User behavior-based point shopping mall service method and system

Abstract

The invention discloses a point shopping mall service method and system based on user behaviors, and relates to the technical field of point shopping malls. Comprising the steps of collecting behavior data of a user in real time and generating a behavior feature vector; on the basis of the behavior feature vector and the life cycle stage of the user, personalized integral commodities and activities are recommended; user integral history is analyzed, an integral strategy is dynamically adjusted, and customized incentives are provided for different types of users; automatically managing the period of validity of the points, reminding the user of the points to be expired, and guiding the user to consume the points in time in combination with the personalized point commodities and activities; the integral strategy is updated regularly, the excitation mode is optimized according to the user behavior change, and the adaptability and the user experience are improved. According to the invention, by dynamically adjusting the point strategy and combining point validity period management, the user is helped to timely consume points about to be expired, and point loss is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of points mall, and in particular to a points mall service method and system based on user behavior. Background Art

[0002] As a new type of consumption and marketing tool, the points mall uses data such as user consumption behavior and interactive behavior on the platform to encourage users to participate in product redemption or activities through the points system, thereby improving user activity and loyalty. The existing points mall system usually attracts users through the acquisition and redemption of points, but with the intensification of market competition, simple point rewards can no longer meet the increasingly complex user needs. Therefore, the points mall has gradually turned to more personalized recommendations and incentive mechanisms to improve the efficiency of point use and user participation.

[0003] The existing points mall system still has some obvious shortcomings. Although most platforms can recommend products based on users' historical behavior data, these recommendations often ignore the differences in users' needs at different stages of their life cycle and fail to effectively provide personalized incentive plans for different types of users. In addition, in the traditional points system, the validity management and reminder mechanism of points are relatively simple, and there is a lack of an intelligent points expiration management system, which often leads to users failing to consume points that are about to expire in a timely manner, resulting in problems such as point waste and user loss. Summary of the invention

[0004] In view of the problems existing in the above background technology, the present invention is proposed.

[0005] Therefore, the problem to be solved by the present invention is how to improve user activity and loyalty through more accurate personalized recommendations, dynamic point strategy adjustment and intelligent point expiration management.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: In the first aspect, an embodiment of the present invention provides a points mall service method based on user behavior, which includes: real-time collection of user behavior data and generation of behavior feature vectors; recommending personalized points products and activities based on the user's behavior feature vectors and life cycle stages; analyzing user points history, dynamically adjusting points strategies, and providing customized incentives for different types of users; automatically managing the validity period of points and reminding users of points that are about to expire, and guiding users to consume points in a timely manner in combination with the personalized points products and activities; regularly updating points strategies, optimizing incentive methods according to changes in user behavior, and improving adaptability and user experience.

[0007] As a preferred solution of the user behavior-based points mall service method of the present invention, the generation of the behavior feature vector includes: setting a preliminary behavior feature vector Represents the behavioral characteristics of user u, including behavioral characteristics of multiple dimensions; quantifies and encodes the behavioral characteristics; weights or fuses different types of behavioral characteristics to generate the final behavioral feature vector When performing weighting or fusion of different types of behavioral features, the behavioral features are standardized or normalized.

[0008] As a preferred solution of the user behavior-based points mall service method of the present invention, wherein: the final behavior feature vector is generated for: ; in, is the behavioral feature vector of user u in the kth time window, is the weight of the kth time window, and K is the total number of behavioral features.

[0009] As a preferred solution of the user behavior-based points mall service method of the present invention, the life cycle stages include new users, active users, silent users and users at risk of churn.

[0010] As a preferred solution of the points mall service method based on user behavior of the present invention, the division process of the life cycle stage is as follows: a weighted decay model is used to reflect the user's behavior: ; in, is the behavioral activity of user u at time t, is the time when user behavior i occurs, t is the time, is the basic activity of user u in behavior i, is the weight of the behavior type, is the decay factor, n is the number of all behavior records of user u; according to Value, divide the user's life cycle stages through different thresholds.

[0011] As a preferred solution of the user behavior-based points mall service method described in the present invention, the recommendation of personalized points products and activities includes the following steps: hierarchical processing of the user's behavior feature vector according to the life cycle stage; and adjusting the points recommendation parameters for each life cycle stage.

[0012] As a preferred solution of the user behavior-based points mall service method of the present invention, the adjustment of the points recommendation parameters at each life cycle stage is as follows: ; in, The recommended value of the points product or activity recommended for user u, is the weight coefficient of user u for the jth interest dimension at the life cycle stage le, is the kth behavior feature vector of user u at life cycle order le, is the weight of the kth behavior feature vector of user u at the life cycle order le, is the behavioral feature vector of product or activity p, is the interest of user u in product or activity p at time t, generated based on the user’s historical behavior, and le is the life cycle stage.

[0013] In the second aspect, an embodiment of the present invention provides a points mall service system based on user behavior, which includes: a behavior data collection module, which is used to collect user behavior data in real time and generate a behavior feature vector; a personalized recommendation engine module, which is used to recommend personalized points products and activities based on the user's behavior feature vector and life cycle stage; a points strategy analysis and adjustment module, which is used to analyze the user's points history, dynamically adjust the points strategy, and provide customized incentives for different types of users; a points management and reminder module, which is used to automatically manage the validity period of points and remind users of points that are about to expire, and guide users to consume points in time in combination with the personalized points products and activities; a points strategy optimization and adaptation module, which is used to regularly update the points strategy, optimize the incentive method according to changes in user behavior, and improve adaptability and user experience.

[0014] In a third aspect, an embodiment of the present invention provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program instructions are executed by the processor, the steps of the user behavior-based points mall service method as described in the first aspect of the present invention are implemented.

[0015] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program instructions are executed by a processor, the steps of the point mall service method based on user behavior as described in the first aspect of the present invention are implemented.

[0016] The beneficial effects of the present invention are as follows: the present invention collects user behavior data in real time and generates behavior feature vectors, combines the user life cycle stage, recommends personalized points products and activities, optimizes the points acquisition and use strategy, and realizes refined management of user behavior; the present invention also helps users consume points that are about to expire in time and reduce the loss of points by dynamically adjusting the points strategy and combining the points validity period management. The present invention has significant advantages in accurately capturing user needs, dynamically adjusting incentive strategies, and intelligently managing the validity period of points. It not only improves the use efficiency of the points mall, but also optimizes the user's participation experience, and has a high market application potential. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0018] Figure 1 The present invention is a flowchart of a points mall service method based on user behavior.

[0019] Figure 2 This is a structural diagram of the points mall service system based on user behavior. DETAILED DESCRIPTION

[0020] In order to make the above-mentioned objectives, behavioral characteristics and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0021] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0022] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific behavior, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0023] Embodiment 1: Reference Figure 1~Figure 2 , which is the first embodiment of the present invention, and provides a points mall service method based on user behavior, including: S1: Collect user behavior data in real time and generate behavior feature vectors to provide basic data support for personalized services.

[0024] Use multiple channels to collect user behavior data in the points mall, including but not limited to browsing products, searching keywords, clicking activities, adding to shopping carts, redeeming points, etc.; through interface integration or event monitoring, record every user behavior in real time.

[0025] The privacy policy must be followed during data collection to ensure data security and user privacy protection.

[0026] Clean the collected raw behavior data to remove noise data and redundant information, such as repeated operations, abnormal behavior records, etc., and classify them according to behavior types (such as purchasing behavior, browsing behavior, evaluation behavior, etc.) to facilitate subsequent processing.

[0027] Based on the cleaned behavioral data, the user's behavioral characteristics are extracted. These behavioral characteristics can be defined according to the specific business needs of the mall. Common behavioral characteristics include: single behavioral characteristics: such as the number of occurrences of a specific behavior (such as the number of times a certain product is browsed), the time interval between behaviors; comprehensive behavioral characteristics: such as user activity, purchase tendency, loyalty, etc.; contextual behavioral characteristics: extracting behavioral characteristics based on different situations, such as whether to participate in the purchase during the promotion period, whether the purchase time is related to holidays or specific activities, etc.

[0028] Assume there is a preliminary behavior feature vector Represents the behavioral characteristics of user u, including behavioral characteristics of multiple dimensions.

[0029] Furthermore, to quantify and encode behavioral features, such as purchase categories and product types, they need to be converted into numerical values ​​through One-Hot Encoding or Embedding. In addition, if the time series of behaviors is involved (for example, the user's active period), it can be processed through the sliding window method or the time series behavior feature extraction method.

[0030] Generate the final behavioral feature vector When generating a user's behavior, you may need to weight or fuse different types of behavior features. For example, you can generate a weighted behavior feature vector based on the user's behavior history (such as the behavior in the last 30 days): ; in, is the behavioral feature vector of user u in the kth time window, is the weight of the kth time window, and K is the total number of behavioral features.

[0031] In order to ensure the dimensional consistency between the dimensions of various behavioral characteristics, the behavioral characteristics are usually standardized or normalized; finally, the generated behavioral feature vectors are stored in the database.

[0032] S2: Based on the user's behavioral feature vector and life cycle stage, recommend personalized points products and activities to motivate users to participate more actively.

[0033] S2.1: Divide the user's life cycle stages based on the user's behavioral feature vector and interaction time in the system.

[0034] The division of user life cycle stages is achieved based on the user's behavioral feature vector and the interaction time in the system. The present invention mainly divides the life cycle stages into four levels: new users, active users, silent users, and users at risk of churn.

[0035] Preferably, a dynamic weighting mechanism is used to consider the user's behavior changes, and the corresponding life cycle stage is dynamically adjusted according to the decreasing effect and timing characteristics of the user's behavior. The specific operation steps are as follows: User behavior usually weakens over time. In order to more accurately reflect this attenuation, a weighted attenuation model is used: ; in, is the behavioral activity of user u at time t, is the time when user behavior i occurs, t is the time, is the basic activity of user u in behavior i, is the weight of the behavior type, is the decay factor, and n is the number of all behavior records of user u.

[0036] according to Values, different thresholds can be used to divide the user's life cycle stages.

[0037] S2.2: Perform hierarchical processing on the user’s behavioral feature vectors according to the life cycle stage.

[0038] Define a weighted behavioral feature vector for each stage and dynamically adjust the weight when making recommendations , ensuring that important behavioral features are highlighted at different stages. For example, for active users, high-frequency interactive behavioral features can be extracted first, while for new users, the focus can be on first-time purchases and browsing behaviors.

[0039] S2.3: Based on the behavioral feature vectors of the stratified users, set the points recommendation parameters for each life cycle level.

[0040] Adjust the points recommendation parameters for each life cycle stage. For example, the types of recommended products and the frequency of activity triggering for new users may be higher than those for active users, but recommendations for active users will focus more on interactivity and high-value products.

[0041] Specifically, at each life cycle stage, when setting recommended parameters for users, dynamic adjustments are made based on the weights of the behavioral feature vectors in S2.2: ; in, The recommended value of the points product or activity recommended for user u, is the weight coefficient of user u for the jth interest dimension at the life cycle stage le, is the kth behavior feature vector of user u at life cycle order le, is the weight of the kth behavior feature vector of user u at the life cycle order le, is the behavioral feature vector of product or activity p, is the interest of user u in product or activity p at time t, generated based on the user’s historical behavior, and le is the life cycle stage.

[0042] Among them, higher-value point products are allocated first to new users and users at risk of churn, while interactive activities are recommended to active users to increase the frequency of their point usage.

[0043] Furthermore, the recommendation parameters and hierarchical behavior feature vectors are called to dynamically generate a list of recommended points products and activities, and update it in real time on the user interface.

[0044] S3: Analyze user points history, dynamically adjust points strategy, and provide customized incentives for different types of users.

[0045] S3.1: Collect and store the user's historical points acquisition and usage records, and extract the user's points usage behavior characteristics based on historical data such as the user's points accumulation speed, redemption frequency, and redemption product types.

[0046] S3.2: Identify the preferences of different types of users based on the characteristics of point usage behavior. For example, users with high redemption frequency prefer to redeem points for goods, while low-frequency users prefer to accumulate points. Set different point strategy adjustment parameters for each user type.

[0047] Specifically, users with a high redemption frequency tend to redeem goods frequently, so their participation needs to be incentivized by increasing the speed of obtaining points or increasing the types of redemption goods; users with a low redemption frequency focus more on the accumulation of points, so their activity can be incentivized by providing redemption discounts for high-value goods and increasing the channels for obtaining points.

[0048] Among them, high redemption frequency users and low redemption frequency users are classified based on the users' redemption frequency.

[0049] Furthermore, the integral strategy adjustment parameters include: Adjust the points earning rate for high redemption frequency users : ; in, It is the rate at which points are earned that is increased for users with a high redemption frequency.

[0050] Provide points redemption discounts for low redemption frequency users : ; in, is the discount factor, which is dynamically adjusted according to the user's point balance. is the user's point balance.

[0051] Display the points strategy through notifications or in the user interface to guide users to use points more frequently to encourage them to participate in mall activities.

[0052] S4: Automatically manage the validity period of points and remind users of points that are about to expire, combine recommended products with personalized points products and activities, and guide users to consume points in a timely manner.

[0053] Specifically, the time when each user obtains points and the validity period are recorded, the points that are about to expire are monitored in real time, and the expiration reminder date is automatically calculated. The validity period of the points is usually related to the date when the points are obtained, and there is a fixed expiration time (such as 1 year, 6 months, etc.). Every time a user obtains new points, the system needs to record the time when the points are obtained and the validity period, and monitor when the points are close to expiration.

[0054] When a user's points are about to expire, an expiration reminder notification is generated and sent to the user via APP message, email or SMS to ensure that the user is aware of the points status in a timely manner and reduce the points loss rate.

[0055] In the reminder notification, the user's behavioral feature vector and points balance are combined to recommend redeemable high-value goods or activities, further guiding users to consume points that are about to expire and improving the efficiency of point usage.

[0056] Furthermore, we can track users’ behavior after they receive reminders about their points expiration, analyze whether they click on recommended products or redeem points, and evaluate the effectiveness of reminder strategies based on users’ point redemption behavior after the reminders, such as the time of reminders, notification methods, etc. Based on behavioral data, we can optimize reminder frequency and content to improve user response rates.

[0057] The effectiveness of the reminder strategy is judged as follows: ; in, is the total number of users who sent reminders. To evaluate the effect, Is an indicator function that returns 1 if the user redeems after the reminder, otherwise 0.

[0058] S5: Update the points strategy regularly, optimize the incentive method according to changes in user behavior, and improve adaptability and user experience.

[0059] Preferably, data on changes in user behavior in the points mall is collected regularly, including behavioral characteristics such as points acquisition, consumption habits, and activity participation, as basic data for optimizing the points strategy.

[0060] Based on behavior change data, analyze the changing trends of user needs, identify possible incentive needs, and adjust the way points are obtained, such as adding diversified tasks or increasing point rewards for specific behaviors.

[0061] Based on the analysis results, the points usage strategy is dynamically updated to recommend personalized points redemption options or limited-time discounts for different user groups to increase users' willingness to redeem.

[0062] The updated points strategy is fed back to the user interface display, and personalized points activities or recommended products are pushed to users through notifications to continuously optimize the user experience and enhance user stickiness.

[0063] Furthermore, this embodiment also provides a points mall service system based on user behavior, including: Behavior data collection module, used to collect user behavior data in real time and generate behavior feature vectors; Personalized recommendation engine module, used to recommend personalized points products and activities based on the user's behavioral feature vector and life cycle stage; Points strategy analysis and adjustment module, used to analyze user points history, dynamically adjust points strategy, and provide customized incentives for different types of users; The points management and reminder module is used to automatically manage the validity period of points and remind users of points that are about to expire, and guide users to consume points in a timely manner in combination with the personalized points products and activities; The points strategy optimization and adaptation module is used to regularly update the points strategy, optimize the incentive method according to changes in user behavior, and improve adaptability and user experience.

[0064] This embodiment also provides a computer device, which is suitable for the case of a points mall service method based on user behavior, including a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the points mall service method based on user behavior as proposed in the above embodiment.

[0065] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covered on the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0066] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for implementing a points mall service based on user behavior as proposed in the above embodiment is implemented.

[0067] In summary, the present invention collects user behavior data in real time and generates behavior feature vectors, combines the user life cycle stage, recommends personalized points products and activities, optimizes the points acquisition and use strategy, and realizes refined management of user behavior; the present invention also helps users consume points that are about to expire in time and reduce points loss by dynamically adjusting the points strategy and combining the points validity period management. The present invention has significant advantages in accurately capturing user needs, dynamically adjusting incentive strategies, and intelligently managing the validity period of points. It not only improves the use efficiency of the points mall, but also optimizes the user's participation experience, and has high market application potential.

[0068] Embodiment 2: Referring to Table 1 and Table 2, which are the second embodiment of the present invention, this embodiment provides a points mall service method based on user behavior. In order to verify the beneficial effects of the present invention, a scientific demonstration is carried out through simulation experiments.

[0069] This embodiment takes an online points mall as an example, and improves user activity and points usage efficiency by implementing a "points mall service method based on user behavior". The experimental data comes from 1,000 users on the platform, and all users are randomly divided into four life cycle stages (new users, active users, silent users, and users at risk of churn). The experiment collects user behavior data in real time, including browsing products, searching keywords, clicking activities, adding shopping carts, points redemption, etc., and generates personalized behavior feature vectors based on these data.

[0070] The experimental process is as follows: By connecting to the mall's behavior tracking system, various operations of users in the mall are recorded in real time, including but not limited to product browsing, keyword searching, ad clicking, adding to shopping cart, and redeeming points.

[0071] Clean the original data and remove redundant and abnormal data, such as repeated behaviors and invalid operation records. Classify user behaviors, mainly including purchasing behaviors, browsing behaviors, and interactive behaviors.

[0072] The cleaned data is quantified and encoded, and behavioral features are generated through independent encoding and embedding vectors. For example, if a user browses a certain category of goods many times within 30 days, this information is converted into numerical features and combined with the sliding window method to process time series data to generate a weighted behavioral feature vector.

[0073] Based on each user's behavioral feature vector and active time on the platform, a weighted decay model is used to divide the user's life cycle stage. Based on this model, the user's life cycle stage is dynamically adjusted.

[0074] Generate personalized recommendations for points products and activities based on life cycle stages and behavioral feature vectors. For example, recommend discounts for high-value products for the first redemption to new users, and recommend highly interactive activities to active users.

[0075] Adjust the points strategy based on the user's points acquisition, usage records and points history. Provide higher points acquisition rates or commodity redemption types to users with high redemption frequency, and provide higher value points redemption discounts to users with low redemption frequency.

[0076] For points that are about to expire, the system will automatically generate a reminder notification and recommend high-value products or activities that are about to expire based on the user's behavioral characteristics to guide users to consume their points in a timely manner.

[0077] By comparing the changes in user points usage rate and activity before and after the experiment, the effectiveness of the points mall service method is evaluated.

[0078] During the experiment, we collected and recorded data on each user's point acquisition and redemption behavior, user activity, and point consumption frequency before and after the implementation. The data will be displayed in the following two tables. Table 1 User life cycle stages and behavioral characteristics User ID Life cycle stage Behavioral characteristic 1 (number of times a product is viewed) Behavioral feature 2 (number of keyword searches) Behavioral feature 3 (number of click activities) Behavioral characteristic 4 (number of points redeemed) User activity rating 001 New Users 10 5 3 0 0.45 002 Active Users 50 30 20 15 0.85 003 Silence User 5 2 1 1 0.20 004 Users at risk of churn 3 1 1 0 0.15 005 Active Users 45 28 18 12 0.80 006 New Users 8 4 2 0 0.35 Table 2 User points history and redemption status User ID Total points Points earning rate (points / day) Points redemption frequency (times / month) Points balance Number of redeemed products Churn Risk Score 001 500 10 0 500 0 0.90 002 1500 30 10 800 5 0.05 003 200 5 0 200 1 0.75 004 50 2 0 50 0 0.95 005 1200 25 8 500 4 0.10 006 600 8 2 400 2 0.60 By comparing the points acquisition, redemption and activity changes of users before and after the experiment, we can see that: As shown in Table 1, user ID 001 and ID 006 are new users, and their behavioral characteristics are mainly reflected in the low number of browsing, searching and clicking activities, and the low activity score. Active users (ID 002 and ID 005) show a high interaction frequency, richer behavioral characteristics, and significantly improved activity scores. Through the dynamic division of life cycle stages and the weighted attenuation mechanism, the system can accurately identify and classify user needs at different stages, providing data support for personalized recommendations.

[0079] As can be seen from Table 2, active users (such as ID 002 and ID 005) not only have a high point acquisition rate, but also a significantly increased frequency of point redemption. In contrast, new users (such as ID 001 and ID 006) have a low point acquisition rate, but a zero redemption frequency, indicating that the system can effectively motivate high-frequency redemption users by dynamically adjusting the point acquisition rate and redemption strategy, and provide low-frequency users with high-value product discounts, thereby improving user engagement.

[0080] For points that are about to expire, personalized product recommendations are combined in reminder notifications to guide users to consume in time, which greatly reduces the loss rate of points. For example, the points of ID 001 and ID 006 have not been redeemed, but the system can improve the efficiency of point use through accurate recommendations and guidance of high-value products. Experimental data shows that the frequency of point redemption for active users is significantly higher, while the point consumption of users at risk of churn is lower, which shows that the strategy of point validity period management and personalized recommendation effectively reduces the waste of points.

[0081] In summary, the points mall service method of the present invention improves user activity, points consumption frequency and the overall efficiency of the mall through refined user life cycle division, personalized points strategy adjustment, intelligent points validity period management and accurate product and activity recommendations.

[0082] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A points mall service method based on user behavior, the behavior characteristics of which are: include, Collect user behavior data in real time and generate behavior feature vectors; Recommend personalized points products and activities based on the user's behavioral feature vector and life cycle stage; Analyze user points history, dynamically adjust points strategy, and provide customized incentives for different types of users; Automatically manage the validity period of points and remind users of points that are about to expire, and guide users to consume points in a timely manner in combination with the personalized points products and activities; Update the points strategy regularly, optimize the incentive method according to changes in user behavior, and improve adaptability and user experience.

2. The user behavior-based points mall service method according to claim 1, wherein the behavior is characterized by: The generation of the behavior feature vector includes: Set up preliminary behavior feature vector Represents the behavioral characteristics of user u, including behavioral characteristics of multiple dimensions; Quantify and code behavioral characteristics; Weight or fuse different types of behavior features to generate the final behavior feature vector When , different types of behavioral features are weighted or integrated; Standardize or normalize behavioral characteristics.

3. The method for providing a points mall service based on user behavior as claimed in claim 2, wherein the behavior characteristic is that: the generating of the final behavior characteristic vector for: ； in, is the behavioral feature vector of user u in the kth time window, is the weight of the kth time window, and K is the total number of behavioral features.

4. The user behavior-based points mall service method as described in claim 3, wherein the behavior characteristics are: the life cycle stages include new users, active users, silent users, and users at risk of churn.

5. The method for providing a points mall service based on user behavior as claimed in claim 4, wherein the behavior is characterized in that the life cycle stages are divided as follows: A weighted attenuation model is used to reflect user behavior: ； in, is the behavioral activity of user u at time t, is the time when user behavior i occurs, t is the moment, is the basic activity of user u in behavior i, is the weight of the behavior type, is the decay factor, n is the number of all behavior records of user u; according to Value, divide the user's life cycle stages through different thresholds.

6. The user behavior-based points mall service method according to claim 5, wherein the behavior is characterized by: The recommendation of personalized points products and activities includes the following steps: Perform hierarchical processing on the user's behavioral feature vectors according to the life cycle stage; Adjust the score recommendation parameters for each life cycle stage.

7. The user behavior-based points mall service method according to claim 6, wherein the behavior characteristics are as follows: the adjustment of the points recommendation parameters at each life cycle stage is: ； in, The recommended value of the points product or activity recommended for user u, is the weight coefficient of user u for the jth interest dimension at the life cycle stage le, is the kth behavior feature vector of user u at life cycle order le, is the weight of the kth behavior feature vector of user u at the life cycle order le, is the behavioral feature vector of product or activity p, is the interest of user u in product or activity p at time t, generated based on the user’s historical behavior, and le is the life cycle stage.

8. A points mall service system based on user behavior, based on the points mall service method based on user behavior according to any one of claims 1 to 7, wherein the behavior characteristics are: Also includes, Behavior data collection module, used to collect user behavior data in real time and generate behavior feature vectors; Personalized recommendation engine module, used to recommend personalized points products and activities based on the user's behavioral feature vector and life cycle stage; Points strategy analysis and adjustment module, used to analyze user points history, dynamically adjust points strategy, and provide customized incentives for different types of users; The points management and reminder module is used to automatically manage the validity period of points and remind users of points that are about to expire, and guide users to consume points in a timely manner in combination with the personalized points products and activities; The points strategy optimization and adaptation module is used to regularly update the points strategy, optimize the incentive method according to changes in user behavior, and improve adaptability and user experience.

9. A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and wherein the behavior characteristic is that when the processor executes the computer program, the steps of the user behavior-based points mall service method described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, wherein the behavior characteristic is that when the computer program is executed by a processor, the steps of the user behavior-based points mall service method described in any one of claims 1 to 7 are implemented.

Images