A user selection method, apparatus, device, and storage medium

Abstract

This invention discloses a user selection method, apparatus, device, and storage medium, relating to the field of computer technology. The method includes: receiving proprietary data generated by a data platform and externally input data, and obtaining real-time user data based on the proprietary data and the external data; inputting the real-time user data into a rule model pre-generated by a merchant through the data platform to perform data preprocessing operations on the real-time user data, obtaining preprocessed user data; comparing the preprocessed user data with rule data contained in the rule model to determine target user data that meets preset matching conditions from the preprocessed user data, and identifying the user corresponding to the target user data as the target user. This invention, by real-time access to changing user data and matching various rules customized by merchants, tags users in real-time based on massive amounts of data, ensuring the timeliness and stability of the user selection process.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to a user selection method, apparatus, device, and storage medium. Background Technology

[0002] Existing customer segmentation systems primarily rely on offline task mode to parse merchants' top-listed rules, scheduling tasks for customer segmentation on a T+1 basis. Furthermore, in SaaS (Software as a Service) platforms, each merchant has their own rules, potentially resulting in millions of tasks. Offline task execution is difficult to control in terms of time and resources. Therefore, existing capabilities are unsatisfactory in terms of supported task volume, timeliness, and stability. Summary of the Invention

[0003] In view of this, the purpose of this invention is to provide a user selection method, apparatus, device, and storage medium that can ensure the timeliness and stability of the user selection process. The specific solution is as follows:

[0004] In a first aspect, the present invention discloses a user selection method, comprising:

[0005] It receives proprietary data generated by the data platform and externally input data, and obtains real-time user data based on the proprietary data and the external data.

[0006] The real-time user data is input into the rule model pre-generated by the data platform so that data preprocessing operations can be performed on the real-time user data to obtain preprocessed user data.

[0007] The preprocessed user data is compared with the rule data contained in the rule model to determine the target user data that meets the preset matching conditions from the preprocessed user data, and the user corresponding to the target user data is determined as the target user.

[0008] Optionally, the receiving data platform generates its own data and externally input data, and obtains real-time user data based on the own data and the external data, including:

[0009] The system receives proprietary data generated by the data platform and externally input data, and performs extraction, transformation, loading, expansion, and aggregation operations on the proprietary data and the external data to obtain the real-time user data.

[0010] Optionally, before inputting the real-time user data into the rule model pre-generated by the merchant through the data platform, the method further includes:

[0011] Obtain the target user model selected by the merchant in the rule pop-up window of the data platform; wherein, the target user model includes a user attribute model, a user consumption model, and a user behavior model;

[0012] The rule model is generated based on real-time data processing technology and the target user model, and then stored in the database.

[0013] Optionally, the step of inputting the real-time user data into the merchant's pre-generated rule model through the data platform to perform data preprocessing operations on the real-time user data to obtain preprocessed user data includes:

[0014] The real-time user data is input into the rule model pre-generated by the merchant through the data platform, so as to use the interceptor to intercept user change data corresponding to user conditions that meet the rule model;

[0015] Perform preset verification and filtering operations on the user change data to obtain initial user change data;

[0016] The initial user change data is subjected to a preset processing and merging operation to obtain the preprocessed user data.

[0017] Optionally, the step of performing a preset processing and merging operation on the initial user change data to obtain the preprocessed user data includes:

[0018] Based on custom keywords, the mapping relationship of the initial user change data is determined, and the initial user change data with the same custom keywords are identified as data to be merged.

[0019] Compare the data generation times corresponding to the data to be merged, and merge the data to be merged with the data generation time that is latest to obtain the preprocessed user data.

[0020] Optionally, after comparing the preprocessed user data with the rule data contained in the rule model to determine target user data that meets preset matching conditions from the preprocessed user data, and determining the user corresponding to the target user data as the target user, the method further includes:

[0021] A first time point is determined where the target user does not meet the preset matching conditions, and a second time point is determined where the target user meets the preset matching conditions;

[0022] Set a user slide-out trigger at the first time point so that the target user is removed at the first time point;

[0023] Set a user slide-in trigger at the second time point so that the target user is moved in at the second time point.

[0024] Optionally, after comparing the preprocessed user data with the rule data contained in the rule model to determine target user data that meets preset matching conditions from the preprocessed user data, and determining the user corresponding to the target user data as the target user, the method further includes:

[0025] When the rule model changes, all the target user data will be identified as offline compensation data;

[0026] The offline compensation data is sent to the new rule model so that it can be determined whether the offline compensation data is valid based on the data existence status and data update time of the offline compensation data.

[0027] If the offline compensation data is valid, then the offline compensation data will be determined as the target user data of the new rule model.

[0028] Secondly, the present invention discloses a user selection device, comprising:

[0029] The user data acquisition module is used to receive proprietary data generated by the data platform and externally input data, and to obtain real-time user data based on the proprietary data and the external data.

[0030] The user data processing module is used to input the real-time user data into the rule model pre-generated by the merchant through the data platform, so as to perform data preprocessing operations on the real-time user data to obtain preprocessed user data.

[0031] The data comparison module is used to compare the preprocessed user data with the rule data contained in the rule model, so as to determine the target user data that meets the preset matching conditions from the preprocessed user data.

[0032] The user identification module is used to identify the user corresponding to the target user data as the target user.

[0033] Thirdly, the present invention discloses an electronic device, comprising:

[0034] Memory, used to store computer programs;

[0035] A processor for executing the computer program to implement the steps of the user selection method disclosed above.

[0036] Fourthly, the present invention discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, it implements the user selection method disclosed above.

[0037] As can be seen, the present invention provides a user selection method, comprising: receiving proprietary data generated by a data platform and externally input data, and obtaining real-time user data based on the proprietary data and the external data; inputting the real-time user data into a rule model pre-generated by the merchant through the data platform to perform data preprocessing operations on the real-time user data to obtain preprocessed user data; comparing the preprocessed user data with the rule data contained in the rule model to determine target user data that meets preset matching conditions from the preprocessed user data, and identifying the user corresponding to the target user data as the target user. Thus, the present invention, by real-time access to user change data, generating corresponding rule models based on the change data, processing the data using the rule models, matching the preprocessed data with various rules customized by the merchant to determine the target user, and then tagging the user in real time according to different choices, ensures the timeliness and stability of the user selection process. Attached Figure Description

[0038] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0039] Figure 1 This is a flowchart of a user selection method disclosed in this invention;

[0040] Figure 2 This is a flowchart of a specific user selection method disclosed in this invention;

[0041] Figure 3 This is a flowchart of a specific user selection method disclosed in this invention;

[0042] Figure 4 This is a schematic diagram of the user selection device structure provided by the present invention;

[0043] Figure 5 This invention provides a structural diagram of an electronic device. Detailed Implementation

[0044] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0045] Currently, existing user segmentation systems primarily rely on offline task mode to parse merchants' top-listed rules, scheduling tasks for user segmentation on a T+1 basis. Furthermore, in SaaS platforms, each merchant has its own rules, potentially resulting in millions of tasks, making it difficult to control time and resources during offline task execution. Therefore, existing capabilities are unsatisfactory in terms of supported task volume, timeliness, and stability. To address this, this invention provides a user selection method that ensures the timeliness and stability of the user selection process.

[0046] This invention discloses a user selection method, see [link to relevant documentation]. Figure 1 As shown, the method includes:

[0047] Step S11: Receive proprietary data generated by the data platform and external input data, and obtain real-time user data based on the proprietary data and the external data.

[0048] In this embodiment, proprietary data generated by the data platform and externally input data are received, and real-time user data is obtained based on the proprietary data and the external data. Specifically, proprietary data generated by the data platform and externally input data are received, and extraction, transformation, loading, expansion, and aggregation operations are performed on the proprietary data and the external data to obtain the real-time user data. It can be understood that, as Figure 2 As shown, the user's own data generated on the SaaS platform, as well as data input from external channels, are fed into the CDP (Customer Data Platform) data platform. The real-time user data is obtained through ETL data warehouse technology (Extract-Transform-Load), expansion, aggregation and other means. The expansion operation is to obtain other information related to the user information based on the input user information.

[0049] Step S12: Input the real-time user data into the rule model pre-generated by the merchant through the data platform, so as to perform data preprocessing operation on the real-time user data and obtain preprocessed user data.

[0050] In this embodiment, the system receives proprietary data generated by the data platform and externally input data. After obtaining real-time user data based on the proprietary and external data, it inputs the real-time user data into a rule model pre-generated by the merchant through the data platform. This allows for data preprocessing to be performed on the real-time user data, resulting in pre-processed user data. Specifically, the real-time user data is input into the rule model pre-generated by the merchant through the data platform. An interceptor is used to intercept user change data that meets the user conditions of the rule model. Pre-defined verification and filtering operations are performed on the user change data to obtain initial user change data. A pre-defined processing and merging operation is then performed on the initial user change data to obtain the pre-processed user data. It can be understood that the pre-defined processing and merging operation on the initial user change data to obtain the pre-processed user data specifically involves: determining a mapping relationship between the initial user change data and the data with the same custom keyword; comparing the data generation times of the data to be merged; and merging the data with the latest data generation time to obtain the pre-processed user data.

[0051] Understandably, the output is a real-time user model (consumption, behavior, attributes, etc.). The rule matching service in Flink (a framework and distributed processing engine) subscribes to the merchant's rules and sets interceptors in the state to intercept and match all user change data for the merchant to which the rule belongs. The partitioning rule for rules and user model data in Flink is the merchant ID (identity number), thus ensuring merchant-level isolation in computation and data. First, necessary validation and filtering are performed on the massive model data. Then, custom key mapping, state time comparison, and state merging are performed. After completing these steps, the data is validated against the rules.

[0052] Step S13: Compare the preprocessed user data with the rule data contained in the rule model to determine the target user data that meets the preset matching conditions from the preprocessed user data, and determine the user corresponding to the target user data as the target user.

[0053] In this embodiment, the real-time user data is input into the rule model pre-generated by the data platform so as to perform data preprocessing operation on the real-time user data. After obtaining the preprocessed user data, the preprocessed user data is compared with the rule data contained in the rule model so as to determine the target user data that meets the preset matching conditions from the preprocessed user data, and the user corresponding to the target user data is determined as the target user.

[0054] It is understood that after determining the user corresponding to the target user data as the target user, a first time point is determined where the target user does not meet the preset matching conditions, and a second time point is determined where the target user meets the preset matching conditions; a user slide-out trigger is set at the first time point so as to remove the target user at the first time point; a user slide-in trigger is set at the second time point so as to move the target user in at the second time point.

[0055] User rules may use data from multiple models. First, the user status is updated using the merged user status data from the previous step. Then, the user data is compared with the merchant's rule data, and the result is sent out to determine if the conditions are met. Simultaneously, triggers for the next slide-in / slide-out are set. For example, if a user meets the rule at 1:10 PM today but not at 0:00 AM 10 days later, the user needs to be slide-out. If the user meets the rule again 20 days later, the user's slide-in needs to be triggered again.

[0056] After user model changes and condition matching occurs, several scenarios may arise: 1. The match was valid at the time, but needs to be removed in the future; 2. The match was invalid at the time, but will be valid in the future. Since the calculation engine is data-driven, the entry and exit crowd packets triggered in the future need to be triggered by a scheduled task. After the data is matched, the next slide-in / slide-out time trigger is calculated, which is managed by Flink. After the timer is triggered, the user data in the state will be recalled to re-initiate the calculation. The matching result is sent downstream, and the previous step is calculated again at the same time.

[0057] This invention uses Flink as the streaming computing engine. Based on Flink, it implements dynamic compilation and hot reloading mechanisms for components through dynamic compilation using JDK (Java Development Kit) and Janino (Java compiler) scripts. It uses MySQL (relational database management system) to support changes in merchant rules and conditions. Kafka (an open-source stream processing platform) is used as a message middleware to decouple service dependencies and eliminate traffic spikes. Kudu (a column-oriented distributed database) and Elasticsearch (a search server) are used as data storage engines, supporting second-level data ingestion, high-concurrency queries, and flexible multi-dimensional retrieval capabilities. Kudu+Presto (an open-source distributed SQL query engine) is used as the ad-hoc query and user segmentation task engine, providing immediate query results and enabling real-time data updates in Kudu. User segmentation rules are scheduled in batches on demand via Presto.

[0058] This invention uses Flink real-time data processing technology to access various model changes such as user consumption, behavior, and attributes in real time, match them with various rules customized by merchants, and tag users in real time based on massive amounts of data to ensure the stability and timeliness of the merchant tagging system and better reach for precise marketing.

[0059] As can be seen, the present invention provides a user selection method, comprising: receiving proprietary data generated by a data platform and externally input data, and obtaining real-time user data based on the proprietary data and the external data; inputting the real-time user data into a rule model pre-generated by a merchant through the data platform to perform data preprocessing operations on the real-time user data to obtain preprocessed user data; comparing the preprocessed user data with the rule data contained in the rule model to determine target user data that meets preset matching conditions from the preprocessed user data, and identifying the user corresponding to the target user data as the target user. Thus, the present invention, by real-time access to user change data, generating corresponding rule models based on the change data, processing the data using the rule models, matching the preprocessed data with various rules customized by the merchant to determine the target user, and then tagging the user in real time according to different choices, ensures the timeliness and stability of the user selection process.

[0060] See Figure 3 As shown, this embodiment of the invention discloses a user selection method. Compared with the previous embodiment, this embodiment further explains and optimizes the technical solution.

[0061] Step S21: Receive proprietary data generated by the data platform and external input data, and obtain real-time user data based on the proprietary data and the external data.

[0062] Step S22: Obtain the target user model selected by the merchant in the rule pop-up window of the data platform.

[0063] In this embodiment, after obtaining real-time user data based on the proprietary data and the external data, the target user model selected by the merchant in the rule pop-up window of the data platform is obtained. The target user model includes a user attribute model, a user consumption model, and a user behavior model. It can be understood that the target audience for merchant activities is users. When a merchant wants to conduct activities targeting a user group with similar characteristics, they can first create a new audience package in the CDP audience rule pop-up window. The rules are built based on the user model of the CDP data platform, which includes multiple models such as the user attribute model, the user consumption model, and the user behavior model.

[0064] Step S23: Generate the rule model based on real-time data processing technology and the target user model, and store the rule model in the database.

[0065] In this embodiment, after obtaining the target user model selected by the merchant in the rule pop-up window of the data platform, the rule model is generated based on real-time data processing technology and the target user model, and the rule model is stored in the database. It is understood that the CDP audience rule pop-up window and the target user model selected by the merchant are converted into a rule model and stored in MySQL. The rule model supports AND and OR judgments for multiple conditions, and supports multiple self-implemented operators (computer keywords) for single conditions: equal, not empty, greater than or equal to, between, etc. Flink CDC (Change Data Capture) subscribes to the binlog (binary file) of this MySQL database, converts the rule conditions into a JSON-based rule model recognizable by the real-time matching service, sends it to Kafka, and is subscribed to by the real-time matching service, which intercepts real-time data and initiates offline compensation tasks within a certain time.

[0066] Step S24: Input the real-time user data into the rule model pre-generated by the merchant through the data platform, so as to perform data preprocessing operation on the real-time user data and obtain preprocessed user data.

[0067] Step S25: Compare the preprocessed user data with the rule data contained in the rule model to determine the target user data that meets the preset matching conditions from the preprocessed user data, and determine the user corresponding to the target user data as the target user.

[0068] In this embodiment, as described above Figure 2 As shown, data processing uses the wid (a globally unique ID for users, serving as user identifiers) as the dimension for audience rule matching. Data fields are trimmed and pre-encoded according to metadata and saved to the Flink state to reduce state size; offline / real-time data conflicts are resolved through time comparison, discarding outdated and dirty data; after data processing, the next nearest window trigger time is calculated according to rules (the nearest time is taken from multiple windows), and the corresponding timer is set.

[0069] Furthermore, the expected effects of the timed recalculation scheme are as follows: offline tasks only need to perform initialization operations and do not need to be executed daily (e.g., once globally, and once at the BOSS level when the BOSS (a globally unique merchant ID) configuration changes); changes in the window state in the offline data comparison rules are theoretically near real-time and do not require daily traffic smoothing, thus avoiding sudden traffic spikes; it is horizontally scalable and not limited by Kafka partitions. As mentioned above. Figure 2 As shown, the implementation details specifically include: message notification (message notification when configuration changes); scheduled tasks (setting triggers to trigger user slide-in / slide-out) or real-time communication (real-time calculation); scalability (model expansion); table scanning, with associated scenarios using Presto, and without associated scenarios directly scanning Kudu; implementation of the Presto query module, expanding support for conditions (single BOSSI); implementation of the Kudu scanning module, expanding support for conditions (single BOSSI). The data matching rules are as follows: target user data that meets preset matching conditions is determined from the preprocessed user data. When the user corresponding to the target user data is determined as the target user, the matching logic is implemented using operator expansion; the population condition loading mechanism adopts partitioning rules.

[0070] Step S26: When the rule model changes, all the target user data is determined as offline compensation data.

[0071] In this embodiment, target user data that meets preset matching conditions is determined from the preprocessed user data. After the user corresponding to the target user data is identified as the target user, when the rule model changes, all the target user data is identified as offline compensation data. It is understood that the offline compensation data is sent to the new rule model so that the validity of the offline compensation data can be determined based on its data existence status and data update time. If the offline compensation data is valid, it is identified as the target user data of the new rule model.

[0072] Understandably, once a merchant establishes a rule, such as targeting users who haven't spent less than 10 yuan in the past year, the real-time rule-based user targeting is event-triggered. It targets historical users who meet the merchant's conditions. Once the rule changes, the user model data under the merchant will be re-extracted. This data is sent to the matching service as offline compensation data to participate in calculations and status updates. If the data does not exist in the status or the update time is greater than the real-time change time in the status, it is considered valid data, and the matching results, timers, and real-time logic remain unchanged.

[0073] In addition, the real-time rules service is characterized by large data volume, frequent query requests, keyword search, and the need to provide offline contact capabilities. Therefore, a dual-write logic is adopted in the database write service. The CDP uses Elasticsearch's high-level API (Application Program Interface) to batch asynchronously write data to ES (elasticsearch), which improves write performance.

[0074] This invention is applied to the construction of CDP user tagging audience system, supporting millions of custom rule-based audience targeting and tagging tasks for merchants on the SaaS platform. It offers better timeliness and stability than conventional offline audience targeting and tagging, providing merchants with better capabilities for marketing outreach and real-time activities of other business parties.

[0075] For details regarding steps S21 and S24, please refer to the corresponding content disclosed in the foregoing embodiments, which will not be repeated here.

[0076] As can be seen, this application embodiment receives proprietary data generated by a data platform and externally input data, and obtains real-time user data based on the proprietary data and the external data; acquires the target user model selected by the merchant in the rule pop-up window of the data platform; generates the rule model based on real-time data processing technology and the target user model, and stores the rule model in a database; inputs the real-time user data into the rule model pre-generated by the merchant through the data platform, so as to perform data preprocessing operations on the real-time user data to obtain preprocessed user data; compares the preprocessed user data with the rule data contained in the rule model, so as to determine the target user data that meets the preset matching conditions from the preprocessed user data, and determines the user corresponding to the target user data as the target user; when the rule model changes, all the target user data is determined as offline compensation data, ensuring the timeliness and stability of the user selection process.

[0077] See Figure 4 As shown, this embodiment of the invention also discloses a user selection device, including:

[0078] User data acquisition module 11 is used to receive proprietary data generated by the data platform and externally input data, and obtain real-time user data based on the proprietary data and the external data.

[0079] User data processing module 12 is used to input the real-time user data into the rule model pre-generated by the merchant through the data platform, so as to perform data preprocessing operation on the real-time user data to obtain preprocessed user data.

[0080] The data comparison module 13 is used to compare the preprocessed user data with the rule data contained in the rule model, so as to determine the target user data that meets the preset matching conditions from the preprocessed user data.

[0081] User identification module 14 is used to identify the user corresponding to the target user data as the target user.

[0082] As can be seen, this invention includes: receiving proprietary data generated by a data platform and externally input data, and obtaining real-time user data based on the proprietary data and the external data; inputting the real-time user data into a rule model pre-generated by the merchant through the data platform to perform data preprocessing operations on the real-time user data to obtain preprocessed user data; comparing the preprocessed user data with the rule data contained in the rule model to determine target user data that meets preset matching conditions from the preprocessed user data, and identifying the user corresponding to the target user data as the target user. Thus, this invention, by accessing real-time user change data, generating corresponding rule models based on the change data, processing the data using the rule models, matching the preprocessed data with various rules customized by the merchant to determine target users, and then tagging users in real-time according to different choices, ensures the timeliness and stability of the user selection process.

[0083] In some specific embodiments, the user data acquisition module 11 specifically includes:

[0084] The data receiving unit is used to receive proprietary data generated by the data platform and externally input data.

[0085] The data processing unit is used to perform extraction, transformation, loading, expansion, and aggregation operations on the proprietary data and the external data to obtain the real-time user data.

[0086] In some specific embodiments, the user data processing module 12 specifically includes:

[0087] The target user model acquisition unit is used to acquire the target user model selected by the merchant in the rule pop-up window of the data platform; wherein, the target user model includes a user attribute model, a user consumption model, and a user behavior model;

[0088] The rule model generation unit is used to generate the rule model based on real-time data processing technology and the target user model, and store the rule model in the database;

[0089] A real-time user data input unit is used to input the real-time user data into the rule model pre-generated by the merchant through the data platform, so as to use an interceptor to intercept user change data corresponding to user conditions that meet the rule model;

[0090] The initial user change data acquisition unit is used to perform preset verification operations and preset filtering operations on the user change data to obtain initial user change data;

[0091] A mapping relationship determination unit is used to determine the mapping relationship of the initial user change data based on a custom keyword;

[0092] The data to be merged determination unit is used to determine the initial user change data with the same custom keyword as the data to be merged.

[0093] A data generation time comparison unit is used to compare the data generation times corresponding to the data to be merged.

[0094] The data merging unit is used to merge the data to be merged that has the latest data generation time to obtain the preprocessed user data.

[0095] In some specific embodiments, the data comparison module 13 specifically includes:

[0096] The data comparison unit is used to compare the preprocessed user data with the rule data contained in the rule model, so as to determine the target user data that meets the preset matching conditions from the preprocessed user data.

[0097] In some specific embodiments, the user determination module 14 specifically includes:

[0098] The user determination unit is used to determine the user corresponding to the target user data as the target user.

[0099] In some specific embodiments, after the user determination module 14, the system further includes:

[0100] The first time point determination unit is used to determine the first time point at which the target user does not meet the preset matching conditions;

[0101] The second time point determination unit is used to determine the second time point at which the target user meets the preset matching conditions;

[0102] A slide-out trigger social unit is configured to set a user slide-out trigger at the first time point so as to remove the target user at the first time point;

[0103] A slide-in trigger setting unit is used to set a user slide-in trigger at the second time point so that the target user is moved in at the second time point;

[0104] The offline compensation data determination unit is used to determine all the target user data as offline compensation data when the rule model changes.

[0105] A valid data determination unit is used to send the offline compensation data to the new rule model so as to determine whether the offline compensation data is valid based on the data existence status and data update time of the offline compensation data.

[0106] The target user data determination unit is used to determine the offline compensation data as the target user data of the new rule model if the offline compensation data is valid data.

[0107] Furthermore, embodiments of the present invention also provide an electronic device. Figure 5 This is a structural diagram of an electronic device 20 according to an exemplary embodiment. The content of the diagram should not be construed as limiting the scope of the invention.

[0108] Figure 5 This is a schematic diagram of the structure of an electronic device 20 provided in an embodiment of the present invention. Specifically, the electronic device 20 may include: at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input / output interface 25, and a communication bus 26. The memory 22 stores a computer program, which is loaded and executed by the processor 21 to implement the relevant steps in the user selection method disclosed in any of the foregoing embodiments. Furthermore, the electronic device 20 in this embodiment may specifically be an electronic computer.

[0109] In this embodiment, the power supply 23 is used to provide operating voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and external devices, and the communication protocol it follows can be any communication protocol applicable to the technical solution of this invention, and is not specifically limited here; the input / output interface 25 is used to acquire external input data or output data to the outside world, and its specific interface type can be selected according to specific application needs, and is not specifically limited here.

[0110] In addition, the memory 22, as a carrier for resource storage, can be a read-only memory, random access memory, disk or optical disk, etc. The resources stored thereon can include operating system 221, computer program 222, etc., and the storage method can be temporary storage or permanent storage.

[0111] The operating system 221 is used to manage and control the various hardware devices on the electronic device 20 and the computer program 222, which may be Windows Server, Netware, Unix, Linux, etc. In addition to including a computer program capable of performing the user selection method executed by the electronic device 20 as disclosed in any of the foregoing embodiments, the computer program 222 may further include a computer program capable of performing other specific tasks.

[0112] Furthermore, this embodiment of the invention also discloses a storage medium storing a computer program, which, when loaded and executed by a processor, implements the user selection method steps disclosed in any of the foregoing embodiments.

[0113] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section.

[0114] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0115] The above provides a detailed description of a user selection method, apparatus, device, and storage medium provided by the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A user selection method, characterized in that, include: It receives proprietary data generated by the data platform and externally input data, and obtains real-time user data based on the proprietary data and the external data. In a SaaS multi-tenant architecture, the real-time user data is input into the rule model pre-generated by the data platform through a real-time streaming computing engine, so as to perform data preprocessing operations on the real-time user data and obtain preprocessed user data; wherein, the real-time streaming computing engine is partitioned by merchant ID to achieve isolation between merchant-level data and computing. The preprocessed user data is compared with the rule data contained in the rule model by a real-time streaming computing engine, so as to determine the target user data that meets the preset matching conditions from the preprocessed user data, and the user corresponding to the target user data is determined as the target user. The receiving data platform generates its own data and externally input data, and obtains real-time user data based on the own data and the external data, including: The system receives proprietary data generated by the data platform and externally input data, and performs extraction, transformation, loading, expansion, and aggregation operations on the proprietary data and the external data to obtain the real-time user data; wherein, the expansion operation is the operation of obtaining other information associated with the input user information; the extraction, transformation, and loading operations are completed through ETL data warehouse technology; The step of inputting the real-time user data into the merchant's pre-generated rule model through the data platform to perform data preprocessing operations on the real-time user data to obtain preprocessed user data includes: The real-time user data is input into the rule model pre-generated by the merchant through the data platform, so as to use the interceptor to intercept user change data corresponding to user conditions that meet the rule model; Perform preset verification and filtering operations on the user change data to obtain initial user change data; The initial user change data is subjected to a preset processing and merging operation to obtain the preprocessed user data; The step of comparing the preprocessed user data with the rule data contained in the rule model to determine target user data that meets preset matching conditions from the preprocessed user data, and determining the user corresponding to the target user data as the target user, further includes: A first time point is determined where the target user does not meet the preset matching conditions, and a second time point is determined where the target user meets the preset matching conditions; Set a user slide-out trigger at the first time point so that the target user is removed at the first time point; Set a user slide-in trigger at the second time point so that the target user is moved in at the second time point; The step of comparing the preprocessed user data with the rule data contained in the rule model to determine target user data that meets preset matching conditions from the preprocessed user data, and determining the user corresponding to the target user data as the target user, further includes: When the rule model changes, all the target user data will be identified as offline compensation data; The offline compensation data is sent to the new rule model so that it can be determined whether the offline compensation data is valid based on the data existence status and data update time of the offline compensation data. If the offline compensation data is valid, then the offline compensation data will be determined as the target user data of the new rule model.

2. The user selection method according to claim 1, characterized in that, Before inputting the real-time user data into the rule model pre-generated by the merchant through the data platform, the method further includes: Obtain the target user model selected by the merchant in the rule pop-up window of the data platform; wherein, the target user model includes a user attribute model, a user consumption model, and a user behavior model; The rule model is generated based on real-time data processing technology and the target user model, and then stored in the database.

3. The user selection method according to claim 1, characterized in that, The preprocessing and merging operation on the initial user change data to obtain the preprocessed user data includes: Based on custom keywords, the mapping relationship of the initial user change data is determined, and the initial user change data with the same custom keywords are identified as data to be merged. Compare the data generation times corresponding to the data to be merged, and merge the data to be merged with the data generation time that is latest to obtain the preprocessed user data.

4. A user selection device, characterized in that, include: The user data acquisition module is used to receive proprietary data generated by the data platform and externally input data, and to obtain real-time user data based on the proprietary data and the external data. The user data processing module is used in a SaaS multi-tenant architecture to input real-time user data into a rule model pre-generated by the merchant through the data platform via a real-time streaming computing engine, so as to perform data preprocessing operations on the real-time user data to obtain preprocessed user data; wherein, the real-time streaming computing engine is partitioned by merchant ID to achieve isolation between merchant-level data and computing. The data comparison module is used to compare the preprocessed user data with the rule data contained in the rule model through a real-time streaming computing engine, so as to determine the target user data that meets the preset matching conditions from the preprocessed user data. The user identification module is used to identify the user corresponding to the target user data as the target user; The user data acquisition module is specifically used to receive proprietary data generated by the data platform and externally input data, and to perform extraction, transformation, loading, expansion, and aggregation operations on the proprietary data and the external data to obtain the real-time user data; wherein, the expansion operation is the operation of obtaining other information associated with the input user information; the extraction, transformation, and loading operations are completed through ETL data warehouse technology; Specifically, the user data processing module is used to input the real-time user data into the rule model pre-generated by the merchant through the data platform, so as to use an interceptor to intercept user change data corresponding to user conditions that meet the rule model; to perform preset verification and preset filtering operations on the user change data to obtain initial user change data; and to perform preset processing and merging operations on the initial user change data to obtain the preprocessed user data. The user selection device further includes: The user move-in / move-out module is used to determine a first time point at which the target user does not meet the preset matching conditions, and to determine a second time point at which the target user meets the preset matching conditions; a user slide-out trigger is set at the first time point to remove the target user at the first time point; a user slide-in trigger is set at the second time point to move the target user in at the second time point. The user selection device further includes: The offline compensation module is used to determine all target user data as offline compensation data when the rule model changes; send the offline compensation data to the new rule model so as to determine whether the offline compensation data is valid data based on the data existence status and data update time of the offline compensation data; if the offline compensation data is valid data, then the offline compensation data is determined as the target user data of the new rule model.

5. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor for executing the computer program to implement the steps of the user selection method as described in any one of claims 1 to 3.

6. A computer-readable storage medium, characterized in that, Used to store a computer program; wherein, when the computer program is executed by a processor, it implements the user selection method as described in any one of claims 1 to 3.

Images