Customer portrait generation method and device, medium and product

Abstract

The invention discloses a customer portrait generation method and device, a medium and a product, and relates to the application of a large model in the financial science and technology field, and the method comprises the steps: setting a feature list and a portrait granularity list, and obtaining the basic data of each business and stock customer information data; matching the business basic data according to the feature list to generate each business type list; and forming a customer list based on each business type list and the business basic data, and matching stock customer information data by using the customer list to generate a customer portrait list. By defining features and portrait granularity, it is ensured that data having great influence on customer behaviors is selected, and an accurate data basis is provided for subsequent portrait generation. By utilizing feature matching, services are automatically classified according to feature clusters, and manual subjective division deviation is avoided. The customer list is extracted through the service type, and then the high-frequency features are extracted from the stock data according to the portrait granularity to generate the customer portrait list, so that the portrait accuracy and the service suitability are improved.

Description

A method, device, medium, and product for generating customer profiles Technical Field

[0001] This invention relates to the field of big data technology, and in particular to a method, device, medium and product for generating customer profiles. Background Technology

[0002] In the financial sector, the accuracy of customer profiling is crucial for business marketing and service quality. With the development of big data and artificial intelligence technologies, banks and other financial institutions need a method to efficiently and accurately generate customer profiles in order to accurately grasp customer needs and provide personalized services.

[0003] Currently, banks primarily employ methods for recommending financial products and building customer profiles, such as pre-setting rigid rules based on expert experience, generating customer segments based on structured data, or mining potential preferences using customer-product interaction matrices through data tracking. These existing methods rely on manual rule design, are highly susceptible to subjective biases, have long rule adjustment cycles, and struggle to adapt to dynamic market changes, such as shifts in customer demand triggered by sudden economic events. Furthermore, data from a single financial scenario is insufficient to support generalized recommendations across business lines, and there is a lack of target customer profile generation models that can automatically adjust and optimize, eliminate human intervention, and support continuous improvement through feedback data. Summary of the Invention

[0004] This invention provides a customer profile generation method, device, medium, and product. By having experts define business characteristics and profile granularities, and using data training-driven methods, customer profiles directly corresponding to business characteristics are generated through business classification and model training. This achieves the elimination of human intervention and self-evolution, improving the accuracy and timeliness of profiles, and solving the problems of reliance on manual rules and poor adaptability in the traditional customer profile construction of banks.

[0005] According to one aspect of the present invention, a customer profile generation method is provided, the method comprising:

[0006] Set up a feature list and a profile granularity list, and obtain basic business data and existing customer information data. The feature list includes various feature conditions, and the profile granularity list includes customer feature dimensions.

[0007] The basic business data is matched against the feature list to generate a list of each business type.

[0008] Customer lists are generated based on lists of each business type and basic business data, and these customer lists are then matched with existing customer information data to generate customer profile lists.

[0009] Optionally, the basic business data is matched according to the feature list to generate a list of business types, including: taking each basic business data as target business data; determining the target feature conditions in the target business data that match the feature list, wherein the target feature conditions are a single feature or a combination of features; and grouping the basic business data with the same target feature conditions into the same business type to generate a list of business types.

[0010] The advantages of this setup are: it enables the systematic classification and integration of basic business data, accurately identifies business characteristics, and reduces the complexity of data processing.

[0011] Optionally, a customer list can be formed based on each business type list and business base data, including: taking each business type list as a target type list and determining the target business in the target type list; extracting the customer identifiers corresponding to the target business from the business base data, merging the customer identifiers of all target businesses under the same business type, and forming a customer list.

[0012] The advantages of this setup are: to achieve efficient integration and classification of target business customer information, to accurately collect customer identifiers, to avoid information dispersion and duplication, and to improve the efficiency of customer data management.

[0013] Optionally, the customer list is used to match existing customer information data to generate a customer profile list, including: using each customer list as a target customer list; matching the target customer list with existing customer information data to determine the target customer data; extracting each customer feature from the target customer data according to the customer feature dimensions listed in the profile granularity list; and generating a customer profile list corresponding to the target customer list based on each customer feature.

[0014] The advantage of this setup is that it enables the filtering of target customer data, the structured integration of customer characteristics, and the improvement of the accuracy of customer profile generation.

[0015] Optionally, a customer profile list corresponding to the target customer list is generated based on each customer characteristic, including: determining the target feature dimension corresponding to each customer characteristic and calculating the repetition rate of each target feature dimension; taking the target feature dimension with a repetition rate greater than the preset repetition rate as the profile feature dimension, and forming a customer profile list based on the profile feature dimension.

[0016] The advantage of this setup is that it enables the filtering and optimization of customer profile feature dimensions, removes redundant features, focuses on core customer attributes, and enhances the representativeness of customer profiles.

[0017] Optionally, the method also includes: using a temporal convolutional network to obtain user time-series behavior data from business base data; extracting short-term fluctuation and long-term trend features from the time-series behavior data, and determining sudden data based on the short-term fluctuation and long-term trend features; and removing sudden data from the business base data.

[0018] The advantage of this setup is that by identifying and removing sudden abnormal data, the quality and reliability of basic business data are improved, and the interference of abnormal data on business decisions is reduced.

[0019] Optionally, the method also includes: obtaining the business to be decided; determining whether the business to be decided is an existing business; if so, obtaining the corresponding target customer profile list based on the business type list to which the business to be decided belongs, and making promotion decisions based on the target customer profile list; otherwise, obtaining the target business type input by the user based on the business to be decided, obtaining the target customer profile list corresponding to the target business type, making preliminary decisions, and regenerating the real business type after accumulating basic data, and updating the decisions.

[0020] The advantages of this setup are: it enables decision optimization for different types of business processes requiring decision-making; it allows for direct matching of target customer profiles to existing businesses, improving the targeting and conversion rate of promotional decisions; and it enables rapid decision-making, initiation, and dynamic optimization for new businesses, shortening the business launch cycle, reducing the cost of trial and error in decision-making, and improving the overall efficiency and quality of business decision-making.

[0021] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0022] At least one processor;

[0023] and a memory communicatively connected to the at least one processor;

[0024] The memory stores a computer program that can be executed by the at least one processor, which is then executed by the at least one processor to enable the at least one processor to perform a customer profile generation method according to any embodiment of the present invention.

[0025] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement a customer profile generation method according to any embodiment of the present invention.

[0026] According to another aspect of the present invention, a computer program product is provided, the computer program product comprising a computer program that, when executed by a processor, implements a customer profile generation method according to any embodiment of the present invention.

[0027] The technical solution of this invention, by defining features and profile granularity, ensures the selection of data that has a significant impact on customer behavior, providing an accurate data foundation for subsequent profile generation. Utilizing feature matching, businesses are automatically categorized by feature clusters, avoiding biases from subjective human classification. The generated business types reflect stable feature preferences, providing an objective classification basis for subsequent customer group association. Customer groups are extracted through business types, and then high-frequency features are extracted from existing data based on profile granularity. Data-driven feature filtering eliminates human error, and the generated profiles accurately reflect the commonalities of target customers, improving profile accuracy and business adaptability.

[0028] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0029] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0030] Figure 1 is a flowchart of a customer profile generation method according to Embodiment 1 of the present invention;

[0031] Figure 2 is a flowchart of another customer profile generation method provided according to Embodiment 2 of the present invention;

[0032] Figure 3 is a schematic diagram of a customer profile generation device according to Embodiment 3 of the present invention;

[0033] Figure 4 is a schematic diagram of the structure of an electronic device that implements a customer profile generation method according to an embodiment of the present invention. Detailed Implementation

[0034] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. 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 should fall within the scope of protection of the present invention.

[0035] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0036] Example 1

[0037] Figure 1 is a flowchart of a customer profile generation method provided in Embodiment 1 of the present invention. This embodiment is applicable to the scenario of determining target customers and involves the application of large models in the field of financial technology. The method can be executed by a customer profile generation device, which can be implemented in hardware and / or software and can be configured in a computer controller.

[0038] As shown in Figure 1, the method includes:

[0039] S110. Set up a feature list and a profile granularity list, and obtain basic business data and existing customer information data. The feature list includes various feature conditions, and the profile granularity list includes customer feature dimensions.

[0040] The feature list refers to a set of feature categories defined by experts, including data feature conditions that have a significant impact on customer behavior preferences, such as specific feature conditions like yield, commission rate, loss rate, and lock-in period, used for subsequent matching and classification of basic business data. The profile granularity list is also defined by experts, including customer data feature dimensions that have a significant impact on customer behavior styles and are owned by the bank, such as gender and age, used to define the information extraction dimensions when generating customer profiles. Basic business data refers to the basic data information of various business transactions provided by business personnel, which can be used as the basis for business type training and customer profile training. The scope is defined by business personnel, prioritizing recent data to ensure timeliness. Existing customer information data refers to existing customer information data, used as the object of customer information retrieval when generating customer profiles, covering the customer data involved in the profile granularity list.

[0041] Specifically, the feature list can be defined by experts based on business needs, specifying data feature conditions that significantly influence customer behavior preferences. Experts can then input these feature conditions into the business classification module of the large model, forming a feature list for subsequent business classification. The profile granularity list can be created by experts selecting feature dimensions from existing customer data that play a decisive role in customer behavior styles, such as dividing age into ranges like 21-25 years old, 26-30 years old, etc. These customer feature dimensions are then input into the model training module, forming a profile granularity list. Business personnel can input basic data from various business transactions into the business classification module to form basic business data, which can be named "Transaction Data Table - xxx Business," used for business type training and customer profile training. Similarly, business personnel can input existing customer information data into the model training module to form an existing customer information data table, serving as a source for extracting customer profile information.

[0042] It should be noted that the information collected above is information and data authorized by the user or fully authorized by all parties. The collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data all comply with the relevant laws, regulations and standards of the relevant countries and regions, necessary confidentiality measures have been taken, and they do not violate public order and good morals. Corresponding operation portals are provided for users to choose to authorize or refuse.

[0043] Optionally, the method also includes: using a temporal convolutional network to obtain user time-series behavior data from business base data; extracting short-term fluctuation and long-term trend features from the time-series behavior data, and determining sudden data based on the short-term fluctuation and long-term trend features; and removing sudden data from the business base data.

[0044] Optionally, user time-series behavior data can be obtained from basic business data using a temporal convolutional network, including: optimizing the parameters of the temporal convolutional network through the transfer learning capability of a large model to adapt it to the extraction of time-series features from financial business data, and obtaining user time-series behavior data containing short-term fluctuations and long-term trends.

[0045] Temporal convolutional networks (TCNNs) are deep learning models that can model time-series data using causal convolution and dilated convolution. TCNNs can read basic business data and arrange it chronologically to form a sequence of user behavior data, thereby capturing dynamic patterns of data change over time. Specifically, TCNNs can identify sudden, non-persistent behavioral patterns in the short term by analyzing changes in recent data points within a time series. For example, by calculating data variance and peak deviation within a sliding time window, short-term abnormal fluctuations can be located. Furthermore, TCNNs can also capture long-term regular patterns in time series. For example, by identifying periodicity and trend characteristics in the data, a mathematical description of long-term trends can be formed. Then, by combining short-term fluctuations and long-term trend characteristics, sudden data events can be identified.

[0046] Specifically, when short-term fluctuations significantly deviate from long-term trends, and such fluctuations have not exhibited similar patterns in historical long-term trends, they are classified as sudden data. Similarly, for short-term peaks that do not conform to long-term periodic patterns, statistical tests can be used to determine whether they are low-probability events; if so, they are classified as sudden data. Furthermore, when short-term fluctuations appear isolated in a time series and lack continuity, if they are unrelated to long-term trends, they can also be classified as sudden data.

[0047] Furthermore, after identifying sudden data, the controller can remove it from the original business data. After removal, the remaining data better reflects users' stable behavioral preferences and business characteristics, ensuring that the subsequently generated business type lists and customer profile lists are based on real and continuous behavioral patterns, thus improving the accuracy and reliability of the model.

[0048] S120. Match the basic business data according to the feature list to generate a list of each business type.

[0049] The business type list refers to the list generated after matching the basic business data according to the feature list and grouping businesses with the same feature clusters into the same business type. Its quantity is automatically generated by the large model based on the feature list and the entered basic data. It does not need to be named, and the obviousness of the features is only indicated by the occurrence of the features.

[0050] Optionally, the basic business data is matched according to the feature list to generate a list of business types, including: taking each basic business data as target business data; determining the target feature conditions in the target business data that match the feature list, wherein the target feature conditions are a single feature or a combination of features; and grouping the basic business data with the same target feature conditions into the same business type to generate a list of business types.

[0051] Optionally, the target feature conditions in the target business data that match the feature list are determined, including: using a multimodal matching algorithm of a large model to perform cross-dimensional association matching on the text and numerical features of the target business data, and identifying a single feature or feature combination as the target feature condition.

[0052] In one specific implementation, the controller uses the basic data of each business as the target business data. The business classification module of the large model can read the feature list and, based on the preset conditions of the feature list, extract the corresponding attributes of the target business data one by one for comparison. For example, if the feature list contains conditions such as "annual return rate < a%" or "annual return rate between a% and b%", the large model will determine whether each business meets the feature conditions. Then, the large model will classify businesses with the same feature clusters into the same business type. For example, a certain type of business that meets both "annual return rate < a%" and "no handling fee" will be classified into one type; another type that only meets "annual return rate between a% and b%" will be classified into another type. The number of business types is automatically generated by the model based on feature combinations and data. It may be a single feature type, such as only feature 1, or it may be a multi-feature combination type, such as feature 1 + feature 2.

[0053] S130. Generate customer lists based on the lists of each business type and the basic business data, and use the customer lists to match the existing customer information data to generate a customer profile list.

[0054] The customer list refers to a list of customers extracted from all business data within each business type, based on the business type list. This list is used to subsequently match existing customer information data to generate customer profiles. The customer profile list is a list created by matching the customer list with existing customer information data, extracting customer information according to the feature dimensions listed in the profile granularity list, and selecting features with high repetition rates. This achieves a direct correspondence between business features and customer profiles, reflecting the profiles of various characteristic preference groups.

[0055] Specifically, the model training module can synchronize basic business data from the business classification module. Based on the generated business type list, it extracts all customers corresponding to each business type. For example, if business type 1 includes businesses a, b, and c, then customer lists a, b, and c are extracted from these three business types and merged to form customer list 1.

[0056] Furthermore, the controller can use Customer List 1 to match the existing customer information data table, extract customer information according to the dimensions of the profile granularity list, and calculate the repetition rate of each feature, selecting features with high repetition rates as core elements of the profile. For example, if most customers in Customer List 1 are male and aged 21-25, then Customer Profile List 1 is generated: Gender: Male, Age: 21-25, realizing a direct association between business characteristics and customer group profiles, forming accurate profiles of people with each feature preference.

[0057] Optionally, a customer list can be formed based on each business type list and business base data, including: taking each business type list as a target type list and determining the target business in the target type list; extracting the customer identifiers corresponding to the target business from the business base data, merging the customer identifiers of all target businesses under the same business type, and forming a customer list.

[0058] The controller treats each business type list as a target type list. For example, when the business type list contains categories such as "Business Type 1: Only has Feature 1" and "Business Type 2: Only has Feature 2," the controller will process each business type list as a processing object. Taking Business Type 1 as an example, it may contain specific target businesses such as Business a, Business b, and Business c. The target businesses are formed by matching and classifying the basic business data according to the feature list, and each target business corresponds to a set of business data with the same feature conditions.

[0059] Furthermore, the controller extracts customer identifiers corresponding to the target business from the business foundation data. The business foundation data stores detailed information for each business, including the customer identifiers participating in the business. For each target business in the target type list, the controller retrieves all records corresponding to that business from the business foundation data, extracts the customer identifiers, and merges the customer identifiers of all target businesses under the same business type to form a customer list. Taking business type 1 as an example, by deduplicating and merging the customer identifiers corresponding to business a, business b, and business c, customer list 1 is finally obtained, which includes customer A, customer B, and customer C. By associating business types with customer identifiers, customer groups with similar business characteristic preferences are aggregated together, providing a data foundation for subsequent customer profile generation. During merging, the system automatically removes duplicate customer identifiers, ensuring that each customer appears only once in the list, thus accurately reflecting the customer group corresponding to that business type.

[0060] Optionally, the method also includes: obtaining the business to be decided; determining whether the business to be decided is an existing business; if so, obtaining the corresponding target customer profile list based on the business type list to which the business to be decided belongs, and making promotion decisions based on the target customer profile list; otherwise, obtaining the target business type input by the user based on the business to be decided, obtaining the target customer profile list corresponding to the target business type, making preliminary decisions, and regenerating the real business type after accumulating basic data, and updating the decisions.

[0061] When processing pending business transactions, the controller first determines whether it is an existing business. Existing businesses refer to those that already exist in the business system and have accumulated a certain amount of transaction data. If the determination result is an existing business, the controller will obtain the corresponding target customer profile list based on the business type list to which the pending business belongs. The business type list is generated after matching the basic business data using a feature list, and each business type corresponds to a specific customer profile. For example, if an existing business belongs to business type 1, and the customer profile list for business type 1 shows that the customers for this type of business are mainly men aged 21-25, then the controller will obtain the corresponding target customer profile list and make promotion decisions based on this target customer profile list, such as determining which customers to promote the business to and how to formulate promotion strategies.

[0062] Conversely, if the business to be decided is determined to be a new business rather than an existing one, the controller cannot directly determine its business type because there is no existing transaction data for the new business. In this case, the controller will acquire users based on the business to be decided, which is the target business type input by the business expert. After acquiring the target business type, the controller will obtain a list of target customer profiles corresponding to that target business type, and then make preliminary decisions based on the target customer profile list, such as determining the initial target customer group. It is important to note that the preliminary decision for a new business is only based on the business type determined manually and the corresponding customer profile list. After the new business is launched, a certain amount of transaction data, i.e., basic business data, will be accumulated. Once enough basic data has been accumulated, the controller will input the accumulated basic business data into a large model for training, regenerating the actual business type of the business. After generating the actual business type, the controller will update the corresponding customer profile list, thereby updating the promotion decision to make the decision more in line with the actual situation. By utilizing existing business types and customer profile lists to guide business promotion decisions, and for new businesses, making preliminary decisions through manual judgment first, and then optimizing based on actual data accumulation, the accuracy and effectiveness of the decisions are improved.

[0063] The technical solution of this invention, by defining features and profile granularity, ensures the selection of data that has a significant impact on customer behavior, providing an accurate data foundation for subsequent profile generation. Utilizing feature matching, businesses are automatically categorized by feature clusters, avoiding biases from subjective human classification. The generated business types reflect stable feature preferences, providing an objective classification basis for subsequent customer group association. Customer groups are extracted through business types, and then high-frequency features are extracted from existing data based on profile granularity. Data-driven feature filtering eliminates human error, and the generated profiles accurately reflect the commonalities of target customers, improving profile accuracy and business adaptability.

[0064] Example 2

[0065] Figure 2 is a flowchart of a customer profile generation method provided in Embodiment 2 of the present invention. This embodiment adds a specific process of using a customer list to match existing customer information data to generate a customer profile list, based on Embodiment 1. The specific content of steps S210-S220 is largely the same as steps S110-S120 in Embodiment 1, and therefore will not be described again in this embodiment. As shown in Figure 2, the method includes:

[0066] S210. Set up a feature list and a profile granularity list, and obtain basic business data and existing customer information data. The feature list includes various feature conditions, and the profile granularity list includes customer feature dimensions.

[0067] Optionally, the method also includes: using a temporal convolutional network to obtain user time-series behavior data from business base data; extracting short-term fluctuation and long-term trend features from the time-series behavior data, and determining sudden data based on the short-term fluctuation and long-term trend features; and removing sudden data from the business base data.

[0068] S220. Match the basic business data according to the feature list to generate a list of each business type.

[0069] Optionally, the basic business data is matched according to the feature list to generate a list of business types, including: taking each basic business data as target business data; determining the target feature conditions in the target business data that match the feature list, wherein the target feature conditions are a single feature or a combination of features; and grouping the basic business data with the same target feature conditions into the same business type to generate a list of business types.

[0070] S230. Customer lists are formed based on the lists of each business type and the basic business data, respectively.

[0071] Optionally, a customer list can be formed based on each business type list and business base data, including: taking each business type list as a target type list and determining the target business in the target type list; extracting the customer identifiers corresponding to the target business from the business base data, merging the customer identifiers of all target businesses under the same business type, and forming a customer list.

[0072] S240. Each customer list is used as a target customer list.

[0073] S250. Use the target customer list to match existing customer information data to determine target customer data.

[0074] Specifically, the customer list is a set of customer identifiers extracted from the business base data based on the business type list. For example, it might be a list of customer identifiers corresponding to all businesses under a specific business type. Taking Customer List 1 as an example, it contains the identifiers for Customer A, Customer B, and Customer C. Then, the controller uses the target customer list to match existing customer information data to determine the target customer data. Existing customer information data is stored in an existing customer information data table, which contains detailed characteristic dimensions of the customers. The controller can retrieve the complete information of all customers in the target customer list by searching this data table using the customer identifiers.

[0075] S260. Extract each customer feature from the target customer data according to the customer feature dimensions listed in the profile granularity list.

[0076] Specifically, the controller can extract various customer characteristics from the target customer data according to the customer characteristic dimensions listed in the profile granularity checklist. The profile granularity checklist is predefined by experts and includes dimensions that have a significant impact on customer behavior styles. For example, if the dimensions defined in the checklist are gender and age range, the controller will extract the specific values ​​corresponding to gender and age range from each target customer's data, that is, extract customer A's gender as "male" and age range as "21-25 years old".

[0077] S270. Generate a customer profile list corresponding to the target customer list based on each customer's characteristics.

[0078] Optionally, a customer profile list corresponding to the target customer list is generated based on each customer characteristic, including: determining the target feature dimension corresponding to each customer characteristic and calculating the repetition rate of each target feature dimension; taking the target feature dimension with a repetition rate greater than the preset repetition rate as the profile feature dimension, and forming a customer profile list based on the profile feature dimension.

[0079] Specifically, the controller determines the target feature dimensions corresponding to each customer characteristic and calculates the repetition rate of each target feature dimension. The controller extracts the specific value of each feature dimension for each customer data point in the target customer list and then performs statistical analysis. For example, the controller first extracts the gender characteristics of all customers, counts the number of males and females, calculates the repetition rate of males and females in this group, and then performs statistical analysis on age ranges, such as the proportion of customers in each range (21-25 years old, 26-30 years old, etc.), obtaining the repetition rate for each age range. Finally, the controller selects target feature dimensions with a repetition rate greater than a preset repetition rate as profile feature dimensions. The preset repetition rate is a threshold set by the system or business requirements to determine which feature dimensions are sufficiently representative. For example, the preset repetition rate can be set to 70%. If the repetition rate of males in the gender dimension is 80%, which is greater than 70%, then males will be selected as a profile feature dimension; similarly, the repetition rate of "21-25 years old" in the age range is 86%, also greater than 70%, and will also be selected. Finally, the controller combines the selected profile feature dimensions to form a customer profile list. By statistically analyzing data, we identify significant common characteristics among the target customer group, thereby accurately portraying the group's profile. By setting a repetition rate threshold, we ensure that the characteristics included in the profile truly reflect the group's mainstream features, avoiding interference from isolated exceptions and ensuring the generated customer profile list has high accuracy and practicality.

[0080] Optionally, the method also includes: obtaining the business to be decided; determining whether the business to be decided is an existing business; if so, obtaining the corresponding target customer profile list based on the business type list to which the business to be decided belongs, and making promotion decisions based on the target customer profile list; otherwise, obtaining the target business type input by the user based on the business to be decided, obtaining the target customer profile list corresponding to the target business type, making preliminary decisions, and regenerating the real business type after accumulating basic data, and updating the decisions.

[0081] The technical solution of this invention provides a foundation for accurately locating target customers by clearly defining the customer group scope. It accurately associates detailed customer information with customer identifiers, ensuring the completeness and accuracy of the acquired customer data and providing a rich data source for profile generation. Data is extracted based on preset key feature dimensions, ensuring the targeted and effective extraction of customer features and avoiding interference from irrelevant information. Customer profiles are generated through a data-driven approach, achieving a direct correspondence between business characteristics and customer profiles, avoiding misjudgments by manual assessment, and improving profile accuracy.

[0082] Example 3

[0083] Figure 3 is a schematic diagram of a customer profile generation device provided in Embodiment 3 of the present invention. As shown in Figure 3, the device includes: a list setting and business data acquisition module 310, used to set a feature list and a profile granularity list, and acquire basic business data and existing customer information data, wherein the feature list includes various feature conditions, and the profile granularity list includes customer feature dimensions;

[0084] The business type list generation module 320 is used to match the basic business data according to the feature list to generate a list of each business type.

[0085] The customer profile list generation module 330 is used to generate customer lists based on lists of various business types and basic business data, and to match the customer lists with existing customer information data to generate customer profile lists.

[0086] Optionally, the business type list generation module 320 is specifically used to: take each business basic data as target business data; determine the target feature conditions in the target business data that match the feature list, wherein the target feature conditions are a single feature or a combination of features; classify the business basic data with the same target feature conditions into the same business type, and generate a business type list.

[0087] Optionally, the customer profile list generation module 330 specifically includes: a customer list generation unit, used to: take each business type list as a target type list, determine the target business in the target type list; extract the customer identifier corresponding to the target business from the business basic data, merge the customer identifiers of all target businesses under the same business type, and form a customer list.

[0088] Optionally, the customer profile list generation module 330 specifically includes: a profile list generation unit, which further includes: a target customer list determination subunit, used to: use each customer list as a target customer list; a target customer data determination subunit, used to: use the target customer list to match existing customer information data to determine the target customer data; a customer feature extraction subunit, used to: extract each customer feature from the target customer data according to the customer feature dimensions listed in the profile granularity list; and a profile list generation subunit, used to: generate a customer profile list corresponding to the target customer list based on each customer feature.

[0089] Optionally, the profile list generation sub-unit is used to: determine the target feature dimensions corresponding to each customer feature and calculate the repetition rate of each target feature dimension; take the target feature dimensions with a repetition rate greater than the preset repetition rate as profile feature dimensions, and form a customer profile list based on the profile feature dimensions.

[0090] Optionally, the device further includes: a burst data removal module, used to: obtain user time-series behavior data from business basic data using a temporal convolutional network; extract short-term fluctuation and long-term trend features from the time-series behavior data, determine burst data based on the short-term fluctuation and long-term trend features; and remove burst data from business basic data.

[0091] Optionally, the device further includes a business decision module, used to: acquire the business to be decided; determine whether the business to be decided is an existing business; if so, acquire the corresponding target customer profile list based on the business type list to which the business to be decided belongs, and make a promotion decision based on the target customer profile list; otherwise, acquire the target business type input by the user based on the business to be decided, acquire the target customer profile list corresponding to the target business type, make a preliminary decision, and regenerate the real business type after accumulating basic data, and update the decision.

[0092] The technical solution of this invention, by defining features and profile granularity, ensures the selection of data that has a significant impact on customer behavior, providing an accurate data foundation for subsequent profile generation. Utilizing feature matching, businesses are automatically categorized by feature clusters, avoiding biases from subjective human classification. The generated business types reflect stable feature preferences, providing an objective classification basis for subsequent customer group association. Customer groups are extracted through business types, and then high-frequency features are extracted from existing data based on profile granularity. Data-driven feature filtering eliminates human error, and the generated profiles accurately reflect the commonalities of target customers, improving profile accuracy and business adaptability.

[0093] The customer profile generation device provided in this embodiment of the invention can execute a customer profile generation method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0094] Example 4

[0095] Figure 4 illustrates a schematic diagram of an electronic device 10 that can be used to implement embodiments of the present invention. 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 assistants, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0096] As shown in Figure 4, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer programs stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0097] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0098] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as a customer profile generation method.

[0099] In some embodiments, a customer profile generation method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the customer profile generation method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform a customer profile generation method by any other suitable means (e.g., by means of firmware).

[0100] Various embodiments 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-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0101] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0102] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0103] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device 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 provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, 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 sound input, voice input, or tactile input).

[0104] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0105] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system. It addresses the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0106] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0107] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for generating customer profiles, characterized in that, include: Set up a feature list and a profile granularity list, and obtain basic business data and existing customer information data. The feature list includes various feature conditions, and the profile granularity list includes customer feature dimensions. Match the basic business data according to the feature list to generate a list of business types. Form a customer list based on each business type list and the basic business data, and use the customer list to match the existing customer information data to generate a customer profile list.

2. The method according to claim 1, characterized in that, The step of matching the basic business data according to the feature list to generate a list of business types includes: taking each basic business data as target business data; determining the target feature conditions in the target business data that match the feature list, wherein the target feature conditions are a single feature or a combination of features; and grouping basic business data with the same target feature conditions into the same business type to generate a list of business types.

3. The method according to claim 2, characterized in that, The step of forming a customer list based on each of the business type lists and the business basic data includes: taking each of the business type lists as a target type list, determining the target business in the target type list; extracting the customer identifier corresponding to the target business from the business basic data, merging the customer identifiers of all target businesses under the same business type, and forming a customer list.

4. The method according to claim 1, characterized in that, The step of using customer lists to match existing customer information data to generate customer profile lists includes: using each customer list as a target customer list; matching the target customer lists with existing customer information data to determine target customer data; extracting each customer feature from the target customer data according to the customer feature dimensions listed in the profile granularity list; and generating a customer profile list corresponding to the target customer lists based on each of the customer features.

5. The method according to claim 4, characterized in that, The step of generating a customer profile list corresponding to the target customer list based on each of the customer characteristics includes: determining the target feature dimension corresponding to each of the customer characteristics and calculating the repetition rate of each target feature dimension; taking the target feature dimension with a repetition rate greater than a preset repetition rate as the profile feature dimension, and forming a customer profile list based on the profile feature dimension.

6. The method according to claim 1, characterized in that, The method further includes: using a temporal convolutional network to obtain user time-series behavior data from business basic data; extracting short-term fluctuation and long-term trend features from the time-series behavior data, determining sudden data based on the short-term fluctuation and long-term trend features; and removing the sudden data from the business basic data.

7. The method according to claim 1, characterized in that, The method further includes: acquiring the business to be decided; determining whether the business to be decided is an existing business; if so, acquiring a corresponding target customer profile list based on the business type list to which the business to be decided belongs, and making a promotion decision based on the target customer profile list; otherwise, acquiring the target business type input by the user based on the business to be decided, acquiring a target customer profile list corresponding to the target business type, making a preliminary decision, and regenerating the real business type after accumulating basic data, and updating the decision.

8. An electronic device, characterized in that, The electronic device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-7.

9. A computer storage medium, characterized in that, The computer storage medium stores computer instructions that are used to cause a processor to execute the method of any one of claims 1-7.

10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the method according to any one of claims 1-7.

Images