Product recommendation customer group determination method and device, storage medium and electronic equipment

Abstract

The invention discloses a recommendation customer group determination method and device of a product, a storage medium and electronic equipment. Relates to the field of artificial intelligence, and comprises the following steps: obtaining related data of each customer, the related data comprising at least one of attribute data, transaction data and target behavior data; inputting the related data of the customer into a label generation model to obtain a group of labels of the customer; determining a label matching rule of the target product, and matching with a group of labels through the label matching rule to obtain a matching result; and under the condition that the matching result represents that the matching is successful, determining the customer as a to-be-recommended customer, and defining a set of all to-be-recommended customers as a recommended customer group of the target product. Through the method and the device, the problem that the determined recommended customer group of the product is inaccurate due to the fact that the matching rule of the product and the recommended customer is single in the related technology is solved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence, and more specifically, to a method, device, storage medium, and electronic device for determining a recommended customer group for a product. Background Art

[0002] In related technologies, in order to recommend products to users in order to expand the customer base and increase transaction volume, companies will conduct big data analysis on users through their transaction data to form corresponding customer portraits, and then screen target customer portraits related to the products to be promoted from a large number of customer portraits, and push products to customers corresponding to the target customer portraits to promote transactions.

[0003] However, the matching rules for obtaining recommended customers through customer portrait matching in related technologies are single, and recommended customers are screened only based on whether a certain aspect of the customer's characteristics meets the requirements, resulting in inaccurate matching of target customer portraits and inability to achieve accurate product recommendations.

[0004] Currently, no effective solution has been proposed to address the problem in the related art that the recommended customer base for a product is inaccurate due to the single matching rule between products and recommended customers. Summary of the invention

[0005] The main purpose of the present application is to provide a method, device, storage medium and electronic device for determining a recommended customer group for a product, so as to solve the problem in the related art that the recommended customer group of a product is inaccurate due to the single matching rule between products and recommended customers.

[0006] In order to achieve the above-mentioned purpose, according to one aspect of the present application, a method for determining a recommended customer group for a product is provided. The method comprises: obtaining relevant data of each customer, wherein the relevant data comprises at least one of the following: attribute data, transaction data and target behavior data; inputting the relevant data of the customer into a label generation model to obtain a set of labels for the customer, wherein the label generation model is trained by multiple sets of training samples, each set of training samples comprises historical relevant data and historical labels, a set of labels comprises multiple labels, and each label is used to characterize a transaction feature of the customer; determining a label matching rule for the target product, matching a set of labels through the label matching rule, and obtaining a matching result; when the matching result indicates a successful match, determining the customer as a customer to be recommended, and defining the set of all customers to be recommended as the recommended customer group for the target product.

[0007] Optionally, matching a group of tags through a tag matching rule to obtain a matching result includes: for each tag in a group of tags, determining an evaluation value of the tag from the tag matching rule, wherein the tag matching rule contains a mapping relationship between each tag and the evaluation value; calculating the sum of the evaluation values ​​of all tags in a group of tags to obtain a matching evaluation value; when the matching evaluation value is greater than or equal to an evaluation value threshold, determining the matching result as a successful match; when the matching evaluation value is less than the evaluation value threshold, determining the matching result as a failed match.

[0008] Optionally, before determining the label matching rules for the target product, the method also includes: determining the product type and product introduction of the target product, and determining the target conditions that need to be met by the recommended customers through the product type and product introduction; setting an evaluation value for each label mapping according to the target conditions, and constructing the label matching rules for the target product based on all labels and the evaluation values ​​of each label mapping.

[0009] Optionally, the label generation model is trained in the following manner: obtain multiple historical labels set by the target organization for customers, and determine the historical relevant data of each customer; determine the historical label of each customer and the customer's historical relevant data as a set of training samples to obtain multiple groups of training samples; train the machine learning model based on the multiple groups of training samples to obtain a label generation model.

[0010] Optionally, after obtaining the recommended customer group for the target product, the method further includes: for each customer in the recommended customer group, detecting the customer's target behavior data; if a first behavior that affects the purchase intention of the target product is detected in the target behavior data, removing the customer from the recommended customer group.

[0011] Optionally, after obtaining the recommended customer group for the target product, the method further includes: for each customer who does not belong to the recommended customer group, detecting the customer's target behavior data; and adding the customer to the recommended customer group when a second behavior that affects the purchase intention of the target product is detected in the target behavior data.

[0012] Optionally, after obtaining the recommended customer group for the target product, the method also includes: obtaining the product recommendation result for each customer in the recommended customer group; determining the number of customers for whom the product recommendation result represents a successful recommendation, obtaining the number of customers, and calculating the ratio of the number of customers to the total number of the recommended customer group to obtain the recommendation success rate; when the recommendation success rate is less than the success rate threshold, updating the label matching rule of the target product based on the label of the recommended customer group.

[0013] In order to achieve the above-mentioned purpose, according to another aspect of the present application, a device for determining a recommended customer group for a product is provided. The device includes: an acquisition unit, which is used to acquire relevant data of each customer, wherein the relevant data includes at least one of the following: attribute data, transaction data, and target behavior data; an input unit, which is used to input the relevant data of the customer into a label generation model to obtain a set of labels for the customer, wherein the label generation model is trained by multiple groups of training samples, each group of training samples includes historical relevant data and historical labels, a set of labels includes multiple labels, and each label is used to characterize a transaction feature of the customer; a first determination unit, which is used to determine the label matching rule of the target product, and match a set of labels through the label matching rule to obtain a matching result; a second determination unit, which is used to determine the customer as a customer to be recommended when the matching result indicates a successful match, and define the set of all customers to be recommended as the recommended customer group for the target product.

[0014] In an embodiment of the present application, the relevant data of each customer is obtained, wherein the relevant data includes at least one of the following: attribute data, transaction data and target behavior data; the relevant data of the customer is input into a label generation model to obtain a set of labels for the customer, wherein the label generation model is trained by multiple groups of training samples, each group of training samples includes historical relevant data and historical labels, a set of labels includes multiple labels, and each label is used to characterize a transaction feature of the customer; a label matching rule for the target product is determined, and a matching result is obtained by matching a set of labels through the label matching rule; when the matching result indicates a successful match, the customer is determined as a customer to be recommended, and the set of all customers to be recommended is defined as a recommended customer group for the target product. The customer's label is constructed through the customer-related data, and the customer's label is matched based on the product's label matching rule, and then the successfully matched customers to be recommended are screened out, thereby achieving the purpose of enriching the matching rules between the product and the customer to be recommended, thereby achieving the technical effect of improving the accuracy of determining the recommended customer group for the product, and thus solving the technical problem of inaccurate recommended customer group for the determined product due to the single matching rule between the product and the recommended customer. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0016] Figure 1 A hardware structure block diagram of a computer terminal (or mobile device) for implementing a method for determining a recommended customer group for a product is shown;

[0017] Figure 2 is a flow chart of a method for determining a recommended customer group for a product provided in an embodiment of the present application;

[0018] Figure 3 is a schematic diagram of a device for determining a recommended customer group for a product provided in an embodiment of the present application;

[0019] Figure 4 It is a structural block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0020] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0021] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0022] It should be noted that the collected information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for display, data for analysis, etc.) involved in this application are information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of relevant data are in compliance with relevant laws, regulations and standards, necessary confidentiality measures are taken, and public order and good customs are not violated, and corresponding operation entrances are provided for users to choose to authorize or refuse. For example, an interface is set up between this system and relevant users or institutions to provide users with corresponding operation entrances for users to choose to agree or refuse the results of automated decision-making; if the user chooses to refuse, the expert decision-making process will be entered.

[0023] Example 1

[0024] According to an embodiment of the present application, an embodiment of a method for determining a recommended customer group for a product is also provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0025] The method embodiment provided in the first embodiment of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Figure 1 The hardware structure block diagram of a computer terminal (or mobile device) for implementing a method for determining a recommended customer group for a product is shown. Figure 1 As shown, the computer terminal 10 (or mobile device) may include one or more (shown as 102a, 102b, ..., 102n in the figure) processors 102 (the processor 102 may include but is not limited to a processing device such as an MCU (Microcontroller Unit) or an FPGA (Field-Programmable Gate Array), a memory 104 for storing data, and a transmission device 106 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a USB (Universal Serial Bus) port (which may be included as one of the ports of a BUS (Business)), a network interface, a power supply and / or a camera. It will be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0026] It should be noted that the one or more processors 102 and / or other data processing circuits described above may generally be referred to herein as "data processing circuits". The data processing circuits may be embodied in whole or in part as software, hardware, firmware, or any other combination thereof. In addition, the data processing circuit may be a single independent processing module, or may be incorporated in whole or in part into any of the other components in the computer terminal 10 (or mobile device). As described in the embodiments of the present application, the data processing circuit acts as a processor control (e.g., selection of a variable resistor terminal path connected to an interface).

[0027] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the method for determining the recommended customer group of the product in the embodiment of the present application. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, the method for determining the recommended customer group of the product described above is realized. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the computer terminal 10 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0028] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the computer terminal 10. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0029] The display may be, for example, a touch screen liquid crystal display that enables a user to interact with a user interface of the computer terminal 10 (or mobile device).

[0030] In the above operating environment, the present application provides a method for determining a recommended customer group for a product. Figure 2 is a flow chart of a method for determining a recommended customer group for a product provided in an embodiment of the present application. Figure 2 As shown, the method includes:

[0031] Step S201, obtaining relevant data of each customer, wherein the relevant data includes at least one of the following: attribute data, transaction data and target behavior data.

[0032] In step S201, the attribute data in the relevant data may include the identity information of the customer, such as occupation, etc.; the transaction data reflects the specific interaction between the customer and the financial institution, such as the purchase and redemption of financial products, the application and repayment of loans, the use record of credit cards, etc. The target behavior data may be the evaluation and feedback of users on financial products in the application of the financial institution.

[0033] Step S202, inputting the customer's relevant data into the label generation model to obtain a set of labels for the customer, wherein the label generation model is trained by multiple sets of training samples, each set of training samples includes historical relevant data and historical labels, a set of labels includes multiple labels, and each label is used to characterize a transaction feature of the customer.

[0034] In step S202, in order to generate labels that can fully reflect customer characteristics, a label generation model based on deep learning is constructed. The model can automatically identify and generate a set of labels for customers by performing multi-level feature extraction and learning on customer historical related data. The current relevant data of the customer to be analyzed is input into the label generation model, and the model will output a set of labels. In addition, the label can include the specific type of the customer's hobbies. This set of labels can dynamically reflect the customer's transaction characteristics and potential needs. For example, for a group of customers with labels such as "high net worth customers", "wealth management product preferences" and "frequent traders", financial institutions can push customized wealth management products in a targeted manner to improve conversion rates.

[0035] Step S203, determining a tag matching rule for the target product, matching the tag matching rule with a group of tags, and obtaining a matching result.

[0036] In step S203, by analyzing the characteristics of the target product, the characteristics of the target product are determined, including but not limited to product type (such as savings, loans, financial management), risk level, expected return, term, customer suitability, etc. Based on the characteristics of the target product, a reasonable matching logic, that is, a label matching rule, is pre-established. For example, for high-risk, high-return financial products, the matching rules tend to select customer labels of "high risk tolerance", "high net worth" and "active investment". In order to quantify the degree of matching, a threshold for label matching needs to be set. For example, if the target customer needs to have at least 70% matching to be considered a potential customer, then a corresponding threshold needs to be set to filter the matching results. The formulated label matching rules are integrated into the rule base, which can store and manage the matching rules of all products to facilitate subsequent matching operations and rule adjustments.

[0037] After determining the tag matching rules for the target product, the customer's set of tags is compared with the tag matching rules for the target product in the rule library one by one. For each rule, the matching evaluation value between each of the customer's tags and the rule is calculated. According to the set evaluation value threshold, it is determined whether the customer meets the matching criteria of the target product. If the matching evaluation value between the customer's set of tags and the rule exceeds the evaluation value threshold, the customer can be determined as a potential target customer, that is, a customer to be recommended.

[0038] Step S204, when the matching result indicates a successful match, the customer is determined as a customer to be recommended, and the set of all customers to be recommended is defined as a recommended customer group for the target product.

[0039] In step S204, when the matching result indicates a successful match, that is, a set of tags of the customer is highly consistent with the tag matching rules of the target product, indicating that the customer has a high potential purchase intention or meets the product positioning. For customers whose matching results indicate a successful match, they are identified as customers to be recommended. This identification can be stored in the customer database so that subsequent product recommendation activities can quickly locate these customers. Collect all data identified as customers to be recommended to form a set of recommended customer groups. The recommended customer group includes all customers who have successfully matched the target product, and these customers are regarded as the main potential buyers of the target product.

[0040] The method for determining the recommended customer group for a product provided in an embodiment of the present application obtains relevant data of each customer, wherein the relevant data includes at least one of the following: attribute data, transaction data and target behavior data; inputs the relevant data of the customer into a label generation model to obtain a set of labels for the customer, wherein the label generation model is trained by multiple groups of training samples, each group of training samples includes historical relevant data and historical labels, a set of labels includes multiple labels, and each label is used to characterize a transaction feature of the customer; determines a label matching rule for the target product, matches a set of labels through the label matching rule to obtain a matching result; when the matching result indicates a successful match, determines the customer as a customer to be recommended, defines the set of all customers to be recommended as the recommended customer group for the target product, constructs the customer's label through the customer-related data, matches the customer's label based on the product's label matching rule, and then screens out the successfully matched customers to be recommended, thereby achieving the purpose of enriching the matching rules between the product and the customer to be recommended, thereby realizing the technical effect of improving the accuracy of determining the recommended customer group for the product, and thereby solving the technical problem of inaccurate recommended customer group for the determined product due to the single matching rule between the product and the recommended customer.

[0041] The matching result between the customer and the target product can be determined based on the matching evaluation value of the label. Optionally, in the method for determining the recommended customer group for a product provided in an embodiment of the present application, a group of labels are matched through label matching rules to obtain the matching result, including: for each label in a group of labels, determining the evaluation value of the label from the label matching rules, wherein the label matching rules contain a mapping relationship between each label and the evaluation value; calculating the sum of the evaluation values ​​of all labels in a group of labels to obtain a matching evaluation value; when the matching evaluation value is greater than or equal to the evaluation value threshold, determining the matching result as a successful match; when the matching evaluation value is less than the evaluation value threshold, determining the matching result as a failed match.

[0042] In some examples, in order to more accurately evaluate the degree of match between customers and target products, a label matching method based on evaluation values ​​is adopted. Each label in the label matching rule is associated with an evaluation value, and the matching result is determined by calculating the sum of the evaluation values ​​of a set of labels for the customer.

[0043] For example, for a target product, a label matching rule is formulated based on its characteristics. The rule contains a set of preset labels and their corresponding evaluation values. For example, for a financial product with a high risk preference, the rule may contain a mapping of the "high risk tolerance" label with a high evaluation value, while the "low risk preference" label may be associated with a low evaluation value or a negative evaluation value. For a set of labels of a customer, the evaluation value corresponding to each label in the label matching rule is searched one by one. After completing the evaluation value search for all labels, the sum of these evaluation values ​​is calculated to obtain the matching evaluation value. This sum reflects the overall matching degree between the customer and the target product at the label level.

[0044] Finally, the calculated matching evaluation value is compared with the preset evaluation value threshold to determine the matching result. If the matching evaluation value is greater than or equal to the evaluation value threshold, it means that the customer and the target product are highly matched, and it can be determined as a successful match, that is, the customer is a potential recommendation for the target product. On the contrary, if the matching evaluation value is less than the evaluation value threshold, it means that the customer and the target product are not matched enough to be recommended, so the matching result is determined as a failed match.

[0045] This embodiment objectively and quantitatively evaluates the degree of match between customers and target products based on a set of customer tags and preset tag matching rules, thereby more accurately screening out customer groups suitable for recommending target products and improving product recommendation efficiency and recommendation success rate.

[0046] In order to efficiently screen the recommended customer base, it is necessary to construct label matching rules for the target product. Optionally, in the method for determining the recommended customer base for a product provided in an embodiment of the present application, before determining the label matching rules for the target product, the method also includes: determining the product type and product introduction of the target product, and determining the target conditions that need to be met by the recommended customers through the product type and product introduction; setting an evaluation value for each label mapping according to the target conditions, and constructing the label matching rules for the target product based on all labels and the evaluation values ​​of each label mapping.

[0047] In some examples, the tag matching rule is a matching process based on the expert rule, which can be based on the suitable objects of financial products or financial services, or based on the needs of customers themselves. The standards for promoting financial products and services to suit customer groups can include savings preferences, consumption preferences, financial management attitudes, etc., which are in line with financial products or services. For example, the sales targets of precious metals or funds tend to be customers with certain purchasing power, strong risk awareness, flexible financial control capabilities, and no need for redemption in the short term.

[0048] For example, if the target product is a high-risk financial investment product, it is preferred to look for customers with a high tolerance for risk and certain investment knowledge and experience. Based on the product type and introduction, determine a series of target conditions that the target customers need to meet, such as income level, risk preference, investment history, etc. For each condition, identify the relevant tags in the customer profile and set evaluation values ​​for each tag based on these conditions. The setting of evaluation values ​​can take into account the importance of the condition and the specific content of the customer tag. For example, if the target product requires customers to have a high tolerance for risk, then the tag "high risk awareness" will be assigned a higher evaluation value. Construct a tag matching rule for the target product, which is based on all tags and the evaluation values ​​mapped to each tag. Rules can include, but are not limited to: minimum evaluation value requirements for certain tags included in the customer profile, or the sum of all evaluation values ​​must meet or exceed a set threshold.

[0049] This embodiment can create customized target customer screening criteria for different products through the label evaluation and matching rule construction process based on product characteristics. It ensures a high degree of matching with the product, and also makes the matching process more objective and accurate by quantifying the label evaluation value, thereby greatly improving the pertinence and success rate of product recommendations, while also improving customer experience and avoiding the interruption of irrelevant products.

[0050] In order to determine whether a customer is a customer to be recommended for the target product, it is necessary to train a label generation model to determine the customer's label. Optionally, in the method for determining the recommended customer group for a product provided in an embodiment of the present application, the label generation model is trained in the following manner: obtaining a variety of historical labels set for customers by the target organization, and determining the historical relevant data of each customer; determining each customer's historical label and the customer's historical relevant data as a set of training samples to obtain multiple groups of training samples; training a machine learning model based on the multiple groups of training samples to obtain a label generation model.

[0051] In some examples, multiple historical labels set by the target institution for customers are collected. These labels may include the customer's consumption habits, risk preferences, credit ratings, hobbies, etc. The historical relevant data of each customer, such as transaction records, etc., are determined, and these data are associated with the historical labels to reflect the historical changes in customer behavior and attributes. The historical labels and historical relevant data of each customer are combined into a set of training samples, and multiple sets of such training samples are constructed.

[0052] Based on multiple sets of training samples, select appropriate machine learning algorithms (such as random forests, deep neural networks, etc.) to train the model, with the goal of allowing the model to learn to automatically extract and generate labels from the customer's historical data. During the training process, the data needs to be preprocessed, such as standardization, normalization, feature selection, etc., to improve the training effect and generalization ability of the model. After the model training is completed, it is evaluated and verified, and the prediction accuracy and stability of the model are checked using a test data set that was not involved in the training. Multiple iterations and optimizations can be performed to ensure that the label generation model can maintain a high prediction accuracy on new customer data.

[0053] This embodiment uses a machine learning model to automatically generate customer tags, which can generate a more comprehensive and accurate customer portrait based on the customer's historical behavior and attribute data. It reduces the subjectivity and bias of manually set tags, and can also capture more subtle and dynamic changes in customer portraits, thereby providing deeper insights in the product matching process and improving recommendation efficiency and conversion rate.

[0054] In order to ensure the accuracy of the determined recommended customer group, it is necessary to promptly eliminate customers with the first behavior. Optionally, in the method for determining the recommended customer group for a product provided in an embodiment of the present application, after obtaining the recommended customer group for the target product, the method further includes: for each customer in the recommended customer group, detecting the target behavior data of the customer; and eliminating the customer from the recommended customer group when it is detected that the target behavior data contains the first behavior that affects the purchase intention of the target product.

[0055] In some examples, for each customer in the initially screened recommended customer group, the target behavior data is detected regularly or in real time. The target behavior data is analyzed to identify the first behavior that may affect the purchase intention of the target product. The first behavior may include, but is not limited to: the customer expresses a negative evaluation of similar products or services, participates in discussions that conflict with the product theme, or demonstrates interests or needs that are inconsistent with the product promotion goals.

[0056] A set of evaluation criteria can be set to determine whether the detected first behavior is strong or clear enough to influence the customer's purchase intention for the target product. For example, if a customer expresses distrust of the target product on multiple occasions, this can be considered a strong first behavior signal. Based on the evaluation criteria, customers who are detected to have a first behavior that influences purchase intention are removed from the recommended customer base to avoid ineffective or counterproductive product recommendations.

[0057] In some optional embodiments, correction is performed on whether a customer meets the promotion criteria of a target product. Some information types and specific correction algorithms may be set to perform correction based on target behavior data.

[0058] By correcting whether the customer meets the target product promotion standards and determining the final list of promotion customers (i.e. recommended customer group) for product recommendations, the recommended customer group can be dynamically adjusted and optimized to ensure that the recommended customers not only match the product at the static label level, but are also highly consistent at the target behavior and real-time demand level.

[0059] In order to expand the scope of the recommended customer group, the recommended customer group can be updated in time based on the second behavior. Optionally, in the method for determining the recommended customer group for the product provided in the embodiment of the present application, after obtaining the recommended customer group for the target product, the method also includes: for each customer who does not belong to the recommended customer group, detecting the target behavior data of the customer; when it is detected that there is a second behavior in the target behavior data that affects the purchasing intention of the target product, adding the customer to the recommended customer group.

[0060] In some examples, for each customer in the non-recommended customer group, the target behavior data is detected periodically or in real time. A machine learning model or rule engine is applied to analyze whether there is a second behavior in the target behavior data that affects the purchase intention of the target product. These second behaviors may include positive comments from customers on related products or services, or participation in communities or activities related to the product theme to express interest in the product type.

[0061] Set up a set of evaluation criteria to determine whether the detected second behavior is strong or clear enough to indicate that the customer may have potential purchase intention for the target product. For example, if a customer frequently mentions research or investment plans for a certain type of financial product on multiple platforms or occasions, then this can be considered a clear second behavior signal. For customers who are detected to have a second behavior and are judged to have potential purchase intention for the target product based on the evaluation criteria, add them to the recommended customer group.

[0062] By detecting the target behavior of customers in the non-recommended customer group and identifying the second behavior, this embodiment can discover and attract potential high-value customers that may have been overlooked, dynamically expand the recommended customer group, and thus improve the coverage and conversion rate of product recommendation activities.

[0063] In order to ensure the accuracy of the recommended customer group, it is necessary to update the label matching rules in a timely manner. Optionally, in the method for determining the recommended customer group for a product provided in an embodiment of the present application, after obtaining the recommended customer group for the target product, the method further includes: obtaining the product recommendation result for each customer in the recommended customer group; determining the number of customers whose product recommendation results represent successful recommendations, obtaining the number of customers, and calculating the ratio of the number of customers to the total number of recommended customers to obtain the recommendation success rate; when the recommendation success rate is less than the success rate threshold, updating the label matching rules for the target product based on the label of the recommended customer group.

[0064] In some examples, the product recommendation results for each customer in the recommended customer group can include customer feedback on the recommended product, whether consultation was conducted, whether purchase behavior was generated, and other information. Determine the number of customers who have successfully recommended the product in the product recommendation results. For example, a successful recommendation can be defined as the customer ultimately purchasing the product or at least conducting in-depth consultation. Count this number to obtain the number of customers with successful recommendations. Calculate the ratio of the number of successful recommended customers to the total number of recommended customers, i.e., the recommendation success rate. This ratio can intuitively reflect the effectiveness of the recommendation strategy under the current tag matching rules.

[0065] Set a success rate threshold to determine whether the current recommendation strategy needs to be adjusted. If the calculated recommendation success rate is less than the preset success rate threshold, this indicates that the current tag matching rules may not be effective enough and need to be further optimized. When the recommendation success rate is lower than the success rate threshold, the tag matching rules of the target product are updated based on the tag data of the recommended customer group. For example, adjust the weight of a specific tag, introduce a new tag, or redefine the mapping relationship between tags and evaluation values.

[0066] This embodiment introduces a recommendation effect monitoring and rule dynamic adjustment mechanism, which can continuously optimize the label matching rules of the target product according to the actual recommendation feedback, thereby improving the accuracy and success rate of product recommendations.

[0067] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0068] Example 2

[0069] The embodiment of the present application also provides a device for determining a recommended customer group for a product. It should be noted that the device for determining a recommended customer group for a product in the embodiment of the present application can be used to execute the method for determining a recommended customer group for a product provided in the embodiment of the present application. The device for determining a recommended customer group for a product provided in the embodiment of the present application is introduced below.

[0070] According to an embodiment of the present application, a device for implementing the method for determining a recommended customer group for the above product is also provided. Figure 3 is a schematic diagram of a device for determining a recommended customer group for a product provided in an embodiment of the present application, such as Figure 3 As shown, the device comprises:

[0071] The acquisition unit 301 is used to acquire relevant data of each customer, wherein the relevant data includes at least one of the following: attribute data, transaction data and target behavior data;

[0072] An input unit 302 is used to input relevant data of a customer into a label generation model to obtain a set of labels for the customer, wherein the label generation model is trained by multiple sets of training samples, each set of training samples includes historical relevant data and historical labels, and a set of labels includes multiple labels, each label is used to characterize a transaction feature of the customer;

[0073] The first determination unit 303 is used to determine a label matching rule for a target product, and to match a group of labels with the label matching rule to obtain a matching result;

[0074] The second determining unit 304 is configured to determine the customer as a customer to be recommended when the matching result indicates a successful match, and define the set of all customers to be recommended as a recommended customer group for the target product.

[0075] The device for determining the recommended customer group for a product provided in an embodiment of the present application obtains relevant data of each customer through an acquisition unit 301, wherein the relevant data includes at least one of the following: attribute data, transaction data, and target behavior data; an input unit 302 inputs the relevant data of the customer into a label generation model to obtain a set of labels for the customer, wherein the label generation model is trained by multiple sets of training samples, each set of training samples includes historical relevant data and historical labels, a set of labels includes multiple labels, and each label is used to characterize a transaction feature of the customer; a first determination unit 303 determines a label matching rule for the target product, matches a set of labels through the label matching rule, and obtains a matching result; a second determination unit 304 determines the customer as a customer to be recommended when the matching result indicates a successful match, defines the set of all customers to be recommended as the recommended customer group for the target product, constructs the customer's label through the customer-related data, matches the customer's label based on the product's label matching rule, and then screens out the successfully matched customers to be recommended, thereby achieving the purpose of enriching the matching rules between the product and the customer to be recommended, thereby achieving the technical effect of improving the accuracy of determining the recommended customer group for the product, and thereby solving the technical problem of inaccurate recommended customer group for the product due to the single matching rule between the product and the recommended customer.

[0076] Optionally, in the recommended customer group determination device for the product provided in the embodiment of the present application, the first determination unit 303 includes: a first determination module, used to determine the evaluation value of the label from the label matching rule for each label in a group of labels, wherein the label matching rule includes a mapping relationship between each label and the evaluation value; a first calculation module, used to calculate the sum of the evaluation values ​​of all labels in a group of labels to obtain a matching evaluation value; when the matching evaluation value is greater than or equal to the evaluation value threshold, determining the matching result as a successful match; and a second determination module, used to determine the matching result as a failed match when the matching evaluation value is less than the evaluation value threshold.

[0077] Optionally, in the device for determining the recommended customer group for a product provided in an embodiment of the present application, the device also includes: a third determination unit, used to determine the product type and product introduction of the target product, and determine the target conditions that need to be met by the recommended customers through the product type and product introduction; a setting unit, used to set the evaluation value of each label mapping according to the target conditions, and construct a label matching rule for the target product based on all labels and the evaluation value of each label mapping.

[0078] Optionally, in the recommended customer group determination device for the product provided in the embodiment of the present application, the device also includes: a fourth determination unit, used to obtain a variety of historical labels set by the target organization for customers, and determine the historical related data of each customer; a fifth determination unit, used to determine the historical label of each customer and the historical related data of the customer as a group of training samples, and obtain multiple groups of training samples; a training unit, used to train a machine learning model based on multiple groups of training samples to obtain a label generation model.

[0079] Optionally, in the device for determining a recommended customer group for a product provided in an embodiment of the present application, the device also includes: a first detection unit, used to detect target behavior data of the customer for each customer in the recommended customer group; and an elimination unit, used to eliminate the customer from the recommended customer group when a first behavior that affects the purchase intention of the target product is detected in the target behavior data.

[0080] Optionally, in the device for determining the recommended customer group for a product provided in an embodiment of the present application, the device also includes: a second detection unit, for detecting target behavior data of the customer for each customer who does not belong to the recommended customer group; and an adding unit, for adding the customer to the recommended customer group when a second behavior that affects the purchasing intention of the target product is detected in the target behavior data.

[0081] Optionally, in the device for determining the recommended customer group for a product provided in an embodiment of the present application, the device also includes: a result acquisition unit, used to obtain the product recommendation result for each customer in the recommended customer group; a sixth determination unit, used to determine the number of customers for whom the product recommendation result represents a successful recommendation, obtain the number of customers, and calculate the ratio of the number of customers to the total number of the recommended customer group to obtain the recommendation success rate; and an update unit, used to update the label matching rule of the target product based on the label of the recommended customer group when the recommendation success rate is less than a success rate threshold.

[0082] It should be noted that the acquisition unit 301, input unit 302, first determination unit 303 and second determination unit 304 correspond to steps S201 to S204 in Example 1, and the four units and corresponding steps implement the same examples and application scenarios, but are not limited to the contents disclosed in the above-mentioned Example 1. It should be noted that the above-mentioned modules or units may be hardware components or software components stored in a memory (e.g., memory 104) and processed by one or more processors (e.g., processors 102a, 102b, ..., 102n), and the above-mentioned modules or units may also be part of a device and may be run in the computer terminal 10 provided in Example 1.

[0083] Example 3

[0084] An embodiment of the present application may provide an electronic device, Figure 4 is a structural block diagram of an electronic device according to an embodiment of the present application. Figure 4 As shown, the electronic device may include: one or more ( Figure 4 (only one is shown) processor 402, memory 404, storage controller, and peripheral interface, wherein the peripheral interface is connected to the radio frequency module, audio module and display.

[0085] Among them, the memory can be used to store software programs and modules, such as program instructions / modules corresponding to the methods and devices in the embodiments of the present application, and the processor executes various functional applications and data processing by running the software programs and modules stored in the memory, that is, realizing the above-mentioned method. The memory may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include a memory remotely arranged relative to the processor, and these remote memories may be connected to the terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0086] The processor can call the information and application programs stored in the memory through the transmission device to perform the following steps: obtain relevant data of each customer, wherein the relevant data includes at least one of the following: attribute data, transaction data and target behavior data; input the relevant data of the customer into the label generation model to obtain a set of labels for the customer, wherein the label generation model is trained by multiple groups of training samples, each group of training samples includes historical relevant data and historical labels, a group of labels includes multiple labels, and each label is used to characterize a transaction feature of the customer; determine the label matching rule of the target product, match it with a group of labels through the label matching rule, and obtain a matching result; when the matching result indicates a successful match, determine the customer as a customer to be recommended, and define the set of all customers to be recommended as the recommended customer group for the target product.

[0087] The processor can also call the information and application stored in the memory through the transmission device to perform the following steps: for each tag in a group of tags, determine the evaluation value of the tag from the tag matching rule, wherein the tag matching rule includes a mapping relationship between each tag and the evaluation value; calculate the sum of the evaluation values ​​of all tags in a group of tags to obtain a matching evaluation value; when the matching evaluation value is greater than or equal to the evaluation value threshold, determine the matching result as a successful match; when the matching evaluation value is less than the evaluation value threshold, determine the matching result as a failed match.

[0088] The processor can also call the information and application programs stored in the memory through the transmission device to perform the following steps: determine the product type and product introduction of the target product, and determine the target conditions that need to be met by the recommended customers through the product type and product introduction; set the evaluation value of each label mapping according to the target conditions, and build the label matching rules for the target product based on all labels and the evaluation values ​​of each label mapping.

[0089] The processor can also call the information and application programs stored in the memory through the transmission device to perform the following steps: obtain the various historical labels set by the target institution for the customer, and determine the historical related data of each customer; determine the historical label of each customer and the historical related data of the customer as a set of training samples to obtain multiple sets of training samples; train the machine learning model based on the multiple sets of training samples to obtain a label generation model.

[0090] The processor can also call the information and application programs stored in the memory through the transmission device to perform the following steps: for each customer in the recommended customer group, detect the customer's target behavior data; if a first behavior that affects the purchase intention of the target product is detected in the target behavior data, the customer is removed from the recommended customer group.

[0091] The processor can also call the information and application programs stored in the memory through the transmission device to perform the following steps: for each customer who does not belong to the recommended customer group, detect the customer's target behavior data; when a second behavior that affects the purchase intention of the target product is detected in the target behavior data, add the customer to the recommended customer group.

[0092] The processor can also call the information and application programs stored in the memory through the transmission device to perform the following steps: obtain the product recommendation results for each customer in the recommended customer group; determine the number of customers for whom the product recommendation results represent successful recommendations, obtain the number of customers, and calculate the ratio of the number of customers to the total number of recommended customers to obtain the recommendation success rate; when the recommendation success rate is less than the success rate threshold, update the label matching rules of the target product based on the label of the recommended customer group.

[0093] According to the embodiment of the present application, a method for obtaining relevant data of each customer is provided, wherein the relevant data includes at least one of the following: attribute data, transaction data and target behavior data; the relevant data of the customer is input into a label generation model to obtain a set of labels of the customer, wherein the label generation model is trained by multiple sets of training samples, each set of training samples includes historical relevant data and historical labels, a set of labels includes multiple labels, and each label is used to characterize a transaction feature of the customer; the label matching rule of the target product is determined, and a matching result is obtained by matching a set of labels through the label matching rule; when the matching result indicates a successful match, the customer is determined as a customer to be recommended, and the set of all customers to be recommended is defined as a solution for the recommended customer group of the target product. The customer's label is constructed through the customer's relevant data, and the customer's label is matched based on the product's label matching rule, and then the successfully matched customers to be recommended are screened out, so as to achieve the purpose of enriching the matching rules between the product and the customer to be recommended, thereby achieving the technical effect of improving the accuracy of determining the recommended customer group of the product, and then solving the technical problem that the recommended customer group of the determined product is inaccurate due to the single matching rule between the product and the recommended customer.

[0094] It can be understood by those skilled in the art that Figure 4 The structure shown is for illustration only, and the electronic device may also be a smart phone, a tablet computer, a PDA, a mobile Internet device (MID), a PAD or other terminal device. Figure 4 The structure of the electronic device is not limited. Figure 4 More or fewer components (such as network interfaces, display devices, etc.) shown in, or having Figure 4 Different configurations are shown.

[0095] A person of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing the hardware related to the terminal device through a program, and the program can be stored in a computer-readable storage medium, and the storage medium may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0096] Example 4

[0097] The embodiment of the present application further provides a storage medium. Optionally, in this embodiment, the storage medium can be used to store the program code executed by the method for determining the recommended customer group of a product provided in the first embodiment.

[0098] Optionally, in this embodiment, the above storage medium may be located in any computer terminal in a computer terminal group in a computer network, or in any mobile terminal in a mobile terminal group.

[0099] The present application also provides a computer program product, which, when executed on a data processing device, is suitable for executing the steps of the method for determining a recommended customer group for a product.

[0100] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0101] In the above embodiments of the present application, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0102] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0103] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0104] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0105] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk and other media that can store program codes.

[0106] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for determining a recommended customer group for a product, characterized in that: include: Acquire relevant data of each customer, wherein the relevant data includes at least one of the following: attribute data, transaction data, and target behavior data; Inputting the relevant data of the customer into a label generation model to obtain a set of labels for the customer, wherein the label generation model is trained by multiple sets of training samples, each set of training samples includes historical relevant data and historical labels, and the set of labels includes multiple labels, each label is used to characterize a transaction feature of the customer; Determine a label matching rule for the target product, and match the set of labels with the label matching rule to obtain a matching result; When the matching result indicates a successful match, the customer is determined as a customer to be recommended, and the set of all customers to be recommended is defined as a recommended customer group for the target product.

2. The method according to claim 1, characterized in that The tag matching rule is used to match the group of tags, and the matching result obtained includes: For each tag in the set of tags, determining an evaluation value of the tag from the tag matching rule, wherein the tag matching rule includes a mapping relationship between each tag and the evaluation value; Calculate the sum of the evaluation values ​​of all tags in the set of tags to obtain a matching evaluation value; When the matching evaluation value is greater than or equal to the evaluation value threshold, determining the matching result as a successful match; When the matching evaluation value is less than the evaluation value threshold, the matching result is determined as a matching failure.

3. The method according to claim 2, characterized in that Before determining the label matching rule of the target product, the method further includes: Determine the product type and product introduction of the target product, and determine the target conditions that the recommended customer needs to meet through the product type and product introduction; An evaluation value of each label mapping is set according to the target condition, and a label matching rule of the target product is constructed based on all labels and the evaluation value of each label mapping.

4. The method according to claim 1, characterized in that: The label generation model is trained in the following way: Obtain various historical tags set by the target institution for the customer and determine the historical relevant data of each customer; Determine each customer's historical label and the customer's historical related data as a set of training samples to obtain multiple sets of training samples; A machine learning model is trained based on the multiple groups of training samples to obtain the label generation model.

5. The method according to claim 1, characterized in that: After obtaining the recommended customer group for the target product, the method further includes: For each customer in the recommended customer group, detecting target behavior data of the customer; In the case where a first behavior that affects the purchase intention of the target product is detected in the target behavior data, the customer is removed from the recommended customer group.

6. The method according to claim 1, characterized in that After obtaining the recommended customer group for the target product, the method further includes: For each customer who does not belong to the recommended customer group, detecting target behavior data of the customer; When it is detected that a second behavior that affects the purchase intention of the target product exists in the target behavior data, the customer is added to the recommended customer group.

7. The method according to claim 1, characterized in that After obtaining the recommended customer group for the target product, the method further includes: Obtain product recommendation results for each customer in the recommended customer group; Determine the number of customers for whom the product recommendation result represents a successful recommendation, obtain the number of customers, and calculate the ratio of the number of customers to the total number of recommended customers to obtain the recommendation success rate; When the recommendation success rate is less than a success rate threshold, the tag matching rule of the target product is updated based on the tag of the recommended customer group.

8. A device for determining a recommended customer group for a product, characterized in that: include: An acquisition unit, configured to acquire relevant data of each customer, wherein the relevant data includes at least one of the following: attribute data, transaction data, and target behavior data; An input unit, used to input the relevant data of the customer into a label generation model to obtain a set of labels for the customer, wherein the label generation model is trained by multiple sets of training samples, each set of training samples includes historical relevant data and historical labels, and the set of labels includes multiple labels, each label is used to characterize a transaction feature of the customer; A first determining unit, configured to determine a label matching rule for a target product, and to match the set of labels with the label matching rule to obtain a matching result; The second determining unit is configured to determine the customer as a customer to be recommended if the matching result indicates a successful match, and define a set of all customers to be recommended as a recommended customer group for the target product.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored executable program, wherein when the executable program is run, the device where the computer-readable storage medium is located is controlled to execute the method for determining a recommended customer group for a product according to any one of claims 1 to 7.

10. An electronic device, characterized in that: include: A memory storing an executable program; A processor is used to run the program, wherein the program, when running, executes the method for determining a recommended customer group for a product as described in any one of claims 1 to 7.

11. A computer program product comprising computer instructions, characterized in that: When the computer instructions are executed by a processor, the steps of the method for determining a recommended customer group for a product described in any one of claims 1 to 7 are implemented.

Images