702, a threshold module 703 and a sending module 704, wherein: The acquisition module 701 is configured to obtain trajectory features of the target user, wherein the trajectory features include the type, price range and geographical location of the hotels browsed by the target user, and the departure place, destination and time of the air ticket, train ticket or bus ticket searched by the target user; A prediction module 702 is configured to obtain an order probability according to the trajectory characteristics using a preset prediction model, wherein the order probability includes a natural conversion rate and a marketing conversion rate; Threshold module 703, configured to generate multiple red envelope thresholds according to historical data of red envelope distribution; The sending module 704 is configured to determine the red envelope sending strategy of the target user according to the order probability and the multiple red envelope thresholds, and determine the amount of the red envelope according to the red envelope sending strategy.

[0067] Optionally, the system includes a training module, wherein the training module is configured to: Obtaining the original trajectory data of all users before the current moment, and converting the original trajectory data into original features, normalizing the numerical features, and label encoding the discrete features; Use the target feature to split, calculate the difference between the information entropy of the node before the split and the sum of the information entropy of the two child nodes after the split to obtain the information gain, wherein the target feature is any original feature; Accumulating the information gains of all splitting points of the target feature in all decision trees to obtain the importance of the target feature, and screening out a feature set that meets the preset requirements according to the importance; The initial prediction model is trained according to the feature set to obtain the preset prediction model.

[0068] Optionally, the threshold module 703 is configured to: Determine a first user group to which red envelopes have been issued, obtain user data of the first user group, the user data including a first natural conversion rate and a first marketing conversion rate of each user, and calculate a first marketing gain of each user according to the first natural conversion rate and the first marketing conversion rate; Dividing the first user group into a plurality of standard red envelope groups according to the first natural conversion rate, the first marketing conversion rate, and the first marketing gain, wherein the plurality of standard red envelope groups include a non-red envelope group, a small number of red envelope groups, a large number of red envelope groups, and a normal red envelope group; The range of natural conversion rate, marketing conversion rate and marketing gain within the standard red envelope group is counted, and the red envelope threshold is determined based on the range.

[0069] Optionally, the threshold module 703 is configured to: Sort all users in the first user group from large to small according to their natural conversion rates, and obtain a first number of users ranked at the top to form a first user group; Sort the users in the first user group in ascending order of marketing gain, obtain a second number of users at the front to form a second user group, and determine the second user group as a group that does not send red envelopes; Remove users corresponding to the second user group from the first user group to obtain a second user group, sort all users in the second user group from large to small according to the natural conversion rate, obtain the third number of users ranked at the front to form a third user group, and determine the third user group as a group that sends less red envelopes; Removing users corresponding to the third user group from the second user group to obtain a third user group, sorting all users in the third user group from large to small according to marketing gain, obtaining a fourth number of users at the front to form a fourth user group, and determining the fourth user group as a multiple red envelope sending group; The fourth user group is obtained by removing users corresponding to the fourth user group from the third user group, and the fourth user group is determined as a normal red envelope sending group.

[0070] Optionally, the threshold module 703 is configured to: Determine the first natural range of the second natural conversion rate and the first gain range of the second marketing gain in the group that does not send red envelopes, determine the second natural range of the third natural conversion rate in the group that sends fewer red envelopes, and determine the second gain range of the third marketing gain in the group that sends more red envelopes; The first minimum value in the first natural range is used as the first threshold, and the first maximum value in the first gain range is used as the second threshold, so that the payment amount is zero when the natural conversion rate is greater than or equal to the first threshold and the marketing gain is less than the second threshold; The second minimum value in the second natural range is used as the third threshold, so that when the natural conversion rate is less than the first threshold or the marketing gain is greater than or equal to the second threshold, and the natural conversion rate is greater than or equal to the third threshold, the payment amount is greater than zero and less than or equal to the first value; The third minimum value in the second gain range is used as the fourth threshold value, so that when the marketing gain is greater than or equal to the fourth threshold value, the payment amount is greater than or equal to a second value, and the second value is greater than the first value.

[0071] Optionally, the sending module 704 is configured to: Analyze the historical behavior data of the target user, wherein the historical behavior data includes browsing history, purchase history, evaluation history, collection history, sharing history, and hotel check-in history; Building a user profile based on the historical behavior data, the user profile includes the user's preference characteristics, spending power, active time periods, travel habits, and price sensitivity; Based on the user portrait, hotel red envelopes of target type and target amount are recommended to the target user.

[0072] Optionally, the sending module 704 is configured to: Identify target features in the user portrait, the target features including the type of hotel where the target user stays most often, the geographical location where the target user travels most frequently, and the average price range of the user's historical consumption; According to the target feature, a red envelope type matching the target user is selected from a predefined red envelope type library, wherein the red envelope type library includes but is not limited to economy hotel discount coupons, luxury hotel discount coupons, and specific brand hotel exclusive coupons; Combining the natural conversion rate and marketing conversion rate of the target user and the historical conversion rate data of each red envelope type in the red envelope type library, the target amount is obtained through a prediction algorithm; A hotel red envelope is distributed to the target user according to the red envelope type and the target amount.

[0073] It should be noted that: when the device provided in the above embodiment realizes its function, only the division of the above functional modules is used as an example. In actual application, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device and method embodiments provided in the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0074] This embodiment also discloses an electronic device, referring to Figure 8 The electronic device may include: at least one processor 801 , at least one communication bus 802 , a user interface 803 , a network interface 804 , and at least one memory 805 .

[0075] The communication bus 802 is used to realize the connection and communication between these components.

[0076] The user interface 803 may include a display screen (Display) and a camera (Camera). The optional user interface 803 may also include a standard wired interface and a wireless interface.

[0077] The network interface 804 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0078] Among them, the processor 801 may include one or more processing cores. The processor 801 uses various interfaces and lines to connect various parts in the entire server, and executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 805, and calling data stored in the memory 805. Optionally, the processor 801 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 801 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 801, and it can be implemented separately through a chip.

[0079] Among them, the memory 805 may include a random access memory (Random Access Memory, RAM) and may also include a read-only memory (Read-Only Memory). Optionally, the memory 805 includes a non-transitory computer-readable storage medium. The memory 805 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 805 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 805 may also be optionally at least one storage device located away from the aforementioned processor 801. As Figure 8 As shown, the memory 805 as a computer storage medium may include an operating system, a network communication module, a user interface 803 module, and an application program for a hotel red envelope distribution method.

[0080] exist Figure 8 In the electronic device shown, the user interface 803 is mainly used to provide an input interface for the user and obtain data input by the user; and the processor 801 can be used to call the application program for storing the hotel red envelope distribution method in the memory 805. When executed by one or more processors 801, the electronic device executes one or more methods in the above-mentioned embodiments.

[0081] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the order of the actions described, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required for the present application.

[0082] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0083] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0084] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0085] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0086] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a memory 805, including several instructions for a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the various embodiments of the present application. The aforementioned memory 805 includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a magnetic disk or an optical disk.

[0087] The above is only an exemplary embodiment of the present disclosure and cannot be used to limit the scope of the present disclosure. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the disclosure of the specification, those skilled in the art will easily think of other embodiments of the present disclosure. This application is intended to cover any modification, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the technical field that are not recorded in the present disclosure. The description and examples are only regarded as exemplary, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. A method for distributing hotel red envelopes, characterized in that: Applied to a travel platform, the method includes: Acquire trajectory features of the target user, the trajectory features including the type, price range, and geographic location of hotels browsed by the target user, and the departure place, destination, and time of the air ticket, train ticket, or bus ticket searched by the target user; Using a preset prediction model to obtain an order probability according to the trajectory characteristics, the order probability includes a natural conversion rate and a marketing conversion rate; Generate multiple red envelope thresholds based on historical data of red envelope issuance; A red envelope sending strategy for the target user is determined based on the order probability and the multiple red envelope thresholds, and the amount of the red envelope is determined based on the red envelope sending strategy.

2. The hotel red envelope distribution method according to claim 1, characterized in that: The method comprises: Obtaining the original trajectory data of all users before the current moment, and converting the original trajectory data into original features, normalizing the numerical features, and label encoding the discrete features; Use the target feature to split, calculate the difference between the information entropy of the node before the split and the sum of the information entropy of the two child nodes after the split to obtain the information gain, wherein the target feature is any original feature; Accumulating the information gains of all splitting points of the target feature in all decision trees to obtain the importance of the target feature, and screening out a feature set that meets the preset requirements according to the importance; The initial prediction model is trained according to the feature set to obtain the preset prediction model.

3. The hotel red envelope distribution method according to claim 1, characterized in that: Generating multiple red envelope thresholds according to historical data of red envelope issuance includes: Determine a first user group to which red envelopes have been issued, obtain user data of the first user group, the user data including a first natural conversion rate and a first marketing conversion rate of each user, and calculate a first marketing gain of each user according to the first natural conversion rate and the first marketing conversion rate; Dividing the first user group into a plurality of standard red envelope groups according to the first natural conversion rate, the first marketing conversion rate, and the first marketing gain, wherein the plurality of standard red envelope groups include a non-red envelope group, a small number of red envelope groups, a large number of red envelope groups, and a normal red envelope group; The range of natural conversion rate, marketing conversion rate and marketing gain within the standard red envelope group is counted, and the red envelope threshold is determined based on the range.

4. The hotel red envelope distribution method according to claim 3, characterized in that: The dividing the first user group into a plurality of standard red envelope groups according to the first natural conversion rate, the first marketing conversion rate and the first marketing gain comprises: Sort all users in the first user group from large to small according to their natural conversion rates, and obtain a first number of users ranked at the top to form a first user group; Sort the users in the first user group in ascending order of marketing gain, obtain a second number of users at the front to form a second user group, and determine the second user group as a group that does not send red envelopes; Remove users corresponding to the second user group from the first user group to obtain a second user group, sort all users in the second user group from large to small according to the natural conversion rate, obtain the third number of users ranked at the front to form a third user group, and determine the third user group as a group that sends less red envelopes; Removing users corresponding to the third user group from the second user group to obtain a third user group, sorting all users in the third user group from large to small according to marketing gain, obtaining a fourth number of users at the front to form a fourth user group, and determining the fourth user group as a multiple red envelope sending group; The fourth user group is obtained by removing users corresponding to the fourth user group from the third user group, and the fourth user group is determined as a normal red envelope sending group.

5. The hotel red envelope distribution method according to claim 3, characterized in that: The counting of the range of the natural conversion rate, the marketing conversion rate and the marketing gain in the standard red envelope group, and determining the red envelope threshold according to the range includes: Determine the first natural range of the second natural conversion rate and the first gain range of the second marketing gain in the group that does not send red envelopes, determine the second natural range of the third natural conversion rate in the group that sends fewer red envelopes, and determine the second gain range of the third marketing gain in the group that sends more red envelopes; The first minimum value in the first natural range is used as the first threshold, and the first maximum value in the first gain range is used as the second threshold, so that the payment amount is zero when the natural conversion rate is greater than or equal to the first threshold and the marketing gain is less than the second threshold; The second minimum value in the second natural range is used as the third threshold, so that when the natural conversion rate is less than the first threshold or the marketing gain is greater than or equal to the second threshold, and the natural conversion rate is greater than or equal to the third threshold, the payment amount is greater than zero and less than or equal to the first value; The third minimum value in the second gain range is used as the fourth threshold value, so that when the marketing gain is greater than or equal to the fourth threshold value, the payment amount is greater than or equal to a second value, and the second value is greater than the first value.

6. The hotel red envelope distribution method according to claim 1, characterized in that: The determining of the red envelope sending strategy for the target user according to the order probability and the multiple red envelope thresholds further comprises: Analyze the historical behavior data of the target user, wherein the historical behavior data includes browsing history, purchase history, evaluation history, collection history, sharing history, and hotel check-in history; Building a user profile based on the historical behavior data, the user profile includes the user's preference characteristics, spending power, active time periods, travel habits, and price sensitivity; Based on the user portrait, hotel red envelopes of target type and target amount are recommended to the target user.

7. The hotel red envelope distribution method according to claim 6, characterized in that: The recommending of a hotel red packet of a target type and a target amount for the target user according to the user portrait includes: Identify target features in the user portrait, the target features including the type of hotel where the target user stays most often, the geographical location where the target user travels most frequently, and the average price range of the user's historical consumption; According to the target feature, a red envelope type matching the target user is selected from a predefined red envelope type library, wherein the red envelope type library includes but is not limited to economy hotel discount coupons, luxury hotel discount coupons, and specific brand hotel exclusive coupons; Combining the natural conversion rate and marketing conversion rate of the target user and the historical conversion rate data of each red envelope type in the red envelope type library, the target amount is obtained through a prediction algorithm; A hotel red envelope is distributed to the target user according to the red envelope type and the target amount.

8. A hotel red envelope distribution system, characterized in that: It includes a collection module, a prediction module, a threshold module and a sending module, among which: a collection module configured to obtain trajectory features of a target user, the trajectory features including the type, price range, and geographic location of hotels browsed by the target user, and the departure place, destination, and time of air tickets, train tickets, or bus tickets searched by the target user; A prediction module, configured to obtain an order probability according to the trajectory characteristics using a preset prediction model, wherein the order probability includes a natural conversion rate and a marketing conversion rate; A threshold module, configured to generate multiple red envelope thresholds based on historical data of red envelope distribution; The sending module is configured to determine the red envelope sending strategy of the target user according to the order probability and the multiple red envelope thresholds, and determine the amount of the red envelope according to the red envelope sending strategy.

9. An electronic device, characterized in that: It includes a processor, a memory, a user interface and a network interface, the memory is used to store instructions, the user interface and the network interface are both used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed, the method according to any one of claims 1 to 7 is executed.

Images