Real-time label processing method and device, equipment and storage medium

Abstract

The invention provides a real-time label processing method and device, equipment and a storage medium, and the method comprises the steps: obtaining customer data in real time, carrying out the data cleaning of the customer data, and obtaining customer related data; classifying the customer-related data based on a data type, and respectively constructing a basic attribute tag and an event tag according to a classification result; setting a rule tag according to a business demand and the event tag, selecting corresponding entity data in the customer related data based on the rule tag, and generating an entity tag; acquiring offline data of a customer, calculating the offline data based on a first label algorithm, and generating an offline pre-calculation result set according to a calculation result; and selecting corresponding data based on the entity tag and the customer basic data tag, constructing a real-time mining tag based on the corresponding data and the offline pre-calculation result set by using the first tag algorithm, and circling and selecting target data based on the real-time mining tag.

Description

Technical Field

[0001] The present disclosure relates to the field of data processing, and in particular to a real-time tag processing method, device, equipment and storage medium. Background Art

[0002] The existing real-time tag technology solution only provides the real-time capability of simple attribute tags, and lacks the real-time capability for complex tags. Specifically, the existing so-called real-time tags actually only provide simple attribute tags, such as gender, age, birthday, city, membership level, and personal attributes such as cumulative consumption and cumulative visits, as well as cumulative indicators. These data indicators can be obtained through simple business processing. For complex tags, such as rule-based (number of transactions in the past 30 days>2 times, etc.) and mining-based tags (customer value, customer stage, prediction tags), real-time support is not provided in the existing real-time tags. In current business, such tags are all implemented through offline calculations, resulting in errors in business use. For example, if a customer currently places an order, the number of orders placed by the corresponding customer in the tag may still be 0. Then, marketing for customers with 0 orders will cause errors. Without a complete real-time tag system, it is impossible to meet the rapidly changing customer needs. Summary of the invention

[0003] The present disclosure provides a real-time tag processing method, device, equipment and storage medium to at least solve the above technical problems existing in the prior art.

[0004] According to a first aspect of the present disclosure, a real-time tag processing method is provided, the method comprising:

[0005] Acquire customer data in real time, perform data cleaning on the customer data, and obtain customer-related data;

[0006] Classify the customer-related data based on data types, and construct basic attribute labels and event labels according to the classification results;

[0007] Setting rule tags according to business requirements and the event tags, circle corresponding entity data in the customer-related data based on the rule tags and generate entity tags;

[0008] Obtaining offline data of the customer, calculating the offline data based on a first label algorithm, and generating an offline pre-calculation result set according to the calculation result;

[0009] Corresponding data is selected based on the entity tag and the customer basic data tag, real-time mining tags are constructed based on the corresponding data and the offline pre-calculated result set using the first tag algorithm, and target data is circled based on the real-time mining tags.

[0010] In one possible implementation, constructing basic attribute labels and event labels respectively according to the classification results includes:

[0011] Selecting business data from the customer-related data, constructing the basic attribute tag based on the business data, marking the business data based on the basic attribute tag and storing the data in a basic data tag pool;

[0012] The customer events in the customer-related data are counted, event-related data in the customer-related data are selected based on the customer events, event tags are constructed based on the event-related data, and the event-related data are marked based on the event tags and stored in an event tag pool.

[0013] In one possible implementation, the step of selecting corresponding entity data in the customer-related data based on the rule tag and generating an entity tag includes:

[0014] Selecting first business data and first event-related data matching the rule tag from the basic data tag pool and the event tag pool based on the rule tag;

[0015] An entity tag is constructed based on the first business data and the first event-related data, and data corresponding to the entity tag is saved.

[0016] In one possible implementation manner, the step of inputting the customer offline data into a first algorithm model for processing to obtain an offline pre-calculation result set includes:

[0017] Selecting a first algorithm model based on the business requirements, inputting the offline data into the first algorithm model, and constructing an offline mining label;

[0018] The offline pre-calculation result set is constructed based on the offline mining tags and the historical data.

[0019] In one embodiment, the method further comprises:

[0020] The rule tag is set according to the business requirement and the event tag, and a rule tag pool is established based on the rule tag;

[0021] According to the business requirements, select the corresponding rule tag from the rule tag pool;

[0022] Based on the selected rule tag, corresponding entity data in the customer-related data is circled and an entity tag is generated, and an entity tag pool is constructed based on the entity tag;

[0023] A first entity tag is selected from the entity tag pool according to the business requirement, and corresponding data is selected based on the first entity tag and the customer basic data tag.

[0024] According to a second aspect of the present disclosure, a real-time tag processing device is provided, the device comprising:

[0025] A data processing unit, used to obtain customer data in real time, clean the customer data, and obtain customer-related data;

[0026] A real-time data label construction unit is used to classify the customer-related data based on data types, and to construct basic attribute labels and event labels according to the classification results; to set rule labels according to business requirements and the event labels, and to circle corresponding entity data in the customer-related data based on the rule labels and generate entity labels;

[0027] An offline pre-calculation result set construction unit, used to obtain offline data of a customer, calculate the offline data based on a first label algorithm, and generate an offline pre-calculation result set according to the calculation result;

[0028] The real-time prediction unit is used to select corresponding data based on the entity label and the customer basic data label, use the first label algorithm to build a real-time mining label based on the corresponding data and the offline pre-calculated result set, and select target data based on the real-time mining label.

[0029] In one possible implementation, the real-time data tag construction unit is further used to select business data from the customer-related data, construct the basic attribute tag based on the business data, mark the business data based on the basic attribute tag and store it in a basic data tag pool; count customer events in the customer-related data, select event-related data from the customer-related data based on the customer events, construct event tags based on the event-related data, mark the event-related data based on the event tags and store it in an event tag pool; select the first business data and the first event-related data that match the rule tag from the basic data tag pool and the event tag pool based on the rule tag; construct an entity tag based on the first business data and the first event-related data, and save the data corresponding to the entity tag;

[0030] The offline pre-calculation result set construction unit is also used to select a first algorithm model based on the business requirements, input the offline data into the first algorithm model, and construct offline mining tags; and construct the offline pre-calculation result set based on the offline mining tags and the historical data.

[0031] In one embodiment, the device further comprises:

[0032] A data selection unit is used to set the rule tag according to the business needs and the event tag, and establish a rule tag pool based on the rule tag; select the corresponding rule tag from the rule tag pool according to the business needs; circle the corresponding entity data in the customer-related data based on the selected rule tag and generate an entity tag, and build an entity tag pool based on the entity tag; select a first entity tag from the entity tag pool according to the business needs, and select corresponding data based on the first entity tag and the customer basic data tag.

[0033] According to a third aspect of the present disclosure, there is provided an electronic device, including:

[0034] at least one processor; and

[0035] a memory communicatively connected to the at least one processor; wherein,

[0036] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method described in the present disclosure.

[0037] According to a fourth aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to cause the computer to execute the method described in the present disclosure.

[0038] The disclosed real-time label processing method, device, equipment and storage medium construct customer basic attribute labels and behavior event labels from real-time acquired customer data, construct rule labels and entity labels according to label corresponding data and business requirements, construct offline mining labels from offline data, and generate prediction results according to the data corresponding to basic attribute labels, behavior event labels, rule labels, entity labels and offline mining labels to select target groups. This solves the problem that existing customer labels cannot be real-time, allowing enterprises to use labels more conveniently and accurately for precision marketing, as well as various products combined with labels.

[0039] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The above and other objects, features and advantages of the exemplary embodiments of the present disclosure will become readily understood by reading the detailed description below with reference to the accompanying drawings. In the accompanying drawings, several embodiments of the present disclosure are shown in an exemplary and non-limiting manner, in which:

[0041] In the drawings, the same or corresponding reference numerals represent the same or corresponding parts.

[0042] Figure 1 The following is a schematic diagram showing the implementation process of a real-time tag processing method according to an embodiment of the present disclosure. Figure 1 ;

[0043] Figure 2 The following is a schematic diagram showing the implementation process of a real-time tag processing method according to an embodiment of the present disclosure. Figure 2 ;

[0044] Figure 3 A schematic diagram of real-time label construction according to an embodiment of the present disclosure is shown;

[0045] Figure 4 A schematic diagram of a real-time tag processing device according to an embodiment of the present disclosure is shown;

[0046] Figure 5 A schematic diagram of the structure of an electronic device according to an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0047] In order to make the purpose, features, and advantages of the present disclosure more obvious and easy to understand, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present disclosure.

[0048] Figure 1 The following is a schematic diagram showing the implementation process of a real-time tag processing method according to an embodiment of the present disclosure. Figure 1 ,like Figure 1 As shown, a real-time tag processing method according to an embodiment of the present disclosure includes the following steps:

[0049] Step 101, acquiring customer data in real time, performing data cleaning on the customer data, and obtaining customer-related data.

[0050] In the disclosed embodiment, a real-time data source is accessed based on a distributed data flow engine (Flink) to obtain customer data in real time. The customer data includes customer basic data, customer attribute data, order data, tracking data, marketing data, interactive data, etc.

[0051] In the disclosed embodiments, the data cleaning operation includes: data preprocessing, formatting, normalizing, deduplication, and removal of outliers, such as the gender field, which normally contains male, female, and unknown; if the original data contains the above three types of data, the data needs to be normalized. Similarly, date of birth, mobile phone number, etc. need to process invalid or abnormally formatted data. Data conversion, encoding conversion, type conversion and other operations are performed on the data. For example, the data format of the mobile phone number from the original mobile phone is a string type, which needs to be converted to a long type for storage to facilitate subsequent use. Data verification, after the above steps are processed, the data is verified to ensure that the above processing process is normal and accurate. The verification content includes specification checking, type checking, etc.

[0052] Step 102: classify the customer-related data based on the data type, and construct basic attribute labels and event labels according to the classification results.

[0053] In the disclosed embodiment, the customer-related data is classified based on the data type, and the classification results include business data and event-related data. Among them, the business data includes relevant information of the customer, specifically: gender, age, birthday, address, source and other information. Business data in the customer-related data is selected, the basic attribute label is constructed based on the business data, the business data is marked based on the basic attribute label and stored in the basic data label pool; customer events in the customer-related data are counted, wherein customer events include: user order events, user order modification events, etc. Event-related data in the customer-related data is selected based on the customer events, and the event-related data contains all the customer's behavior trajectory processes and behavior times. Event labels are constructed based on the event-related data, and the event-related data are marked based on the event labels and stored in the event label pool.

[0054] Step 103: setting a rule tag according to business requirements and the event tag, circle corresponding entity data in the customer-related data based on the rule tag, and generate an entity tag.

[0055] In the disclosed embodiment, the required rule tags are set according to the business requirements and the event tags generated above, for example, "number of orders placed in the last 30 days" is used as the rule tag.

[0056] In the disclosed embodiment, the first business data and the first event-related data that match the rule tag are selected from the basic data tag pool and the event tag pool based on the rule tag; specifically, the first business data and the first event-related data respectively include the relevant data of the above rule tag "number of orders in the last 30 days". An entity tag is constructed based on the first business data and the first event-related data, and the data corresponding to the entity tag is saved. Among them, the entity tag is a tag that can be directly created using the rule in the application, that is, the relevant data corresponding to the "number of orders in the last 30 days" tag.

[0057] In the disclosed embodiment, the rule tag is set according to the business needs and the event tag, and a rule tag pool is established based on the rule tag; wherein, the rule tag that has been set is stored in the rule tag pool, and when the rule is generated again, the existing rule tag can be directly selected from the tag pool. According to the business needs, the corresponding rule tag is selected from the rule tag pool; based on the selected rule tag, the corresponding entity data in the customer-related data is circled and an entity tag is generated, and an entity tag pool is constructed based on the entity tag; according to the business needs, a first entity tag is selected from the entity tag pool, and corresponding data is selected based on the first entity tag and the customer basic data tag.

[0058] Step 104: Obtain offline data of the customer, calculate the offline data based on the first label algorithm, and generate an offline pre-calculation result set according to the calculation result.

[0059] In the disclosed embodiment, the offline data is similar to the above-mentioned real-time acquired data, which are event-related data, customer-related data, etc.

[0060] The first algorithm model is selected based on the business needs, wherein the first algorithm model selects different algorithm models according to different analysis situations. For example, when analyzing customer value tags, a clustering algorithm can be selected for processing. The offline data is input into the first algorithm model to construct offline mining tags; the offline pre-calculated result set is constructed based on the offline mining tags and the historical data. Among them, the offline mining tags correspond to the customers. Taking the customer value tags as offline mining tags as an example, the tags include: high-value loyal customers, important retained customers, high-quality development customers and other tags. The pre-calculated result set is composed of the results of scoring users based on the above tags and / or customer-related data.

[0061] Step 105: Select corresponding data based on the entity tag and the customer basic data tag, use the first tag algorithm to build a real-time mining tag based on the corresponding data and the offline pre-calculated result set, and select target data based on the real-time mining tag.

[0062] In the disclosed embodiment, when the real-time computing service processes the corresponding customer data, it obtains the offline pre-calculation result set, combines the basic attribute tags, event tags, rule tags and entity tags constructed above with the corresponding real-time data that can be obtained, and uses the first tag algorithm to calculate the real-time data and the above-mentioned offline pre-calculation result set to obtain the latest real-time mining tags. Among them, the calculation rules of the real-time mining tags are the same as the calculation rules of the offline mining tags.

[0063] Figure 2 The following is a schematic diagram showing the implementation process of a real-time tag processing method according to an embodiment of the present disclosure. Figure 2 , Figure 3 A schematic diagram of real-time label construction according to an embodiment of the present disclosure is shown. Figure 2 and Figure 3 As shown, a real-time tag processing method according to an embodiment of the present disclosure includes the following steps:

[0064] Step 201: access a real-time data source to obtain data.

[0065] In the disclosed embodiment, the time tag computing service (based on the Flink engine) is connected to the real-time data source, and the real-time data source includes customer basic data, customer attribute data, order data, tracking data, marketing data, interaction data, etc.

[0066] Table 1. Customer basic information data

[0067]

[0068]

[0069] As shown in Table 1, customer basic data: contains the basic information of the customer, including customer ID, name, mobile phone number, WeChat ID and other information.

[0070] Table 2. Customer attribute data (i.e., customized attribute information in addition to basic customer attributes)

[0071]

[0072] As shown in Table 2, customer attribute data: includes custom attribute information of customers in addition to basic information, such as other mobile phone numbers, baby information, membership information, etc. Different companies can customize different customer attribute information.

[0073] Table 3. Order data

[0074]

[0075]

[0076] As shown in Table 3, order data: includes order information from different channels, such as Taobao, Douyin, Kuaishou, etc. The order information includes order number, order time, amount, product name, transaction status and other information.

[0077] Table 4. Buried point data

[0078]

[0079]

[0080] As shown in Table 4, the embedded data includes the browsing information data of customers on the mini program, such as login, search, browsing products, adding to cart, favorites, participating in activities, exiting and other information.

[0081] Table 5. Marketing data

[0082]

[0083] As shown in Table 5, marketing data: includes marketing data for the customer, such as WeChat group messages, SMS messages, phone messages, mini-program template messages, and other customer contact behaviors based on specific basis or intelligent rules.

[0084] Table 6. Interaction data

[0085]

[0086] As shown in Table 6, interactive data: such as activity interaction information, chat interaction data, group chat interaction data, etc.

[0087] Step 202: data cleaning.

[0088] In the disclosed embodiment, after access, real-time data cleaning and data processing are performed, and the real-time data cleaning process is as follows: data preprocessing, formatting, normalizing, deduplication, and removal of outliers on the data, such as the gender field, which normally contains male, female, and unknown; if the original data contains the above three types of data, the data needs to be normalized. Similarly, date of birth, mobile phone number, etc. need to process invalid or abnormally formatted data. Data conversion, encoding conversion, type conversion and other operations are performed on the data. For example, the data format of the mobile phone number from the original mobile phone is a string type, which needs to be converted to a long type for storage to facilitate subsequent use. Data verification, after the above steps are processed, the data is verified to ensure that the above processing process is normal and accurate. The verification content includes specification checking, type checking, etc.

[0089] Step 203: construct customer basic attribute labels.

[0090] In the disclosed embodiment, customer basic data and customer attribute data are used to construct customer basic attribute labels according to specific business process requirements and are stored in a basic data label pool of a data analysis engine (clickhouse).

[0091] Table 7. Basic attribute label examples

[0092]

[0093]

[0094]

[0095] Step 204: construct a customer behavior event label.

[0096] In the disclosed embodiment, 3. After real-time data cleaning and data processing, the order, embedding, marketing, and interactive data are supplemented with user information to filter invalid data, and customer behavior event tag data is constructed and stored in the event tag pool. During the processing, temporary data will be stored in the redis cache for easy and quick access. The event data contains all the customer's behavior trajectory processes and behavior time, so that rule tags can be flexibly constructed by selecting different time ranges.

[0097] Step 205: construct a rule tag.

[0098] In the embodiment of this disclosure, taking the label "Number of orders placed in the past 30 days" as an example, the total number of orders placed by customers through all channels in the past 30 days (excluding the current day) is calculated based on the main order. The default hierarchical divisions are: 0 times, 1 time, 2-3 times, 4-5 times, 6-10 times, and more than 10 times. We select the order event to construct the above label (order time: createTime, today is 2024-09-01 00:00:00).

[0099] Table 8. Rule tags

[0100]

[0101]

[0102] In the disclosed embodiment, after configuring the rules of the tag, when using this tag, the tag will be dynamically generated, and the specified rules will generate the sql statement created by the corresponding tag, as follows. Each layer will build a similar sql statement to generate the corresponding layered data. When the customer uses the tag, the tag will be dynamically executed to generate results and use. The rule selector can construct rule tags by selecting customer basic attribute tags and customer behavior event tags. Taking the transaction amount in the last 30 days as an example, as long as the time range of the order event is selected to be within the last 30 days and the order has been completed, the creation of the rule tag can be completed. After the rule tag is created, the real-time tag calculation service reads the configured rule requirements and builds the entity tag. The entity data corresponding to the tag created by the rule can be directly used in the application.

[0103] Step 206: Offline mining label construction.

[0104] In the disclosed embodiment, the offline computing service loads offline data sources and data warehouse data models, combines specific label requirements, and constructs offline mining labels through algorithm calculation, such as customer value labels, and uses clustering algorithms to process and separate high-value customers and low-value customers. The offline data source is similar to the real-time data source, which is customer basic data, attribute data, and behavior data. The offline label uses the big data technology framework Hadoop as the underlying technology, builds a data warehouse through Hive, and uses the Hive / Spark engine as the computing engine. The task is developed in the SQL type. After the data calculation is completed, it is stored in the MySQL database or the ClickHouse engine.

[0105] In the disclosed embodiment, taking the customer value label as an example, the customer value label calculates the customer's R value, F value, and M value (R represents the number of days since the last transaction, F represents the transaction frequency, and M represents the cumulative transaction amount) according to the RFM model weighted mechanism method, and then uses the quantile integration method to obtain the customer's current points (the maximum score is 30 points, and the higher the customer's points, the greater the value). The default stratification is divided into: high-value loyal customers (27-30 points): loyal customers who continuously generate consumption behaviors and each consumption amount is relatively high. For example, new customers have completely become old customers, or even become loyal members. Important retained customers (23-26 points): retained customers who start high-frequency consumption or generate high-value consumption. For example, the first order amount of new customers is high & new customers begin to continuously generate repurchase behavior. High-quality development customers (19-22 points): customers who have recently had transactions and the transaction amount is higher than the average customer unit price, and the number of transactions is significantly lower than the average. For example, customers have recently generated a single high-priced order. High-quality ordinary customers (15-18 points): customers begin to generate consumption behaviors, but the consumption frequency or amount is low. If a new customer has made their first order, they can be stimulated to continue to generate consumption conversions. General retention customers (11-14 points): Customers who have made high-frequency or high-amount consumption behaviors, but have not made any consumption in the recent period. If the consumption demand of old customers has decreased in the recent period, they need to be stimulated. Customers to be activated (7-10 points): Customers who have made high-frequency or high-amount consumption behaviors, but have not made any consumption in a long period of time. If old customers who have made high-frequency consumption have not traded again for a long time. Dormant customers (4-6 points): Customers who have made high-amount consumption but have a relatively low consumption frequency and have not made any consumption in the recent period. If the customer's willingness to consume continues to decrease or even disappear. Retained customers (0-3 points): Users who have not made any consumption behaviors for a long time and whose consumption frequency and amount are not high. If the marketing means have not achieved the effect of awakening old customers. The customer value label stratification includes: high-value loyal customers, important retained customers, high-quality development customers, high-quality ordinary customers, general retained customers, customers to be activated, dormant customers, and retained customers.

[0106] Step 207, real-time mining tag construction.

[0107] In the disclosed embodiment, when the real-time computing service processes the corresponding customer data, it obtains the offline pre-calculated result set, combines the real-time data, obtains the latest mining label (such as customer value) result, and stores the result in the data analysis engine (ClickHouse). Step 206 has explained the calculation process of the offline customer value label, so the real-time mining label only needs to calculate the score of a single customer according to the offline rules, and then set the real-time stratification of the customer according to the stratification. Why is the participation of offline labels required here? The reason is that it relies on the average value of the offline full calculation results, and then weights it with a single customer to obtain the customer's true score and corresponding stratification. In this way, the real-time construction of customer value is realized (actually relying on offline pre-processing combing + real-time single customer data). After the above steps are processed and executed, all real-time label construction is completed, and the labels include customer basic labels, customer consumption labels, customer marketing labels, customer interaction labels, customer value labels, and customer behavior event labels; by selecting between different labels, the appropriate group of people can be selected to achieve business.

[0108] In the disclosed embodiment, compared with offline tags, offline tags use batch data processing technology to store and process the generated data for a certain period of time, and then produce and update the tags. This causes offline tag data to have a large lag and cannot meet the scenarios that require rapid response, such as real-time search, real-time recommendation, and real-time marketing in e-commerce scenarios. Real-time tags are different. When the data changes, the real-time computing service can quickly get the changed content and immediately calculate and update the tags, which helps to make changes quickly in business scenarios. For example, in e-commerce scenarios, real-time tags can recommend relevant products and discount information in real time based on the customer's real-time browsing and purchasing behavior. This personalized promotion can enhance the customer's purchasing experience and increase the customer's willingness to buy. Compared with existing real-time tag solutions, existing real-time tag solutions are based on traditional technical architectures, including business processing and database architecture solutions, which only provide simple customer attribute tag real-time processing capabilities, and fixed behavioral indicators as real-time scenarios, such as RFM models and other specific scenario functions; there is no way to flexibly construct rule-based tags through real-time basic tags and behavioral events through rule configuration, not to mention the more difficult offline + incremental-based solutions to achieve real-time mining tags. First, the existing real-time labeling solutions are still based on traditional underlying technical solutions, which have great disadvantages in processing massive data and large-scale real-time computing. Secondly, there is no effective real-time integration solution. There is no effective and feasible solution for implementing complex rule labels and mining class labels, which requires high costs or offline implementation.

[0109] Figure 4 A schematic diagram of a real-time tag processing device according to an embodiment of the present disclosure is shown. Figure 4 As shown, a real-time label processing device according to an embodiment of the present disclosure includes:

[0110] The data processing unit 401 is used to obtain customer data in real time, clean the customer data, and obtain customer-related data;

[0111] The real-time data label construction unit 402 is used to classify the customer-related data based on the data type, and construct basic attribute labels and event labels according to the classification results; set rule labels according to business needs and the event labels, circle the corresponding entity data in the customer-related data based on the rule labels and generate entity labels.

[0112] The real-time data label construction unit 402 is also used to select business data from the customer-related data, construct the basic attribute label based on the business data, mark the business data based on the basic attribute label and store it in the basic data label pool; count customer events in the customer-related data, select event-related data from the customer-related data based on the customer events, construct event labels based on the event-related data, mark the event-related data based on the event labels and store them in the event label pool; select the first business data and the first event-related data that match the rule label from the basic data label pool and the event label pool based on the rule label; construct an entity label based on the first business data and the first event-related data, and save the data corresponding to the entity label.

[0113] The offline pre-calculation result set construction unit 403 is used to obtain the offline data of the customer, calculate the offline data based on the first label algorithm, and generate the offline pre-calculation result set according to the calculation result.

[0114] The offline pre-calculation result set construction unit 403 is also used to select a first algorithm model based on the business requirements, input the offline data into the first algorithm model, and construct offline mining tags; and construct the offline pre-calculation result set based on the offline mining tags and the historical data.

[0115] The real-time prediction unit 404 is used to select corresponding data based on the entity tag and the customer basic data tag, use the first tag algorithm to build a real-time mining tag based on the corresponding data and the offline pre-calculated result set, and select target data based on the real-time mining tag.

[0116] The data selection unit 405 is used to set the rule tag according to the business needs and the event tags, and establish a rule tag pool based on the rule tags; select the corresponding rule tag from the rule tag pool according to the business needs; circle the corresponding entity data in the customer-related data based on the selected rule tag and generate an entity tag, and build an entity tag pool based on the entity tag; select a first entity tag from the entity tag pool according to the business needs, and select corresponding data based on the first entity tag and the customer basic data tag.

[0117] In an exemplary embodiment, the data processing unit 401, the real-time data label construction unit 402, the offline pre-calculation result set construction unit 403, the real-time prediction unit 404 and the data selection unit 405 can be implemented by one or more central processing units (CPU, Central Processing Unit), graphics processing unit (GPU, Graphics Processing Unit), application specific integrated circuit (ASIC, Application Specific Integrated Circuit), DSP, programmable logic device (PLD, Programmable Logic Device), complex programmable logic device (CPLD, Complex Programmable Logic Device), field programmable gate array (FPGA, Field-Programmable Gate Array), general processor, controller, microcontroller (MCU, Micro Controller Unit), microprocessor (Microprocessor), or other electronic components.

[0118] Regarding the device in the above embodiment, the specific manner in which each module and unit performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0119] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device and a readable storage medium.

[0120] Figure 4A schematic block diagram of an example electronic device 800 that can be used to implement an embodiment of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0121] like Figure 4 As shown, the device 800 includes a computing unit 801, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 802 or a computer program loaded from a storage unit 808 into a random access memory (RAM) 803. In the RAM 803, various programs and data required for the operation of the device 800 can also be stored. The computing unit 801, the ROM 802, and the RAM 803 are connected to each other via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0122] A number of components in the device 800 are connected to the I / O interface 805, including: an input unit 806, such as a keyboard, a mouse, etc.; an output unit 807, such as various types of displays, speakers, etc.; a storage unit 808, such as a disk, an optical disk, etc.; and a communication unit 809, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 809 allows the device 800 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0123] The computing unit 801 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 801 performs the various methods and processes described above, such as a real-time tag processing method. For example, in some embodiments, a real-time tag processing method may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 808. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 800 via ROM 802 and / or communication unit 809. When the computer program is loaded into RAM 803 and executed by the computing unit 801, one or more steps of a real-time tag processing method described above may be performed. Alternatively, in other embodiments, the computing unit 801 may be configured to perform a real-time tag processing method in any other appropriate manner (e.g., by means of firmware).

[0124] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0125] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0126] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0127] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0128] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0129] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0130] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0131] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of the present disclosure, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0132] The above is only a specific embodiment of the present disclosure, but the protection scope of the present disclosure is not limited thereto. Any person skilled in the art who is familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present disclosure, which should be included in the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be based on the protection scope of the claims.

Claims

1. A real-time label processing method, characterized in that: The method comprises: Acquire customer data in real time, perform data cleaning on the customer data, and obtain customer-related data; Classify the customer-related data based on data types, and construct basic attribute labels and event labels according to the classification results; Setting rule tags according to business requirements and the event tags, circle corresponding entity data in the customer-related data based on the rule tags and generate entity tags; Obtaining offline data of the customer, calculating the offline data based on a first label algorithm, and generating an offline pre-calculation result set according to the calculation result; Corresponding data is selected based on the entity tag and the customer basic data tag, real-time mining tags are constructed based on the corresponding data and the offline pre-calculated result set using the first tag algorithm, and target data is circled based on the real-time mining tags.

2. The method according to claim 1, characterized in that The basic attribute labels and event labels are constructed according to the classification results, including: Selecting business data from the customer-related data, constructing the basic attribute tag based on the business data, marking the business data based on the basic attribute tag and storing the data in a basic data tag pool; The customer events in the customer-related data are counted, event-related data in the customer-related data are selected based on the customer events, event tags are constructed based on the event-related data, and the event-related data are marked based on the event tags and stored in an event tag pool.

3. The method according to claim 2, characterized in that The step of selecting corresponding entity data in the customer-related data based on the rule tag and generating an entity tag includes: Selecting first business data and first event-related data matching the rule tag from the basic data tag pool and the event tag pool based on the rule tag; An entity tag is constructed based on the first business data and the first event-related data, and data corresponding to the entity tag is saved.

4. The method according to claim 1, characterized in that: The step of inputting the customer offline data into a first algorithm model for processing to obtain an offline pre-calculation result set includes: Selecting a first algorithm model based on the business requirements, inputting the offline data into the first algorithm model, and constructing an offline mining label; The offline pre-calculation result set is constructed based on the offline mining tags and the historical data.

5. The method according to claim 1, characterized in that The method further comprises: The rule tag is set according to the business requirement and the event tag, and a rule tag pool is established based on the rule tag; According to the business requirements, select the corresponding rule tag from the rule tag pool; Based on the selected rule tag, corresponding entity data in the customer-related data is circled and an entity tag is generated, and an entity tag pool is constructed based on the entity tag; A first entity tag is selected from the entity tag pool according to the business requirement, and corresponding data is selected based on the first entity tag and the customer basic data tag.

6. A real-time label processing device, characterized in that: The device comprises: A data processing unit, used to obtain customer data in real time, clean the customer data, and obtain customer-related data; A real-time data label construction unit is used to classify the customer-related data based on data types, and respectively construct basic attribute labels and event labels according to the classification results; set rule labels according to business requirements and the event labels, circle corresponding entity data in the customer-related data based on the rule labels, and generate entity labels; An offline pre-calculation result set construction unit, used to obtain offline data of a customer, calculate the offline data based on a first label algorithm, and generate an offline pre-calculation result set according to the calculation result; The real-time prediction unit is used to select corresponding data based on the entity label and the customer basic data label, use the first label algorithm to build a real-time mining label based on the corresponding data and the offline pre-calculated result set, and select target data based on the real-time mining label.

7. The device according to claim 6, characterized in that The real-time data label construction unit is further used to select business data from the customer-related data, construct the basic attribute label based on the business data, mark the business data based on the basic attribute label and store it in the basic data label pool; Counting customer events in the customer-related data, selecting event-related data in the customer-related data based on the customer events, constructing event tags based on the event-related data, marking the event-related data based on the event tags and storing them in an event tag pool; Selecting first business data and first event-related data matching the rule tag from the basic data tag pool and the event tag pool based on the rule tag; Building an entity tag based on the first business data and the first event-related data, and saving data corresponding to the entity tag; The offline pre-calculation result set construction unit is further used to select a first algorithm model based on the business requirements, input the offline data into the first algorithm model, and construct an offline mining label; The offline pre-calculation result set is constructed based on the offline mining tags and the historical data.

8. The device according to claim 6, characterized in that The device also includes: A data selection unit is used to set the rule tag according to the business needs and the event tag, and establish a rule tag pool based on the rule tag; select the corresponding rule tag from the rule tag pool according to the business needs; circle the corresponding entity data in the customer-related data based on the selected rule tag and generate an entity tag, and build an entity tag pool based on the entity tag; select a first entity tag from the entity tag pool according to the business needs, and select corresponding data based on the first entity tag and the customer basic data tag.

9. An electronic device, characterized in that: include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 5.

10. A non-transitory computer-readable storage medium storing computer instructions, characterized in that: The computer instructions are used to make a computer execute the method according to any one of claims 1-5.

Images