Method and device for predicting user loss probability and electronic equipment

The timing large language model predicts the probability of user churn, captures changes in user behavior, and adjusts the recommendation system strategy, solves the problem of inaccurate user churn prediction in the existing technology, and improves the user retention rate and the sustainability of the recommendation system.

CN119006040AActive Publication Date: 2024-11-22ZHIZHESIHAIBEIJINGTECH CO LTD

Patent Information

Application Number
CN202411455724.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-18
Publication Date
2024-11-22
Estimated Expiration
2044-10-18

AI Technical Summary

Technical Problem

The existing recommendation algorithm cannot capture changes in user behavior in a timely manner, resulting in the inconsistent content of recommendations that are not consistent with user needs, thereby accelerating user churn. The existing solutions to analyze user churn behavior fail to fully consider the dynamic changes in user behavior and individual differences, which limits the Accuracy of prediction.

Method used

By monitoring the user's multi-dimensional behavior data in real time, using the time-sequence large language model to predict the probability of user churn, capturing the changing trend of user behavior, and adjusting the control strategy of the recommendation system to improve user retention.

Benefits of technology

It improves the real-time and accuracy of user churn probability prediction, improves user retention through personalized control strategies, reduces user churn rate, and enhances the sustainability of the recommendation system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119006040A_ABST
    Figure CN119006040A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of data processing, and provides a user loss probability prediction method and device and electronic equipment. The method comprises the following steps: acquiring each current refreshing moment when a target user refreshes a recommended page on a target terminal and a historical behavior record for the recommended page on the target terminal in a historical time period; analyzing the behavior characteristics of the linear behaviors occurring in the historical time period to obtain a first activeness score of the target user at the current refreshing moment; processing the behavior characteristics of the nonlinear behaviors to obtain a second activeness score of the target user in a historical time period; determining a day-level activeness sequence of the target user based on the first activeness score of each current refreshing moment and the corresponding second activeness score; and based on the time sequence large language model, predicting the day-level activeness sequence to obtain the user loss probability of the target user in the preset future time period. According to the method, the real-time performance and the accuracy of predicting the user loss probability are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method, device and electronic device for predicting user churn probability. Background Art

[0002] With the rapid development of network technology, various websites, APPs and other network products have emerged to meet the experience needs of different users.

[0003] Currently, these network products are not necessarily in a stable operation and maintenance state after they are launched. Users' interests and behaviors will change over time, and existing recommendation algorithms often cannot capture these changes in a timely manner, resulting in recommended content that does not meet users' actual needs, thereby accelerating user churn.

[0004] However, existing solutions for analyzing user churn behavior cannot effectively predict user churn behavior and generally fail to fully consider the dynamic changes in user behavior and individual differences, which limits the accuracy of predictions. Summary of the Invention

[0005] The purpose of the embodiments of the present application is to provide a method, device and electronic device for predicting user churn probability, which is used to monitor multi-dimensional user behavior data such as user activity patterns and login status in real time, and use advanced large language models to predict user churn probability, thereby more finely capturing the changing trends of user behavior and improving the real-time and accuracy of the prediction.

[0006] In a first aspect, a method for predicting user churn probability is provided, which may include: Obtaining each current refresh time of a target user refreshing a recommended page on a target terminal and a historical record of the target user's behavior on the recommended page on the target terminal within a historical time period; the historical record of behavior includes linear behaviors that change linearly with time and nonlinear behaviors that do not change linearly with time; each historical time period is a historical time period adjacent to the corresponding current refresh time; For any current refresh moment, analyze the behavioral characteristics of linear behaviors that occurred in the historical time period to obtain the target user's first activity score at the current refresh moment; Processing the behavioral characteristics of the nonlinear behavior to obtain a second activity score of the target user in a historical time period; Determining a daily activity sequence for the target user based on the first activity score and the corresponding second activity score at each current refresh time, the daily activity sequence including a correspondence between each natural day arranged in chronological order and the corresponding target activity score; Based on the time series large language model, the daily activity sequence is predicted to obtain the user churn probability of the target user in a preset future time period. The preset future time period is the time period starting from the last current refresh time among the current refresh times.

[0007] In a possible implementation, the behavior characteristics of the linear behavior include the time when the linear behavior occurs; Analyze the behavioral characteristics of linear behaviors that occurred within a historical period to obtain the target user's first activity score at the current refresh time, including: Using a preset behavior activity algorithm, the current refresh time and the time when the linear behavior occurred in the historical time period are processed to obtain the initial activity score corresponding to the linear behavior at the current refresh time; Based on the configured time aggregation weights of each linear behavior, the initial activity scores of the corresponding linear behaviors are aggregated to obtain the first activity score of the target user at the current refresh time.

[0008] In a possible implementation, the linear behavior is reading behavior; the preset behavior activity algorithm is expressed as:

[0009] in, The current refresh time when the target user refreshes the page, represents the first activity score of the i-th reading behavior at time T, They are the configuration parameters, is the time when the i-th reading behavior occurs, , n represents the number of reading behaviors in the historical period.

[0010] In a possible implementation, the behavioral characteristics of the nonlinear behavior include the target user's number of interactions, number of login days, and current refresh times; The calculation formula of the second activity score is expressed as:

[0011] in, Indicates the second activity score of the target user's j-th nonlinear behavior in the historical time period, For the configuration parameters, , It represents the number of the j-th nonlinear behavior, where j=1 represents interactive nonlinear behavior, j=2 represents login nonlinear behavior, and j=3 represents refresh nonlinear behavior.

[0012] In one possible implementation, determining the target user's daily activity sequence based on the first activity score and the corresponding second activity score at each current refresh moment includes: Using the trained activity estimation model, the first activity score and the corresponding second activity score at each current refresh time are processed to obtain a real-time activity sequence of the target user, the real-time activity sequence including a correspondence between each current refresh time and the real-time activity score; The real-time activity scores in the real-time activity sequence are aggregated and calculated according to the aggregation granularity of natural days to obtain a daily level activity sequence.

[0013] In one possible implementation, before predicting the daily activity sequence based on the time series large language model, the method further includes: According to a preset convolution algorithm, a preset time convolution kernel is used to perform a convolution operation on the target activity score in the daily activity sequence to obtain a new daily activity sequence; the new daily activity sequence includes the correspondence between each natural day arranged in chronological order and the corresponding new activity score.

[0014] In one possible implementation, based on a time series large language model, the daily activity sequence is predicted to obtain the user churn probability of the target user in a preset future time period, including: determining a scaling factor based on a non-empty target activity score in the daily activity sequence; Using the scaling factor, normalizing the non-empty target activity scores in the daily activity sequence to obtain a normalized activity sequence; Use the torch.bucketize function to map the normalized target activity scores in the normalized activity sequence into a set of predefined buckets. Each predefined bucket has the same size and its own bucket boundary value. Determining a token sequence based on the bucket boundary values ​​of the predefined buckets and the number of configured sequence special tags; The token sequence is input into the time series large language model to obtain the user churn probability of the target user in a preset future time period.

[0015] In a possible implementation, after obtaining the user churn probability of the target user within a preset future time period, the method further includes: Based on the user churn probability, the control strategies in the recall stage, rough ranking stage, fine ranking stage, and re-ranking stage of the recommendation system are adjusted respectively to obtain new control strategies for the corresponding stages; Among them, the new control strategy in the recall stage includes adjusting the recall quantity of the content that the target user is interested in; the new control strategy in the rough ranking stage includes adjusting the weight value of the target user's consumption behavior; the new control strategy in the fine ranking stage includes adjusting the weight value that represents the user value and content value of the target user; the new control strategy in the rearrangement stage includes adjusting the display order of the recommended content that the target user is interested in.

[0016] In a second aspect, a device for predicting user churn probability is provided, which may include: an acquisition unit, configured to acquire each current refresh moment of a target user refreshing a recommended page on a target terminal and a historical record of the target user's behavior on the recommended page on the target terminal within a historical time period; the historical record of the behavior includes linear behaviors that change linearly with time and nonlinear behaviors that do not change linearly with time; each historical time period is a historical time period adjacent to the corresponding current refresh moment; An analysis unit, configured to analyze, for any current refresh moment, the behavior characteristics of linear behaviors occurring within a historical time period, and obtain a first activity score of the target user at the current refresh moment; a processing unit, configured to process the behavioral characteristics of the nonlinear behavior to obtain a second activity score of the target user in a historical time period; a determining unit, configured to determine a daily activity sequence of the target user based on the first activity score and the corresponding second activity score at each current refresh time, the daily activity sequence comprising a correspondence between each natural day arranged in chronological order and the corresponding target activity score; A prediction unit is used to predict the daily activity sequence based on a time series large language model to obtain the user churn probability of the target user within a preset future time period, where the preset future time period is a time period starting from the last current refresh moment among each current refresh moment.

[0017] In a third aspect, an electronic device is provided, the electronic device including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus; Memory for storing computer programs; The processor is configured to implement any of the method steps described in the first aspect when executing a program stored in the memory.

[0018] In a fourth aspect, a computer-readable storage medium is provided, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, any of the method steps described in the first aspect is implemented.

[0019] The present application provides a method for predicting user churn probability, wherein each current refresh time at which a target user refreshes a recommended page on a target terminal and a historical record of the target user's behavior on the recommended page on the target terminal within a historical time period are obtained; the historical behavior record includes linear behaviors that change linearly with time and nonlinear behaviors that do not change linearly with time; each historical time period is a historical time period adjacent to the corresponding current refresh time; for any current refresh time, the behavioral characteristics of the linear behaviors occurring within the historical time period are analyzed to obtain a first activity score of the target user at the current refresh time; the behavioral characteristics of the nonlinear behaviors are processed to obtain a second activity score of the target user within the historical time period; based on the first activity score and the corresponding second activity score at each current refresh time, a daily activity sequence of the target user is determined, the daily activity sequence including a correspondence between each natural day arranged in chronological order and the corresponding target activity score; based on a time series large language model, the daily activity sequence is predicted to obtain the user churn probability of the target user within a preset future time period, the preset future time period being the time period starting at the last current refresh time among the current refresh times. This method improves the real-time and accuracy of predicting user churn probability. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments of the present application. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without creative work.

[0021] Figure 1 A flowchart of a method for predicting user churn probability provided in an embodiment of the present application; Figure 2 A schematic diagram of the structure of a device for predicting user churn probability provided in an embodiment of the present application; Figure 3 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0022] The following, in conjunction with the accompanying drawings, provides a clear and complete description of the technical solutions in the embodiments of this application. Obviously, the described embodiments represent only a portion of the embodiments of this application and do not constitute a complete set of embodiments. Based on the embodiments of this application, all other embodiments derived by persons of ordinary skill in the art without inventive effort are intended to fall within the scope of protection of this application. Unless otherwise defined, technical or scientific terms used in this application should have the same ordinary meanings as those understood by persons of ordinary skill in the art. The terms "first," "second," and similar expressions used in this application do not denote any order, quantity, or importance; they are merely used to distinguish between different components. Terms such as "include" or "comprising" mean that the element or object preceding the term includes the elements or objects listed after the term, and their equivalents, without excluding other elements or objects. Terms such as "connect," "couple," or "connected" are not limited to physical or mechanical connections but may include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used solely to indicate relative positional relationships. When the absolute position of the described objects changes, the relative positional relationships may also change accordingly.

[0023] User retention is a crucial metric for mobile apps, directly impacting the system's long-term effectiveness and commercial value. However, existing recommendation systems often focus on improving short-term user engagement and satisfaction, while neglecting the importance of long-term user retention. As a result, even recommendation systems that perform well in the short term can face the problem of gradual user churn, impacting the system's overall effectiveness and sustainability.

[0024] Existing technologies use a variety of different approaches to predict user churn probability, including: 1. Methods based on historical behavior analysis: This type of method mainly relies on users' historical behavior data (such as browsing and consumption records), and predicts the probability of user churn through statistical analysis or simple machine learning models.

[0025] 2. User portrait-based methods: This type of method builds user portraits to understand users' interests, preferences, social attributes, etc., and combines this information to predict the probability of user churn.

[0026] 3. Methods based on time series analysis: This type of method uses time series analysis techniques (such as ARIMA, state space model, etc.) to capture the changing patterns of user behavior over time and predict future behavior patterns.

[0027] 4. Deep learning-based methods: This type of method uses deep learning technology (such as recurrent neural networks (RNNs) and long short-term memory (LSTMs)) to automatically extract complex features of user behavior and predict the probability of user churn.

[0028] While existing technologies can predict user churn to a certain extent, they often fail to fully consider the dynamic nature of user behavior and individual differences. Therefore, there is still room for improvement in the accuracy of predicting user churn probability. Furthermore, many methods are limited by the amount of data and computing resources required, and face certain challenges in practical application.

[0029] To address the aforementioned issues, this application provides a method for predicting user churn probability. In this method, when a target user refreshes a recommendation page, the target terminal automatically collects user behavior data and calculates a user activity score to assess the target user's activity score at the current refresh moment. Subsequently, by analyzing the user's activity score data, a time-series large language model is used to predict their future churn probability. This time-series large language model combines time series analysis and natural language processing techniques to better understand how user behavior changes over time. In particular, it leverages the powerful predictive capabilities of the large language model (LLM) to achieve accurate user churn prediction.

[0030] Furthermore, based on the estimated user churn probability, personalized control strategies are applied to each stage module of the recommendation system to improve user retention. Each stage module can include a recall control system, a coarse ranking control system, a fine ranking control system, and a re-ranking control system.

[0031] The preferred embodiments of the present application are described below in conjunction with the drawings in the specification. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present application and are not used to limit the present application. In addition, the embodiments and features in the embodiments of the present application can be combined with each other if there is no conflict.

[0032] Figure 1 This is a flow chart of a method for predicting the probability of user churn provided by an embodiment of the present application. Figure 1 As shown, the method may include: Step S110: Acquire each current refresh time when the target user refreshes the recommended page on the target terminal and the target user's historical behavior record on the recommended page on the target terminal within a historical time period corresponding to the current refresh time.

[0033] In the specific implementation, after obtaining the current refresh time T of the target user refreshing the recommended page on the target terminal, for any current refresh time T, the user portrait database is called to obtain the historical behavior records of the target user corresponding to each current refresh time on the recommended page on the target terminal within the historical time period.

[0034] The historical time period is the historical time period adjacent to the current refresh time. The historical behavior records may include linear behaviors that change linearly over time (such as reading content) and non-linear behaviors that do not change linearly over time (such as refresh behaviors, user interaction behaviors, and login behaviors).

[0035] Step S120: Analyze the behavior characteristics of each linear behavior that occurred in the historical time period to obtain the first activity score of the target user at the current refresh moment of the corresponding linear behavior.

[0036] The behavior characteristics of the linear behavior may include the time when the linear behavior occurs.

[0037] In specific implementations, for any current refresh time T, a preset behavioral activity algorithm is used to process the current refresh time and the time when linear behaviors occurred within the corresponding historical time period to obtain the initial activity score corresponding to the linear behaviors at the current refresh time. Subsequently, based on the configured time aggregation weights for each linear behavior, the initial activity scores of the corresponding linear behaviors are aggregated to obtain the target user's first activity score at the current refresh time.

[0038] Specifically: First, obtain the time when each reading behavior occurred in the historical time period to obtain the historical reading behavior set. For example, assuming that the user exist Always read the content , then the user A collection of historical reading behaviors in the last month , is the reading content corresponding to the i-th reading behavior The time point of the moment.

[0039] Secondly, for any reading behavior, a preset activity algorithm is used to process the current refresh time and the time when the reading behavior occurred in the corresponding historical time period to obtain the initial activity score corresponding to the reading behavior at the current refresh time. The time when the reading behavior occurred can be understood as the time when the reading behavior started.

[0040] Then, the product of the configured time-aggregated weights for each linear behavior and the initial activity scores for the corresponding linear behaviors is added together to obtain the target user's first activity score at the current refresh time. The first activity score G represents the target user's activity score at the current refresh time after the historical time period.

[0041] In an example, when the linear behavior is reading behavior, the calculation formula for the first activity score G can be expressed as:

[0042] in, It represents the time aggregation weight of the i-th reading behavior, and its value can be set to , you can also set more complex exponential weights, or fixed values, such as , this application is not limited here.

[0043] In an example, when the linear behavior is reading behavior, the preset behavior activity algorithm is expressed as:

[0044] in, The current refresh time when the target user refreshes the page, represents the first activity score of the i-th reading behavior at time T, They are the configuration parameters, is the time when the i-th reading behavior occurs, , n represents the number of reading behaviors in the historical period.

[0045] Step S130: Process the behavioral features of the nonlinear behavior to obtain a second activity score of the target user in a historical time period.

[0046] The behavioral characteristics of nonlinear behaviors may include the number of interactions, login days, and current refresh times of the target user. Accordingly, the calculation formula for the second activity score can be expressed as:

[0047] in, Indicates the second activity score of the target user's j-th nonlinear behavior in the historical time period, For the configuration parameters, , It represents the number of the j-th nonlinear behavior, where j=1 represents interactive nonlinear behavior, j=2 represents login nonlinear behavior, and j=3 represents refresh nonlinear behavior.

[0048] Step S140: Determine the target user's daily activity sequence based on the first activity score and the corresponding second activity score at each current refresh moment.

[0049] In a specific implementation, a trained activity estimation model is used to process the first activity score G and the corresponding second activity score Z at each current refresh time to obtain the target user's real-time activity sequence. The real-time activity sequence can include the corresponding relationship between each current refresh time and the real-time activity score; the activity estimation model is a linear regression model.

[0050] Next, the real-time activity scores at each refresh time in the real-time activity sequence are aggregated and calculated at the natural day aggregation granularity to obtain a daily activity sequence. The daily activity sequence may include a chronological correspondence between each natural day and the corresponding target activity score.

[0051] In one example, for any historical time period corresponding to the current refresh time T, the corresponding second activity score Z can be obtained, including the activity score of the interactive behavior. , the activity score of login behavior and the activity score of the current refresh behavior When , the calculation formula of the real-time activity score s at the current refresh time T can be expressed as:

[0052] in, and is the weight parameter trained by the model, The value range is .

[0053] After calculating the real-time activity score s at each current refresh time, the target user's ID and the corresponding real-time activity sequence can be updated to the user activity score database in real time through the Kafka message sending mechanism. That is, the user activity database stores: user ID, each current refresh time T and the corresponding real-time activity score s. As time goes by, the user Activity score Will be continuously recorded, thus forming a real-time activity sequence This step ensures the real-time and accuracy of user activity score information, making it possible to adjust the recommendation strategy in a timely manner based on the latest activity score information, thereby improving the relevance of recommendations and user satisfaction.

[0054] Furthermore, after sorting the real-time activity scores in the obtained real-time activity sequence in chronological order, the real-time activity scores are aggregated and calculated according to the aggregation granularity of natural days (days) to obtain the target activity scores corresponding to each natural day, thereby generating a day-level activity sequence S, that is, the day-level activity sequence can include the correspondence between each natural day arranged in chronological order and the corresponding target activity score.

[0055] In one example, to calculate the target activity score for day d The calculation formula is as follows:

[0056] in, Represents a user In the Day's The real-time activity score s at the moment, Represents a user In the The number of activity scores per day.

[0057] Step S150: Based on the time series large language model, the daily activity sequence is predicted to obtain the user churn probability of the target user in a preset future time period.

[0058] Since the data format of the target activity scores in the daily activity sequence S is different from the data format of the input data of the time series large language model, the sequence needs to be tokenized, that is, the numerical sequence is converted into a data token form suitable for processing by the large language model. This conversion method includes three steps: calculating the scaling factor, scaling the context, and bucketing.

[0059] (1) Calculate the scaling factor : Determine a scaling factor based on the non-empty target activity scores in the daily activity sequence S. Specifically, obtain the cumulative sum of the absolute values ​​of all non-empty target activity scores in the daily activity sequence S, and divide the sum by the number of all non-empty target activity scores; If the calculated quotient is greater than 0, the quotient is determined as the scaling factor. ; If the calculated quotient is not greater than 0, the scaling factor Set to 1.0 to ensure the validity of subsequent calculations, the scaling factor The calculation formula is as follows:

[0060] in, is the number of all non-empty activity scores, is the kth non-empty target activity score in the daily activity sequence S.

[0061] (2) Scaling context: The scaling factor is used to normalize the non-empty target activity scores in the daily activity sequence to obtain the corresponding normalized target activity scores, thereby obtaining a normalized activity sequence; specifically, the normalized target activity score is the sum of the corresponding non-empty target activity scores and the scaling factor The quotient of the division.

[0062] (3) Bucketing: The torch.bucketize function is used to map the normalized target activity scores in the normalized activity sequence into a set of predefined buckets. Each predefined bucket has the same size and its own bucket boundary value. The token sequence is determined based on the bucket boundary value of each predefined bucket and the configured number of target sequence tokens.

[0063] Specifically, the torch.bucketize function is used to map the scaled sequence into a set of predefined buckets. The mapping process also requires consideration of the number of sequence-specific markers. Sequence-specific markers refer to distinctive markers in the input sequence of a large temporal language model, such as the start-of-sequence marker [CLS] and the end-of-sequence marker [SEP].

[0064] First, the bucket boundary value of each predefined bucket is determined (for example, after determining that the size of each bucket is approximately 0.0073, the bucket boundary value of each predefined bucket is determined), which can be determined by uniform distribution or based on the characteristics of data distribution.

[0065] Second, use the torch.bucketize function to map the scaled sequences into the corresponding predefined buckets.

[0066] Afterwards, the bucket index determined by the bucket boundary value of each predefined bucket Convert to the corresponding token and add the number of special tags in the sequence to get the user Token sequence. Bucket index The calculation formula can be expressed as:

[0067] in, is a list of bucket boundary values, is the number of special tokens in the sequence.

[0068] Finally, the token sequence is input into the time series large language model to obtain the target user's user churn probability within a preset future time period. The preset future time period is the time period starting from the last current refresh time among the current refresh times.

[0069] In some embodiments, in order to better capture the changing pattern of user activity scores, before executing step S150, a time convolution kernel may be used according to a preset convolution algorithm. , performing a convolution operation on the target activity score in the daily activity sequence. This process not only extracts features but also effectively smooths the noise in the daily activity sequence, resulting in a smoother new user activity sequence. Then, executing step S150.

[0070] Among them, the preset convolution algorithm can be expressed as: Set the convolution kernel The length is , then the temporal convolution kernel Can write , then based on the new user activity sequence , the target user is the first in the sequence The new activity score for the day is , the calculation formula can be:

[0071] in, The first in the day-level activity sequence The target activity score for the day.

[0072] In some embodiments, after obtaining the user churn probability of the target user within a preset future time period, the control strategies in the recall stage, coarse ranking stage, fine ranking stage and re-ranking stage of the recommendation system can be adjusted based on the user churn probability to obtain new control strategies for the corresponding stages; wherein, the new control strategy for the recall stage may include adjusting the recall quantity of content that the target user is interested in; the new control strategy for the coarse ranking stage may include adjusting the weight value of the target user's consumption behavior; the new control strategy for the fine ranking stage may include adjusting the weight value that characterizes the user value and content value of the target user; the new control strategy for the re-ranking stage may include adjusting the display order of recommended content that the target user is interested in.

[0073] Based on user churn probability, personalized control strategies are applied to each stage of the recommendation system to improve user retention. The corresponding adjustment systems for each stage can include: recall control system, coarse ranking control system, and fine ranking control system. Strategies for adjusting the display order of content types that target users are interested in can be applied to the re-ranking control system.

[0074] (1) Recall control system Responsible for adjusting the number of recommended resources recalled based on user preferences. By analyzing the user's historical behavior data, the user's preferred resource types are identified and the number of resources of the corresponding type is increased accordingly.

[0075] Strategy: For users with a high probability of churn, increase the number of recalls based on the resource types that the user prefers. For resource types that the user is less interested in, appropriately reduce the number of recalls.

[0076] (2) Coarse drainage control system This is used during the rough ranking phase to optimize recommended content by individually adjusting the weights of multiple metrics. By adjusting the weights of multiple metrics (CTR, likes, favorites, shares, creations, comments, and page views) for each user, the recommendation system is more likely to recommend content that users will enjoy.

[0077] Strategy: For users with a higher churn risk (i.e., a higher probability of user churn), increase the weights corresponding to the CTR, likes, favorites, and sharing target scores in the multi-goal system, and reduce the weights corresponding to the creation, comment, and personal page entry target scores, so that the system is more likely to recommend content that users like.

[0078] (3) Precision control system In the refined ranking stage, there are four comprehensive value goals: user value, creative value, content value, and commercial value. By increasing the weight of user value and content value goals and reducing the weight of commercial value and creative value goals, we can reduce the recommendation of content that users are not interested in or may cause churn.

[0079] Strategy: For users with a high churn risk (i.e., a high probability of user churn), reduce the weight of ecological goals (such as recommendation diversity and novelty) and the weight of commercial value goals (such as promotion and advertising); strengthen the value of user-preferred content to improve user satisfaction.

[0080] (4) Rearrangement control system Responsible for adjusting business strategies and commercial strategies during the final presentation of recommendation results to ensure that the recommended content can both meet user needs and reduce negative impacts.

[0081] Strategy: For users with a high probability of churn, adjust the order of recommended content, prioritizing content that the user is more interested in. Reduce the display of advertisements or other commercial content that may cause user disgust.

[0082] Through the coordinated work of the above four control systems, personalized recommendations can be effectively adjusted based on the probability of user churn, thereby improving user satisfaction and retention, reducing user churn, and bringing better business results to the App.

[0083] Alternatively, for potential churn users with a higher probability of churn, additional incentives such as limited-time discounts, membership privileges, and point rewards can be provided to enhance user stickiness.

[0084] In some embodiments, in order to further improve the accuracy of the prediction, after obtaining the user churn probability of the target user in a preset future time period, the target user's attribute information (such as basic information such as age, gender, occupation, etc.), the attribute information of the target user's friends on the social network, and the content that the target user and each friend are interested in can be obtained.

[0085] Based on the configured first cluster center and first cluster radius, the target user's attribute information and the friends' attribute information are clustered to obtain target friend information similar to the target user; and based on the configured second cluster center and second cluster radius, the target user's attention content and the attention content of each friend are clustered to obtain target attention content similar to the attention content of the target user; Then, check whether the target content exists in the content of interest corresponding to the target friend information; If so, the target user churn probability corresponding to the target friend information corresponding to the target content is obtained, and the average of the target user churn probabilities is determined as the final user churn probability of the target user in a preset future time period.

[0086] If not, the user churn probability obtained in step S150 is determined as the final user churn probability of the target user in the preset future time period.

[0087] In some embodiments, after obtaining a new content distribution strategy, a quick response mechanism can be established to promptly collect feedback information from target users regarding the new content distribution strategy, and further adjust the new content distribution strategy accordingly, so that users feel that their opinions are valued, which can effectively improve satisfaction.

[0088] Corresponding to the above method, the embodiment of the present application also provides a device for predicting the probability of user churn, such as Figure 2 As shown, the device includes: An acquisition unit 210 is configured to acquire each current refresh time of a target user refreshing a recommended page on a target terminal and a historical record of the target user's behavior on the recommended page on the target terminal within a historical time period; the historical record of the behavior includes linear behaviors that change linearly with time and non-linear behaviors that do not change linearly with time; each historical time period is a historical time period adjacent to the corresponding current refresh time; The analysis unit 220 is configured to analyze the behavior characteristics of linear behaviors occurring within a historical time period at any current refresh time, and obtain a first activity score of the target user at the current refresh time; The processing unit 230 is configured to process the behavioral characteristics of the nonlinear behavior to obtain a second activity score of the target user in a historical time period; A determining unit 240 is configured to determine a daily activity sequence for the target user based on the first activity score and the corresponding second activity score at each current refresh time, the daily activity sequence comprising a correspondence between each natural day arranged in chronological order and the corresponding target activity score; The prediction unit 250 is used to predict the daily activity sequence based on the time series large language model to obtain the user churn probability of the target user in a preset future time period. The preset future time period is the time period with the last current refresh time as the starting time among each current refresh time.

[0089] The functions of the various functional units of the user churn probability prediction device provided in the above-mentioned embodiment of the present application can be realized through the above-mentioned method steps. Therefore, the specific working process and beneficial effects of each unit in the user churn probability prediction device provided in the embodiment of the present application will not be repeated here.

[0090] The present application also provides an electronic device, such as Figure 3 As shown, it includes a processor 310 , a communication interface 320 , a memory 330 and a communication bus 340 , wherein the processor 310 , the communication interface 320 , and the memory 330 communicate with each other via the communication bus 340 .

[0091] Memory 330, for storing computer programs; The processor 310 is configured to execute the program stored in the memory 330 by performing the following steps: Obtaining each current refresh time of a target user refreshing a recommended page on a target terminal and a historical record of the target user's behavior on the recommended page on the target terminal within a historical time period; the historical record of behavior includes linear behaviors that change linearly with time and nonlinear behaviors that do not change linearly with time; each historical time period is a historical time period adjacent to the corresponding current refresh time; For any current refresh moment, analyze the behavioral characteristics of linear behaviors that occurred in the historical time period to obtain the target user's first activity score at the current refresh moment; Processing the behavioral characteristics of the nonlinear behavior to obtain a second activity score of the target user in a historical time period; Determining a daily activity sequence for the target user based on the first activity score and the corresponding second activity score at each current refresh time, the daily activity sequence including a correspondence between each natural day arranged in chronological order and the corresponding target activity score; Based on the time series large language model, the daily activity sequence is predicted to obtain the user churn probability of the target user in a preset future time period. The preset future time period is the time period starting from the last current refresh time among the current refresh times.

[0092] The communication bus mentioned above can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. This communication bus can be divided into address buses, data buses, and control buses. For ease of illustration, the figure uses only one thick line, but this does not mean that there is only one bus or only one type of bus.

[0093] The communication interface is used for communication between the above electronic device and other devices.

[0094] The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage. Alternatively, the memory may be at least one storage device located away from the processor.

[0095] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components.

[0096] The implementation methods and beneficial effects of the various components of the electronic device in the above embodiments to solve the problems can be found in Figure 1 The various steps in the embodiment shown are implemented, therefore, the specific working process and beneficial effects of the electronic device provided by the embodiment of the present application are not repeated here.

[0097] In another embodiment provided by the present application, a computer-readable storage medium is further provided, in which instructions are stored. When the computer-readable storage medium is executed on a computer, the computer executes the method for predicting user churn probability described in any of the above embodiments.

[0098] In another embodiment provided by the present application, a computer program product including instructions is also provided, which, when executed on a computer, enables the computer to execute the method for predicting user churn probability described in any one of the above embodiments.

[0099] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the embodiments of the present application may take the form of entirely hardware embodiments, entirely software embodiments, or embodiments combining software and hardware. Furthermore, the embodiments of the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0100] The embodiments of the present application are described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0101] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0102] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0103] Although preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they become aware of the basic creative concepts. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present invention.

[0104] Obviously, those skilled in the art can make various changes and modifications to the embodiments of the present application without departing from the spirit and scope of the embodiments of the present application. Thus, if these modifications and variations of the embodiments of the present application fall within the scope of the claims and their equivalents, the embodiments of the present application are also intended to include these modifications and variations.

Claims

1. A method for predicting user churn probability, characterized in that: The method comprises: Obtaining each current refresh time of the target user refreshing the recommended page on the target terminal and the historical behavior record of the target user on the recommended page on the target terminal in the historical time period; the historical behavior record includes linear behaviors that change linearly with time and non-linear behaviors that do not change linearly with time; each historical time period is a historical time period adjacent to the corresponding current refresh time; For any current refresh time, analyze the behavior characteristics of linear behaviors that occurred in the historical time period to obtain the first activity score of the target user at the current refresh time; Processing the behavior characteristics of the nonlinear behavior to obtain a second activity score of the target user in a historical time period; Determine a daily activity sequence of the target user based on the first activity score and the corresponding second activity score at each current refresh time, wherein the daily activity sequence includes a correspondence between each natural day arranged in chronological order and the corresponding target activity score; Based on the time series large language model, the daily level activity sequence is predicted to obtain the user churn probability of the target user in a preset future time period, and the preset future time period is a time period starting from the last current refresh time among the current refresh times.

2. The method according to claim 1, characterized in that The behavior characteristics of the linear behavior include the time when the linear behavior occurs; Analyze the behavior characteristics of linear behaviors that occurred in the historical time period to obtain the first activity score of the target user at the current refresh time, including: Using a preset behavior activity algorithm, the current refresh time and the behavior occurrence time of the linear behavior in the historical time period are processed to obtain an initial activity score corresponding to the linear behavior at the current refresh time; Based on the configured time aggregation weights of each linear behavior, the initial activity scores of the corresponding linear behaviors are aggregated to obtain the first activity score of the target user at the current refresh time.

3. The method according to claim 2, characterized in that The linear behavior is reading behavior; the preset behavior activity algorithm is expressed as: in, The current refresh time when the target user refreshes the page. represents the first activity score of the i-th reading behavior at time T, They are the configuration parameters, is the time when the i-th reading behavior occurs, , n represents the number of reading behaviors in the historical period.

4. The method according to claim 1, characterized in that The behavior characteristics of the non-linear behavior include the number of interactions, number of login days and current refresh times of the target user; The calculation formula of the second activity score is expressed as: in, Indicates the second activity score of the target user's j-th nonlinear behavior in the historical time period, For the configuration parameters, , It represents the number of the j-th nonlinear behavior, j=1 represents interactive nonlinear behavior, j=2 represents logged-in nonlinear behavior, and j=3 represents refresh nonlinear behavior.

5. The method according to claim 1, characterized in that Determining a daily level activity sequence of the target user based on the first activity score and the corresponding second activity score at each current refresh time includes: Using the trained activity estimation model, the first activity score and the corresponding second activity score at each current refresh time are processed to obtain a real-time activity sequence of the target user, wherein the real-time activity sequence includes a corresponding relationship between each current refresh time and the real-time activity score; According to the aggregation granularity of natural days, the real-time activity scores in the real-time activity sequence are aggregated and calculated to obtain a day-level activity sequence.

6. The method according to claim 1, characterized in that Before predicting the day-level activity sequence based on the time series large language model, the method further includes: According to a preset convolution algorithm, a preset time convolution kernel is used to perform a convolution operation on the target activity score in the day-level activity sequence to obtain a new day-level activity sequence; the new day-level activity sequence includes the correspondence between each natural day arranged in chronological order and the corresponding new activity score.

7. The method according to claim 1, characterized in that Based on the time series large language model, the daily level activity sequence is predicted to obtain the user churn probability of the target user in a preset future time period, including: Determining a scaling factor based on a non-empty target activity score in the day-level activity sequence; Using the scaling factor, normalizing the non-empty target activity scores in the daily activity sequence to obtain a normalized activity sequence; Use torch.bucketize function to map the normalized target activity scores in the normalized activity sequence into a set of predefined buckets; each predefined bucket has the same size and has its own bucket boundary value; Determine a token sequence based on the bucket boundary values ​​of the predefined buckets and the number of configured sequence special tags; The token sequence is input into the time series large language model to obtain the user churn probability of the target user in a preset future time period.

8. The method according to claim 1, characterized in that After obtaining the user churn probability of the target user in a preset future time period, the method further includes: Based on the user churn probability, the control strategies in the recall stage, the rough ranking stage, the fine ranking stage and the re-ranking stage of the recommendation system are adjusted respectively to obtain new control strategies for the corresponding stages; Among them, the new control strategy in the recall stage includes adjusting the recall quantity of the content that the target user is interested in; the new control strategy in the rough ranking stage includes adjusting the weight value of the target user's consumption behavior; the new control strategy in the fine ranking stage includes adjusting the weight value that characterizes the user value and content value of the target user; the new control strategy in the rearrangement stage includes adjusting the display order of the recommended content that the target user is interested in.

9. A device for predicting user churn probability, characterized in that: The device comprises: An acquisition unit, used to acquire each current refresh time when a target user refreshes a recommended page on a target terminal and a historical behavior record of the target user on the recommended page on the target terminal in a historical time period; the historical behavior record includes a linear behavior that changes linearly with time and a non-linear behavior that does not change linearly with time; each historical time period is a historical time period adjacent to the corresponding current refresh time; An analysis unit, configured to analyze the behavior characteristics of linear behaviors occurring in a historical time period at any current refresh time, and obtain a first activity score of the target user at the current refresh time; A processing unit, configured to process the behavior characteristics of the nonlinear behavior to obtain a second activity score of the target user in a historical time period; a determining unit, configured to determine a day-level activity sequence of the target user based on the first activity score and the corresponding second activity score at each current refresh time, wherein the day-level activity sequence includes a correspondence between each natural day arranged in chronological order and the corresponding target activity score; A prediction unit is used to predict the daily level activity sequence based on a time series large language model to obtain the user churn probability of the target user in a preset future time period, where the preset future time period is a time period starting from the last current refresh time among the current refresh times.

10. An electronic device, characterized in that: The electronic device comprises a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; Memory, used to store computer programs; A processor, for implementing any of the methods described in claims 1-8 when executing a program stored in a memory.

Citation Information

Patent Citations

  • User information acquisition method and device, electronic equipment and storage medium

    CN111401433A

  • Method and device for predicting user loss, equipment and medium

    CN114897557A

  • Marketing platform risk early warning management system for Internet promotion

    CN118674278A

Cited By

  • Loss prediction method and device, electronic equipment, computer readable storage medium and computer program product

    CN119941317A