Group state tracking method, device, equipment, computer medium and product

Abstract

The invention discloses a group state tracking method and device, equipment, a computer medium and a product, which are applied to the field of data processing, and the method comprises the following steps: obtaining user data from a plurality of data sources, and carrying out tagging processing on the user data to obtain a multi-dimensional tag data set; in response to a request message sent by a requester and used for querying a target group, detecting whether the target group exists or not; under the condition that the target group does not exist, obtaining at least one piece of label data meeting the requirement of the target group, and combining the label data to obtain the target group and identification information of the target group; determining a tracking condition of the target group based on the type of the target group; the tracking condition comprises at least one tracking index and a corresponding tracking frequency; under the condition that the tracking frequency is met, obtaining index data of the target group based on the identification information; and generating trend data for representing a tracking index change trend of the target group based on the index data. The method improves the flexibility and adaptability of group tracking.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of data processing, and particularly relates to a group state tracking method and device, equipment, computer storage medium and computer program product. BACKGROUND

[0002] With the development of science and technology, how to realize stable tracking of a user group is of great significance.

[0003] A conventional group tracking method usually adopts a manner of predefining fixed group conditions and tracking indicators to perform state tracking. This method needs to fix the tracking conditions and tracking indicators of the user group during program development, and data is imported into a relational database through an ETL (Extract, Transform, Load) process for precalculation. However, the above group tracking manner has certain limitations, and can only track certain indicators of certain user groups, and cannot support business personnel to dynamically create or adjust user groups and tracking indicators according to real-time needs, and has low flexibility and poor adaptability.

[0004] Therefore, how to provide a group tracking method to realize the flexibility and adaptability of user group tracking is a problem to be solved at present. SUMMARY

[0005] Embodiments of the present application provide a group state tracking method, device, equipment, computer storage medium and computer program product, which can improve the flexibility and adaptability of group tracking.

[0006] In a first aspect, embodiments of the present application provide a group state tracking method, which comprises: Obtaining user data from a plurality of data sources, and performing labelization processing on the user data to obtain a multi-dimensional label data set; In response to a request message for querying a target group sent by a requestor, detecting whether the target group exists; In the case where the target group does not exist, obtaining at least one label data satisfying the demand of the target group from the multi-dimensional label data set, combining the at least one label data to obtain the target group and identification information of the target group; Determining a tracking condition of the target group based on a type of the target group; the tracking condition comprises at least one tracking indicator and a corresponding tracking frequency; In the case where the tracking frequency is satisfied, obtaining indicator data of the target group based on the identification information; Generating trend data for representing a tracking indicator change trend of the target group based on the indicator data.

[0007] In some possible implementation manners, the tracking frequency is determined in the following manner: obtaining a value level of the target group; determining an initial tracking frequency of the target group based on the value level; obtaining a historical index data fluctuation condition of the target group; determining a coefficient of variation of the target group based on the historical index data fluctuation condition; performing an increase processing on the initial tracking frequency in a case where the coefficient of variation is higher than a first threshold value; performing a decrease processing on the initial tracking frequency in a case where the coefficient of variation is not higher than a second threshold value; maintaining the initial tracking frequency unchanged in a case where the coefficient of variation is between the first threshold value and the second threshold value.

[0008] In some possible implementation manners, the multi-dimensional label data set includes at least one of the following: user basic information, asset liability information, product holding information, transaction behavior, and channel characteristics. The obtaining, from the multi-dimensional label data set, at least one label data satisfying the demand of the target group includes: constructing a data wide table used to represent an association relationship between label data; obtaining, based on the data wide table, at least one label data satisfying the demand of the target group.

[0009] In some possible implementation manners, the method further includes: determining a tracking time based on the tracking frequency; The obtaining, based on the identification information, of the index data of the target group includes: constructing a time series data partition mapping table used to represent an association relationship between identification information-time-tracking index-index data; obtaining, based on the tracking time, a target time series data partition matching the tracking time from the time series data partition mapping table; obtaining, based on the identification information, the index data of the target group from the target time series data partition.

[0010] In some possible implementation manners, after the obtaining of the index data of the target group, the method further includes: constructing structured data including identification information, tracking time, tracking index, and index data; storing the structured data in a preset relational database.

[0011] In some possible implementation manners, the generating, based on the index data, of the trend data for representing a change trend of the tracking index of the target group comprises: sequentially sorting the index data according to time to generate a time series of the index data; rendering, by using a chart rendering engine, the time series of the index data into a statistical chart; the statistical chart comprises at least one of a line chart, a column chart or an area chart.

[0012] In some possible implementation manners, after the generating, based on the index data, of the trend data for representing a change trend of the tracking index of the target group, the method further comprises: inputting the trend data into a preset decision model to obtain a decision result; performing a preset target operation on the target group based on the decision result; the target operation comprises at least one of adjusting a recommended resource of the target group and modifying a value level of the target group.

[0013] In a second aspect, an embodiment of the present application provides a group state tracking device, which comprises: an acquisition module, configured to acquire user data from a plurality of data sources, and perform label processing on the user data to obtain a multi-dimensional label data set; a detection module, configured to detect, in response to a request message for querying a target group sent by a requester, whether the target group exists; The acquisition module is further configured to, in a case where the target group does not exist, acquire at least one label data satisfying a requirement of the target group from the multi-dimensional label data set, and combine the at least one label data to obtain the target group and identification information of the target group; a determination module, configured to determine a tracking condition of the target group based on a type of the target group; the tracking condition comprises at least one tracking index and a corresponding tracking frequency; The acquisition module is further configured to, in a case where the tracking frequency is satisfied, acquire index data of the target group based on the identification information; a generation module, configured to generate, based on the index data, trend data for representing a change trend of a tracking index of the target group.

[0014] In a third aspect, an embodiment of the present application provides a group state tracking device, which comprises: a processor and a memory storing computer program instructions; the processor implements the group state tracking method of any one of the preceding embodiments when executing the computer program instructions.

[0015] In a fourth aspect, an embodiment of the present application provides a computer storage medium, and the computer storage medium stores computer program instructions. When the computer program instructions are executed by a processor, the group state tracking method of any one of the preceding aspects is implemented.

[0016] In a fifth aspect, an embodiment of the present application provides a computer program product, and instructions in the computer program product are executed by a processor of an electronic device, so that the electronic device can execute the group state tracking method of any one of the preceding aspects.

[0017] The group state tracking method, device, equipment, computer storage medium and computer program product provided by the embodiments of the present application, after responding to the request message sent by the requester for querying the target group, first detect whether the target group exists. In the case that the target group does not exist, at least one label data meeting the demand of the target group is obtained from the multi-dimensional label data set, and the target group is obtained by combining the at least one label data. Then, based on the type of the target group, the tracking condition of the target group is determined, and when the tracking condition is reached, the index data of the target group is obtained based on the identification information. Finally, based on the index data, the trend data for representing the tracking index change trend of the target group is generated. This method fundamentally subverts the rigid mode of the traditional pre-solidified group definition. By responding to the business request and automatically detecting whether the target group exists, when the group does not exist, the labels meeting the business demand can be filtered and combined from the multi-dimensional label data set in time, and a new target group is dynamically generated. Therefore, the business operation personnel do not need to rely on the technical team to redevelop the data model or the ETL (Extract, Transform, Load) process, and can complete the autonomous definition and creation of the group within minutes or even seconds according to the demand, which significantly improves the flexibility and adaptability of group creation. At the same time, the personalized tracking index and tracking frequency are determined according to the group type, which optimizes the resource allocation and is conducive to further improving the flexibility and effectiveness of group tracking. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments of the present application. For those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0019] Figure 1 is a flowchart of the group state tracking method provided by an embodiment of the present application; Figure 2 is a flowchart of the tracking frequency determination method provided by an embodiment of the present application; Figure 3is a flowchart of S170 provided by an embodiment of the present application; Figure 4 is a structural diagram of a group state tracking device provided by another embodiment of the present application; Figure 5 is a structural diagram of an electronic device provided by yet another embodiment of the present application. DETAILED DESCRIPTION

[0020] The features and exemplary embodiments of various aspects of the present application will be described in detail below with reference to the drawings. To make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain the present application, but not to limit the present application. The present application can be implemented without some of these specific details by those skilled in the art. The following description of the embodiments is only to provide a better understanding of the present application by showing examples of the present application.

[0021] It should be noted that, in this document, relational terms such as first and second and the like can be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without more limitations, an element defined by the phrase "comprising a... " does not exclude the existence of additional identical elements in the process, method, article, or apparatus that includes the element.

[0022] It should be noted that the acquisition, storage, use and processing of data in the embodiments of the present application all comply with the relevant provisions of national laws and regulations.

[0023] It should be noted that in the embodiments of the present application, some industry existing solutions, components, models, etc. may be mentioned, which should be considered as exemplary, and the purpose is only to illustrate the feasibility of the implementation of the technical solutions of the present application, but it does not mean that the applicant has or will necessarily use the solution.

[0024] Traditional group tracking methods usually track the state in a way of pre-defining fixed group conditions and tracking indicators. This method needs to fix the tracking conditions and tracking indicators of the user group at the program development time, and the data is imported into the relational database through the ETL (Extract, Transform, Load) process for pre-computation. However, the above group tracking method has certain limitations, and can only track some indicators of some fixed user groups, and cannot support business personnel to dynamically create or adjust user groups and tracking indicators according to real-time needs, and has low flexibility and poor adaptability.

[0025] In order to solve the problems in the prior art, the embodiments of the present application provide a group state tracking method, device, equipment and computer storage medium.

[0026] Firstly, the group state tracking method provided by the embodiments of the present application is introduced.

[0027] Figure 1 The flowchart of the group state tracking method provided by an embodiment of the present application is shown. As shown in Figure 1 The method can include the following steps: S110, obtaining user data from a plurality of data sources, and performing labelization processing on the user data to obtain a multi-dimensional label data set.

[0028] Exemplarily, the original user data can be obtained from a plurality of data sources such as internal user information library, transaction system, financial product library and external channel behavior log through data interface or log collection tool in a timing or real-time manner.

[0029] As an example, when performing labelization processing on the user data, a pre-defined label rule engine (such as processing logic based on SQL or rule configuration file) can be used to perform labelization processing on the original data to obtain multi-dimensional label data, and finally form a unified multi-dimensional label data set.

[0030] Exemplarily, the multi-dimensional label data set includes at least one of the following: user basic information, asset liability information, product holding information, transaction behavior and channel characteristics. For example, "age interval: 30-40 years old", "asset level: high net worth", "holding product: fund A", "monthly transaction frequency: >10 times", "main channel: mobile terminal", etc. The multi-dimensional label data set can be stored in a distributed data warehouse (such as Hive or ClickHouse).

[0031] S120, in response to the request message for querying the target group sent by the requestor, detecting whether the target group exists.

[0032] In the absence of the target group, S130 is performed; in the presence of the target group, S140 is performed.

[0033] Exemplarily, taking the target group "young high net worth users who frequently transact on mobile terminals recently but do not hold any wealth management products" as an example, the requestor (operation personnel, business personnel) can input a query request through a front-end Web interface. In response to the query request of the requestor, first, it can be detected in a relational database whether the target group specified in the request exists, and the relational database pre-stores multiple types of groups. If not, the target group dynamic creation process is entered, and after the target group is created, the tracking conditions are determined according to the group type. If it exists, the target group can be directly obtained, and the tracking conditions are determined according to the group type.

[0034] S130, from the multi-dimensional label data set, at least one label data satisfying the demand of the target group is obtained, at least one label data is combined, and the target group and the identification information of the target group are obtained.

[0035] Exemplarily, in the case where the target group does not exist in the relational database, the request semantics can be parsed and converted into a query condition for the multi-dimensional label data set: channel feature ='mobile terminal' AND transaction behavior = 'high frequency' AND product holding = 'no wealth management product' AND asset level = 'high net worth' AND age interval = 'young'.

[0036] Based on the query condition, a list of user IDs satisfying all conditions is filtered out from the multi-dimensional label data set, the list is defined as a new target group, and a globally unique group ID (such as Group001) is generated for it. At the same time, the group definition (i.e., the above label combination condition) is saved to the relational database.

[0037] In order to improve the query efficiency, in some embodiments, S130 can include: A data wide table for representing the association relationship between label data is constructed; Based on the data wide table, at least one label data satisfying the demand of the target group is obtained.

[0038] It should be noted that the original label data is relatively scattered, and the key fields can be associated by constructing a data wide table. First, a plurality of label data is obtained, for example: user basic information (user ID, age, city), product holding table (user ID, product ID, holding amount), transaction behavior table (user ID, transaction time, transaction type, amount). The key fields of the above-mentioned multiple tables are pre-connected (joined) together with the user ID as the primary key to form a new data wide table containing all related fields: {user ID: "U001", age: "35", city: XX, holding product: "finance A | fund B", transaction frequency in the last 30 days: 15, the latest transaction channel: "APP"}.

[0039] When a specific group creation request is received, for example, when an operator needs to create a group of "users in XX area, holding financial products, and having more than 10 times of APP transactions in the last 30 days and being over 35 years old", the above query condition can be converted into a query condition for the data wide table, and only a single wide table needs to be scanned and filtered without multiple table association (JOIN), so that all user ID lists meeting the conditions can be quickly filtered from the wide table, and this list is the label data constituting the target group.

[0040] By constructing a data wide table integrating multi-dimensional labels in the data preprocessing stage, the complex and time-consuming multi-table association (JOIN) calculation required during online query can be completed in the offline stage. When a business personnel dynamically creates a group, only a pre-connected wide table needs to be efficiently scanned and filtered, and a precise user list can be returned instantly (usually within sub-seconds). This fundamentally solves the performance problem of dynamically combining multi-dimensional conditions for real-time grouping under massive data, and greatly improves the query efficiency.

[0041] S140, determining a tracking condition of the target group based on a type of the target group.

[0042] Illustratively, according to the label features of the target group, the type of the target group can be determined, for example, the label features of the target group are high net worth and young, and the type of the target group is high net worth user. The tracking condition can be determined according to the type of the newly created or existing target group. The tracking condition includes at least one tracking index and a corresponding tracking frequency. The tracking index can be total assets, transaction frequency, etc., and the tracking frequency can be tracking by time, by day, by week, by month or custom setting tracking period, or can be intelligently set according to the value level of the group. The tracking frequency is not fixed, but can be adjusted adaptively by analyzing the value level of the group and the fluctuation of historical index data (such as calculating the coefficient of variation), so as to realize the optimal allocation of resources.

[0043] It should be noted that different types of group tracking indicators are also different. For the high net worth group, the tracking indicators can be "daily asset balance" and "monthly total transaction amount". For the potential loss group, the indicators can be "login activity" and "product holding quantity change rate". At the same time, the tracking frequency corresponding to different types of groups is also different. For the high net worth group, the tracking frequency can be relatively frequent, for example, it can be tracked daily. The group identification information and the corresponding tracking frequency and tracking indicators can be saved in a relational database.

[0044] As an example, as shown in Figure 2 The tracking frequency can be determined in the following manner: S210, obtaining the value level of the target group.

[0045] Exemplarily, the value level is a comprehensive and quantitative business priority label, which is used to systematically classify and sort the importance, potential contribution and strategic significance of a group in the entire business system.

[0046] As an example, the value level of the target group can be determined according to the label data of the target group and a preset rule base. The rule base pre-stores the value levels corresponding to different label data. The value levels from high to low can be S level, A level, B level, C level and D level. For example, rule 1: IF (user level IN ['diamond VIP', 'platinum VIP']) AND (total consumption in the past year > 1,000,000) THEN value level = S level; rule 2: IF (product type held IN ['private bank', 'high-end insurance']) THEN value level = A level; rule 3: IF (user source = partner recommendation) AND (registration time < 30 days) THEN value level = B level (high potential new customers); rule 4: IF (the last interaction > 180 days) THEN value level = D level (silent / loss risk); default rule: the group that fails to match any of the above rules, value level = C level (ordinary group).

[0047] As another example, the value level of the target group can be calculated according to the label data of the target group by a pre-trained machine learning model (classification model or regression model). The model predicts the value score or level. For example, a user lifetime value prediction model is used to predict the total profit that the group can bring to the enterprise in the future period of time, and the level is divided according to the profit interval. A loss risk prediction model is used to predict the loss probability of the group. Important groups with high loss risk can be temporarily promoted to a higher level for focused maintenance.

[0048] It should be noted that there are many factors affecting the value level, including the asset size of the group, the product holding situation, the behavior activity and loyalty, and the loss risk, etc.

[0049] S220, determine an initial tracking frequency of the target group based on the value level.

[0050] After determining the value level of the target group, different initial tracking frequencies are configured for groups of different levels.

[0051] Exemplarily, a mapping table between value levels and initial tracking frequencies can be constructed in advance, for example: S level (top level): frequency = per hour / real-time; A level (high value): frequency = per day; B level (ordinary): frequency = per week; C level (low value): frequency = per month. Based on the mapping table and the value level of the target group, the initial tracking frequency of the target group can be determined.

[0052] S230, obtain the historical index data fluctuation of the target group.

[0053] It should be noted that when determining the tracking frequency, the stability of the target group also needs to be considered. Exemplarily, the stability of the target group can be determined according to the historical index data fluctuation of the target group.

[0054] Exemplarily, the historical index data fluctuation of the target group can be determined according to the historical value sequence of the tracking index (such as "daily asset" and "transaction frequency") of the group in the past period of time (for example, the past 30 days).

[0055] S240, determine the coefficient of variation of the target group based on the historical index data fluctuation.

[0056] Exemplarily, the coefficient of variation (CV) is the ratio of the standard deviation to the average value (CV = standard deviation / average value). The coefficient of variation is a relative volatility indicator, which overcomes the problem that the absolute value of the index cannot be compared horizontally when only the standard deviation is used. For example, a group with a daily transaction amount of 1 million, a standard deviation of 50,000 (CV = 5%), has much lower volatility than a group with a daily transaction amount of 10,000, a standard deviation of 10,000 (CV = 100%). The coefficient of variation represents the relative fluctuation degree of the historical behavior of the group (CV value).

[0057] S250, determine whether the coefficient of variation is higher than a first threshold or lower than a second threshold, execute S260 if the coefficient of variation is higher than the first threshold, execute S270 if the coefficient of variation is not higher than the second threshold, and execute S280 if the coefficient of variation is between the first threshold and the second threshold.

[0058] S260, perform an increase processing on the initial tracking frequency.

[0059] S270, performing a frequency reduction process on the initial tracking frequency.

[0060] S280, keeping the initial tracking frequency unchanged.

[0061] It should be noted that when dynamic adjustment (adaptive optimization) is based on fluctuation, the initial frequency can be corrected according to the comparison of the CV value and the preset threshold. Usually, there are two thresholds, the first threshold H (high fluctuation threshold, such as CV>30%) and the second threshold L (low fluctuation threshold, such as CV<10%). When the coefficient of variation is greater than the first threshold H, it indicates that the historical behavior of the group is very unstable and changes dramatically. In order to capture important changes in time, perform a frequency increase process on the initial tracking frequency. For example, temporarily adjust the frequency of the A-level group from "daily" to "hourly".

[0062] When the coefficient of variation is less than or equal to the second threshold L: it indicates that the historical behavior of the group is very stable, almost a straight line. Too frequent tracking is not meaningful, and the initial tracking frequency is reduced. For example, adjust the frequency of the B-level group from "weekly" to "monthly".

[0063] When the second threshold L is less than the coefficient of variation and the first threshold H, it indicates that the fluctuation is within the normal expected range, and the initial tracking frequency is kept unchanged.

[0064] The value level of the group establishes the baseline of resource input, and ensures that the high-value group obtains the necessary attention density. By introducing objective data fluctuation analysis based on the coefficient of variation, the baseline frequency is dynamically calibrated, which can automatically identify and allocate higher tracking frequency to high-value and high-fluctuation groups to capture key changes; at the same time, reduce the tracking frequency for low-fluctuation or low-value groups to avoid unnecessary consumption of computing resources. Thus, under the constraint of limited resources, the global optimization of tracking input and business benefits is realized.

[0065] S150, determining whether the tracking frequency is met, in the case that the tracking frequency is not met, performing S160; in the case that the tracking frequency is met, performing S170.

[0066] S160, without obtaining the index data of the target group.

[0067] Exemplarily, based on the tracking frequency, the tracking time can be determined. When the tracking frequency is daily tracking, the index data can be queried and obtained at a preset time (such as 2 a.m.) every day, and the index data does not need to be obtained when the time is not reached, and the group can be skipped to query the next group.

[0068] It should be noted that when the tracking time is reached, the tracking index can be read, the tracking index forms a query statement according to the group identification information, a query request is initiated, and then data files related to the group tracking index are obtained, the data files can be split according to the dimensions of {date, group, index 1}, {date, group, index 2}, {date, group, index n}, and written back to a relational database.

[0069] S170, based on the identification information, obtaining the index data of the target group.

[0070] When the tracking time is reached, the index data of the target group can be obtained from the distributed database based on the identification information of the target group, such as group ID (e.g., Group001). The index data is the data value corresponding to the tracking index, for example, when the tracking index is the daily average asset balance, the corresponding index data is the specific value of the daily average asset balance.

[0071] Exemplarily, based on the tracking frequency, the tracking time can be determined, for example, when the tracking frequency is daily, as a more specific example, the tracking time can be 2 a.m. every day; when the tracking frequency is weekly, the tracking time can be 2 a.m. every Monday.

[0072] In some embodiments, as shown in Figure 3 S170 can include: S1701, constructing a time series data partition mapping table for representing the association relationship between identification information-time-tracking index-index data.

[0073] Exemplarily, the time series data partition is strictly partitioned according to the date field, and the group ID is indexed, so that the data block can be quickly located during query, and partition pruning can be realized.

[0074] The time series data partition mapping table is partitioned according to the time field. For example, partitioning by day, then all groups, all index data of 2024-01-01 will be physically stored together to form an independent data block (partition); the data of 2024-01-02 is another partition. Thus, the disorganized index data is pre-organized into a plurality of data blocks in chronological order, with clear structure and convenient quick positioning.

[0075] S1702, based on the tracking time, obtaining the target time series data partition matched with the tracking time from the time series data partition mapping table.

[0076] Upon receiving the query request, the request contains the tracking time (e.g., the user wants to view the data on 2024-01-15). After the database engine (query optimizer) receives the query, it does not scan all the data in the entire table. Instead, it first looks at the time 2024-01-15 in the query condition, and then directly goes to the "mapping table" to find the specific data partition named p20250115 (or similar naming rules). In this way, the query range is instantly reduced from "all table data" (which may contain years of data, TB level) to "single day partition" data (which may only be GB or MB level).

[0077] S1703, based on the identification information, obtaining the target group's index data from the target time series data partition.

[0078] After determining the target time series data partition (i.e., all data blocks for 2024-01-15), the group identification information (such as group_id='VIP_001') in the query condition is filtered. Since the data has been aggregated by day, and an index is usually established for group_id within the partition, the filtering becomes very efficient. The database engine can quickly find all records belonging to the VIP_001 group in the data for that day. Finally, the value of one or more indicators of the group on 2024-01-15 is obtained.

[0079] By constructing a four-dimensional association mapping table of "identification information-time-tracking indicator-index data" time series data partition, the traditional full table scanning query is transformed into direct positioning access based on partition. When the index data of a specific group at a specific time point is needed, the corresponding physical data partition is first accurately located according to the tracking time, and then the target data is quickly retrieved within the partition through the group identification. The data query range is rapidly converged from TB / PB level full data to GB / MB level or even smaller target partition, achieving exponential improvement in query performance, and fundamentally solving the performance problem of real-time analysis of massive group tracking data.

[0080] In some embodiments, after obtaining the target group's index data, a structured data including identification information, tracking time, tracking indicator, and index data can also be constructed; the structured data is stored in a preset relational database.

[0081] Exemplarily, after obtaining the target group's index data, the multi-dimensional information generated by this tracking task can be mapped to standard structured data: (group ID, time, indicator, indicator value). The structured data is persistently stored in a relational database, ensuring that the data will not be lost and can be efficiently and stably accessed by subsequent query, analysis, and summary operations.

[0082] By encapsulating the indicator data obtained by each tracking calculation, the corresponding group identifier, timestamp, and indicator name into standardized structured data, and persistently storing them into a relational database, the "data assetization" of massive, multi-source tracking results is realized, which converts the originally scattered and temporary calculation results into data assets with unified format, clear meaning, and long-term accumulation and reuse, laying a solid foundation for subsequent historical trend deep analysis, cross-group comparison, and enterprise-level data warehouse construction.

[0083] S180、Based on the indicator data, trend data representing the tracking indicator change trend of the target group is generated.

[0084] For example, the values of one or more indicators obtained at different time points can be sorted by time to obtain the indicator change trend.

[0085] As an example, when generating trend data representing the tracking indicator change trend of the target group based on the indicator data, the indicator data can be sorted in chronological order to generate a time series data sequence; using a chart rendering engine, the time series data sequence is rendered into a statistical chart; the statistical chart includes at least one of a line chart, a column chart, or an area chart.

[0086] For example, based on the indicator data and the time when the indicator data is obtained, data points containing time-value can be constructed, and these data points can be sorted in chronological order (ascending order) to obtain a time series data sequence. The time series data sequence is an ordered array that reflects how the indicator data changes over time. For example: [[2024-01-01, 100], [2024-01-02, 120], [2024-01-03, 115], … ].

[0087] A chart rendering engine refers to a software library or service specifically used to generate charts, such as ECharts, AntVG2 for the front end, or Matplotlib for the back end, or chart generation API for the server. The chart rendering engine is used to convert numerical sequences into graphics, and according to the data sequence and the preset chart configuration (such as color, axis label), it calculates and draws the corresponding graphical elements (statistical chart) in memory or canvas.

[0088] Exemplarily, the statistical chart can include at least one of a line chart, a column chart, or an area chart. The line chart is most commonly used to show the trend and trend of continuous changes of indicators over time. By connecting the data points, the rising, falling, fluctuation or stable state is clearly displayed. The column chart is suitable for comparing the absolute size of the indicator values at different time points or time periods. It can intuitively display peaks and valleys and differences. The area chart fills the color on the basis of the line chart, emphasizes the cumulative effect over time or the relationship between the part and the whole, and visually highlights the changes in quantity. After obtaining the statistical chart, the trend data of the target group can be displayed.

[0089] By automatically rendering the time-ordered indicator data sequence into intuitive statistical charts such as line charts, column charts, etc., the qualitative change from abstract numbers to visual graphics is completed, so that business operation personnel do not need to have professional data analysis skills, and can intuitively grasp whether the group state is improved, deteriorated or stable through the graphic trend in seconds, which is helpful for quickly finding problems and timely intervention.

[0090] In order to further improve the operation efficiency and avoid lag, in some embodiments, after generating the trend data for representing the tracking indicator change trend of the target group based on the indicator data, the trend data can be input into a preset decision model to obtain a decision result; based on the decision result, a preset target operation is performed on the target group; the target operation includes at least one of the following: adjusting the recommended resources of the target group, modifying the value level of the target group.

[0091] It should be noted that the trend data is not only the original numerical value, but also the structured or vectorized data containing information such as trend direction (up / down), change rate, fluctuation mode. The decision model can be a trained prediction or classification model. For example, a classification model for predicting future churn probability based on historical trends, or an anomaly detection model for identifying abnormal patterns.

[0092] After obtaining the trend data, the trend data is input into the decision model, and the decision model obtains a decision result by interpreting the trend data, for example, abnormal decrease in activity of a high-value group. Based on the decision result, a preset target operation is performed on the target group.

[0093] Exemplarily, the target operation can be adjusting the recommended resources of the target group, for example, when the decision model judges that a group has high conversion potential, the budget, priority or exposure frequency of the advertisements seen by the users of the group is automatically increased; or when it is judged that there is a risk of loss, exclusive coupons or care content are automatically pushed to it.

[0094] The target operation can also be modifying the value level of the target group, for example, when it is found that a group originally marked as "ordinary" has a continuously strong consumption ability trend, the value level of the group can be automatically upgraded to "high value", so that more optimal configurations are obtained in subsequent tracking frequency and service priority.

[0095] By making judgments through a preset data-trained decision model, objective, consistent and traceable business responses can be ensured for the same or similar trend patterns, effectively avoiding the subjectivity of judgments and the randomness of operations due to experience and state differences of different operators, making the execution of operation strategies more accurate, standardized and scientific, and improving the overall operation quality.

[0096] The group state tracking method, device, equipment, computer storage medium and computer program product provided in the embodiments of the present application can first detect whether the target group exists after responding to the request message for querying the target group sent by the requestor, and in the case that the target group does not exist, at least one label data meeting the demand of the target group can be obtained from the multi-dimensional label data set, and the target group is obtained by combining the at least one label data. Then, based on the type of the target group, the tracking condition of the target group is determined, and when the tracking condition is reached, the index data of the target group is obtained based on the identification information. Finally, based on the index data, the trend data for representing the tracking index change trend of the target group is generated. This method fundamentally overturns the rigid mode of traditional pre-solidified group definition. By responding to business requests and automatically detecting whether the target group exists, when the group does not exist, the labels meeting the business demand can be filtered and combined from the multi-dimensional label data set in real time, and a new target group is dynamically generated, so that the business operator does not need to rely on the technical team to redevelop the data model or the ETL (Extract, Transform, Load) process, and can complete the self-definition and creation of the group within minutes or even seconds according to the demand, which significantly improves the flexibility and adaptability of group creation. At the same time, the personalized tracking index and tracking frequency are determined according to the group type, which optimizes the resource allocation and is conducive to further improving the flexibility and effectiveness of group tracking.

[0097] Based on the group state tracking method provided in the above embodiments, the present application also provides a specific implementation mode of a group state tracking device. Please refer to the following embodiments.

[0098] Firstly, referring to Figure 4 The group state tracking device 400 provided in the embodiments of the present application comprises: The acquisition module 410 is configured to acquire user data from a plurality of data sources, and perform labelization processing on the user data to obtain a multi-dimensional label data set. The detection module 420 is configured to detect whether the target group exists in response to a request message for querying the target group sent by the requester. The acquisition module 410 is further configured to acquire at least one label data satisfying the requirement of the target group from the multi-dimensional label data set in the case where the target group does not exist, and combine the at least one label data to obtain the target group and identification information of the target group. The determination module 430 is configured to determine a tracking condition of the target group based on a type of the target group, wherein the tracking condition comprises at least one tracking index and a corresponding tracking frequency. The acquisition module 410 is further configured to acquire index data of the target group based on the identification information in the case where the tracking frequency is satisfied. The generation module 440 is configured to generate trend data for representing a change trend of the tracking index of the target group based on the index data.

[0099] In some possible implementation manners, the tracking frequency is determined in the following manner: A value level of the target group is acquired. An initial tracking frequency of the target group is determined based on the value level. A historical index data fluctuation condition of the target group is acquired. A variation coefficient of the target group is determined based on the historical index data fluctuation condition. In the case where the variation coefficient is higher than a first threshold value, the initial tracking frequency is adjusted to be higher. In the case where the variation coefficient is not higher than a second threshold value, the initial tracking frequency is adjusted to be lower. In the case where the variation coefficient is between the first threshold value and the second threshold value, the initial tracking frequency is kept unchanged.

[0100] In some possible implementation manners, the multi-dimensional label data set comprises at least one of the following: user basic information, asset and liability information, product holding information, transaction behavior and channel characteristics. The acquisition module 410 is further configured to: construct a data wide table for representing an association relationship between the label data. acquire the at least one label data satisfying the requirement of the target group based on the data wide table.

[0101] In some possible implementation manners, the determination module 430 is further configured to: determine a tracking time based on the tracking frequency. The acquisition module 410 is further configured to: construct a time sequence data partition mapping table for representing an association relationship between the identification information, time, tracking index and index data. Based on the tracking time, the target time series data partition matching the tracking time is obtained from the time series data partition mapping table; Based on the identification information, the index data of the target group is obtained from the target time series data partition.

[0102] In some possible implementation ways, after obtaining the index data of the target group, the group state tracking apparatus 400 further includes a construction module. The construction module is configured to construct structured data including the identification information, the tracking time, the tracking index, and the index data. The structured data is stored in a preset relational database.

[0103] In some possible implementation ways, the generation module 440 is further configured to: The index data is sorted in chronological order to generate a time series data sequence. The time series data sequence is rendered into a statistical chart by using a chart rendering engine; the statistical chart includes at least one of a line chart, a column chart, or an area chart.

[0104] In some possible implementation ways, after generating the trend data used to represent the tracking index change trend of the target group based on the index data, the group state tracking apparatus 400 further includes an input module. The input module is configured to input the trend data into a preset decision model to obtain a decision result. Based on the decision result, a preset target operation is performed on the target group; the target operation includes at least one of adjusting a recommended resource of the target group, or modifying a value level of the target group.

[0105] The various modules of the group state tracking apparatus provided in the embodiments of the present application can achieve the functions of Figure 1 The functions of the various steps of the group state tracking method are provided, and the corresponding technical effects can be achieved; for brevity, the details are not described herein.

[0106] Figure 5 FIG. 1 is a structural schematic diagram of an electronic device provided in an embodiment of the present application.

[0107] The electronic device 500 can include a processor 501 and a memory 502 storing computer program instructions.

[0108] Specifically, the processor 501 can include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or can be configured as one or more integrated circuits implementing the embodiments of the present application.

[0109] The memory 502 can include mass storage for data or instructions. As an example and not by way of limitation, the memory 502 can include a Hard Disk Drive (HDD), a floppy disk drive, a flash memory, a Compact Disc (CD) or other optical disk, a Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, the memory 502 can include removable or non-removable (or fixed) media, where appropriate. Where appropriate, the memory 502 can be internal or external to the integrated gateway disaster recovery appliance. In particular embodiments, the memory 502 is non-volatile, solid-state memory.

[0110] In particular embodiments, the memory can include read-only memory (ROM), random access memory (RAM), magnetic disk storage mediums, optical storage mediums, flash memory devices, electrical, optical, or other physical / tangible memory storage devices. Thus, in general, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., a memory device) encoded with software that, when executed (by one or more processors), is operable to

[0111] The processor 501 implements any one of the group state tracking methods in the above embodiments by reading and executing computer program instructions stored in the memory 502.

[0112] In some examples, the electronic device 500 can further include a communication interface 503 and a bus 504. Where appropriate, as shown, the processor 501, the memory 502, and the communication interface 503 are connected by the bus 504 and accomplish communication among each other. Figure 5

[0113] The communication interface 503 is mainly used to realize the communication among the modules, devices, units and / or equipment in the embodiments of the present application.

[0114] ​Bus 504 includes hardware, software, or both, to couple components of the online data traffic billing device to each other in a form, for example, but not limited to, a bus. For example, but not limited to, the bus 504 can include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or another suitable bus or a combination of two or more of these. Where suitable, bus 504 can include one or more buses. Although particular buses are described and shown in the embodiments of the present application, the present application contemplates any suitable bus or interconnect.

[0115] For example, the electronic device 500 can be a mobile phone, a tablet computer, a notebook computer, a palm computer, an in-vehicle electronic device, an ultra-mobile personal computer (UMPC), a netbook, or a personal digital assistant (PDA), etc.

[0116] The electronic device 500 can perform the group state tracking method in the embodiments of the present application, thereby realizing the combination Figure 1 of the group state tracking method described above.

[0117] In addition, in combination with the group state tracking method in the above embodiments, the embodiments of the present application further provide a computer readable storage medium to implement. The computer readable storage medium has computer program instructions stored thereon; the computer program instructions are executed by a processor to implement any one of the group state tracking methods in the above embodiments. Examples of the computer readable storage medium include non-transitory computer readable storage media, such as a portable disc, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, etc.

[0118] The embodiments of the present application further provide a computer program product, which includes a computer program or instructions, and the computer program or instructions are executed by a processor to implement any one of the group state tracking methods in the above embodiments.

[0119] It should also be noted that the example embodiments mentioned in the present application describe some methods or systems based on a series of steps or devices. However, the present application is not limited to the order of the above steps, that is, the steps can be performed in the order mentioned in the embodiments, can be different from the order in the embodiments, or several steps can be performed simultaneously.

[0120] The computer program instructions can also be loaded onto a computer, other programmable data processing apparatus, or other processing device to cause a series of operational steps to be performed on the computer, other programmable apparatus or other processing device to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide processes for implementing the functions / acts specified in the flowchart and / or block diagram block or blocks.

[0121] The above is merely specific implementation of the present application, and those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described system, module and unit can refer to the corresponding process in the foregoing method embodiments, which will not be described herein. It should be understood that the protection scope of the present application is not limited to this, and any person skilled in the art can easily think of various equivalent modifications or replacements within the technical range disclosed in the present application, and these modifications or replacements shall be covered within the protection scope of the present application.

Claims

1. A group state tracking method, characterized in that, include: User data is obtained from multiple data sources, and the user data is tagged to obtain a multidimensional tagged dataset; In response to a request message sent by the requesting party for querying a target group, detect whether the target group exists; In the absence of the target group, at least one tag data that meets the requirements of the target group is obtained from the multidimensional tag dataset, and the at least one tag data is combined to obtain the target group and the identification information of the target group; Based on the type of the target group, determine the tracking conditions for the target group; The tracking conditions include at least one tracking indicator and a corresponding tracking frequency; If the tracking frequency is met, the indicator data of the target group is obtained based on the identification information; Based on the indicator data, trend data is generated to represent the changing trend of the tracking indicators for the target group.

2. The group status tracking method according to claim 1, characterized in that, The tracking frequency is determined in the following manner: Obtain the value level of the target group; Based on the value level, determine the initial tracking frequency of the target group; Obtain historical indicator data fluctuations for the target group; Based on the fluctuations in the historical indicator data, the coefficient of variation of the target group is determined; If the coefficient of variation is higher than a first threshold, the initial tracking frequency is increased. If the coefficient of variation is not higher than the second threshold, the initial tracking frequency is reduced. If the coefficient of variation is between the first threshold and the second threshold, the initial tracking frequency remains unchanged.

3. The group status tracking method according to claim 1, characterized in that, The multidimensional tag dataset includes at least one of the following: basic user information, asset and liability information, product holding information, transaction behavior, and channel characteristics; The step of obtaining at least one tag data that meets the requirements of the target group from the multidimensional tag dataset includes: Construct a wide table to represent the relationships between the label data; Based on the wide data table, obtain at least one tag data that meets the requirements of the target group.

4. The group status tracking method according to claim 1, characterized in that, Also includes: Based on the tracking frequency, the tracking time is determined; The step of obtaining the indicator data of the target group based on the identification information includes: Construct a time-series data partitioning mapping table to represent the relationships between identification information, time, tracking indicators, and indicator data; Based on the tracking time, obtain the target time-series data partition that matches the tracking time from the time-series data partition mapping table; Based on the identification information, the indicator data of the target group is obtained from the target time series data partition.

5. The group state tracking method according to any one of claims 1-4, characterized in that, After obtaining the indicator data of the target group, the method further includes: Construct structured data that includes identification information, tracking time, tracking indicators, and indicator data; The structured data is stored in a pre-defined relational database.

6. The group status tracking method according to any one of claims 1-4, characterized in that, The step of generating trend data based on the indicator data to represent the changing trend of the tracking indicators for the target group includes: The indicator data is sorted in chronological order to generate a time-series data sequence; The time-series data sequence is rendered into a statistical chart using a chart rendering engine; the statistical chart includes at least one of the following: line chart, bar chart, or area chart.

7. The group status tracking method according to any one of claims 1-4, characterized in that, After generating trend data representing the changing trend of tracking indicators for the target group based on the indicator data, the method further includes: The trend data is input into a preset decision model to obtain the decision result; Based on the decision result, a preset target operation is performed on the target group; the target operation includes at least one of the following: adjusting the recommended resources of the target group, modifying the value level of the target group.

8. A group status tracking device, characterized in that, The device includes: The acquisition module is used to acquire user data from multiple data sources, perform tagging processing on the user data, and obtain a multidimensional tag dataset. The detection module is used to detect whether the target group exists in response to a request message sent by the requester for querying the target group; The acquisition module is further configured to, in the absence of the target group, acquire at least one tag data that meets the requirements of the target group from the multidimensional tag dataset, combine at least one tag data to obtain the target group and the identification information of the target group; The determination module is used to determine the tracking conditions of the target group based on the type of the target group; the tracking conditions include at least one tracking indicator and a corresponding tracking frequency; The acquisition module is also used to acquire the indicator data of the target group based on the identification information, provided that the tracking frequency is met; The generation module is used to generate trend data based on the indicator data to represent the trend of changes in the tracking indicators of the target group.

9. A group status tracking device, characterized in that, The device includes: a processor and a memory storing computer program instructions; the processor, when executing the computer program instructions, implements the group state tracking method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer program instructions that, when executed by a processor, implement the group state tracking method as described in any one of claims 1-7.

11. A computer program product, characterized in that, When the instructions in the computer program product are executed by the processor of the electronic device, the electronic device is able to perform the group state tracking method as described in any one of claims 1-7.

Images