Target user determination method and device, medium and product

Abstract

The invention discloses a target user determination method and device, a medium and a product, and relates to the technical field of artificial intelligence. The method comprises the steps of determining reference financial products associated with a target financial product and reference product features of the reference financial products; determining reference user features of historical purchase users of the reference financial products, and determining a first machine learning model based on the reference product features and the reference user features; clustering each reference user feature to obtain a reference user cluster, and processing each reference user cluster to obtain a target user feature; and determining associated users associated with each target user feature, and determining each associated user as a to-be-recommended target user of the target financial product. According to the scheme of the invention, the problem of user loss caused by wrong customer group matching is solved, the user who has a real intention to purchase the target financial product can be accurately determined, and help is provided for improving the purchase rate of the target financial product.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a method, device, medium and product for determining a target user. Background Art

[0002] Financial products refer to tools and products used for investment, wealth management, risk management, and other financial services. With economic development and the deepening of financial markets, the number of financial products has increased significantly. Common financial products include funds, stocks, futures, and wealth management products.

[0003] At present, users are mainly divided into several customer groups based on characteristics such as risk preference or age, and a mapping relationship between different categories of financial products and each customer group is established based on expert rules, and financial products are recommended to each user based on the mapping relationship. However, in actual scenarios, the purchasing population of products often has multiple characteristics, and there is a certain correlation between the characteristics. Therefore, simply establishing a mapping relationship between different categories of financial products and each customer group based on expert rules cannot truly reflect the user's true intention to purchase financial products, and will lead to user loss due to incorrect customer group matching.

[0004] Based on this, how to accurately identify users who have real intentions to purchase target financial products is crucial to increasing the purchase rate of target financial products. Summary of the Invention

[0005] The present invention provides a method, device, medium, and product for determining target users to solve the problem of user loss due to customer group matching errors. It can accurately determine users who have a real intention to purchase the target financial product, thereby helping to increase the purchase rate of the target financial product.

[0006] According to one aspect of the present invention, a method for determining a target user is provided, the method comprising:

[0007] In response to a recommendation instruction for a target financial product, determining each reference financial product associated with the target financial product, and determining a reference product feature of each reference financial product;

[0008] Determining reference user characteristics of historical purchasing users of each reference financial product, and determining a first machine learning model based on each reference product characteristic and each reference user characteristic;

[0009] Clustering the reference user features to obtain reference user clusters, and processing the reference user clusters based on the first machine learning model to obtain target user features;

[0010] Determine associated users associated with each target user feature, and determine each associated user as a target user to be recommended for the target financial product.

[0011] According to another aspect of the present invention, a device for determining a target user is provided, the device comprising:

[0012] a first determining module configured to determine, in response to a recommendation instruction for a target financial product, reference financial products associated with the target financial product and reference product features of each reference financial product;

[0013] A second determination module is configured to determine reference user characteristics of historical purchasers of each reference financial product, and determine a first machine learning model based on the characteristics of each reference product and the characteristics of each reference user;

[0014] a clustering module, configured to cluster the reference user features to obtain reference user clusters, and process the reference user clusters based on the first machine learning model to obtain target user features;

[0015] The third determining module is configured to determine associated users associated with each target user feature, and determine each associated user as a target user to be recommended for the target financial product.

[0016] According to another aspect of the present invention, an electronic device is provided, comprising:

[0017] at least one processor; and

[0018] a memory communicatively connected to the at least one processor; wherein,

[0019] The memory stores a computer program executable by the at least one processor. The computer program is executed by the at least one processor so that the at least one processor can perform the method for determining a target user according to any embodiment of the present invention.

[0020] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for determining a target user according to any embodiment of the present invention when executed.

[0021] According to another aspect of the present invention, a computer program product is provided, comprising a computer program, wherein when the computer program is executed by a processor, the method for determining a target user according to any embodiment of the present invention is implemented.

[0022] The technical solution of the embodiment of the present invention determines the reference financial products associated with the target financial product and the reference product features of each reference financial product in response to a recommendation instruction of the target financial product; determines the product features of the reference financial products associated with the target financial product, and based on the product features of the reference financial products and the user features of the user who purchased the reference financial product; further, a first machine learning model can be determined based on the reference product features and the reference user features, and the importance of each reference user feature can be ranked; each reference user feature is clustered to obtain each reference user cluster cluster, and each reference user cluster cluster is processed based on the first machine learning model to obtain the target user feature; determines the associated users associated with each target user feature, and determines each associated user as the target user to be recommended for the target financial product, which solves the problem of user loss due to customer group matching errors, can accurately determine users who have a real intention to purchase the target financial product, and provide help to improve the purchase rate of the target financial product.

[0023] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present invention, nor is it intended to limit the scope of the present invention. Other features of the present invention will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0025] Figure 1 This is a flowchart of a method for determining a target user provided in accordance with the first embodiment of the present invention;

[0026] Figure 2 This is a flow chart of a method for determining a target user according to a second embodiment of the present invention;

[0027] Figure 3 This is a flowchart of a method for determining a target user provided in accordance with the third embodiment of the present invention;

[0028] Figure 4 is a flowchart of another method for determining a target user provided in accordance with the third embodiment of the present invention;

[0029] Figure 5 This is a schematic diagram of the structure of a device for determining a target user according to a fourth embodiment of the present invention;

[0030] Figure 6It is a structural diagram of an electronic device for implementing the method for determining a target user according to an embodiment of the present invention. DETAILED DESCRIPTION

[0031] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0032] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention 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 numbers used in this way can be interchanged where appropriate so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes 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.

[0033] Example 1

[0034] Figure 1 This is a flow chart of a method for determining a target user according to a first embodiment of the present invention. This embodiment is applicable to the case of determining a target user to be recommended a target financial product. The method can be executed by a target user determination device, which can be implemented in the form of hardware and / or software. The target user determination device can be configured in an electronic device such as a computer, server, or tablet computer. Figure 1 As shown, the method includes:

[0035] Step 110 : In response to the recommendation instruction for the target financial product, determine each reference financial product associated with the target financial product, and determine a reference product feature of each reference financial product.

[0036] Among them, the target financial products can be funds, stocks, bonds, futures or wealth management products, etc.

[0037] Optionally, in this embodiment, before the target financial product is launched, or after the target financial product is launched, if the target financial product needs to be promoted, the financial system can automatically generate a recommendation instruction for the target financial product to determine users with a higher probability of purchasing the target financial product and recommend the target financial product to these users.

[0038] In an optional implementation of this embodiment, after receiving the recommendation instruction for the target financial product, each reference financial product associated with the target financial product may be further determined, and reference product features of each reference financial product may be determined.

[0039] It is understandable that each financial product may contain multiple product information, such as product name, product features, product terms, etc. When one or more product information of a financial product is the same or similar to a target financial product, then it can be determined that the financial product is associated with the target financial product, that is, it is a reference financial product associated with the target financial product.

[0040] For example, if the target financial product is a low-risk current financial product A, then the reference financial products associated with it may be a low-risk current financial product B, a low-risk fixed-term financial product C, and so on.

[0041] Optionally, in this embodiment, after the reference financial product is determined, all product information of the reference financial product may be obtained, and the product information may be represented as a vector to obtain the reference product features.

[0042] It should be noted that, in this embodiment, the product information of each financial product, as well as the user information of each financial product purchased later, is obtained only after authorization by the user, and the method of obtaining it is reasonable and legal.

[0043] Step 120: Determine reference user characteristics of historical purchasing users of each reference financial product, and determine a first machine learning model based on each reference product characteristic and each reference user characteristic.

[0044] Among them, the first machine learning model can be used to rank the feature importance of each reference user feature.

[0045] Optionally, in this embodiment, after determining the reference financial products associated with the target financial product, the historical purchasing users of each reference financial product can be further determined, for example, users who purchased each reference financial product in the past month, three months, or one year; further, the user information of each user (for example, name, address, contact information, occupation, or credit report, etc.) can be determined, and this user information can be represented by a vector to obtain the characteristics of each reference user.

[0046] In this embodiment, after determining the reference product features and the reference user features, the target machine learning model can be further pre-trained based on the reference product features and the reference user features to obtain a first machine learning model; wherein, the target machine learning model can be a centralized learning model, such as a random forest model, an extreme random tree model, a gradient boosting tree model or a deep forest model, etc., which is not limited in this embodiment.

[0047] Step 130: Cluster the reference user features to obtain reference user clusters, and process the reference user clusters based on the first machine learning model to obtain target user features.

[0048] Optionally, in this embodiment, after determining to obtain the first machine learning model and sorting the feature importance of each reference user feature based on the first machine learning model, the sorted reference user features can be further clustered to obtain reference user clusters.

[0049] The solution of this embodiment, after ranking the feature importance of each reference user feature based on the first machine learning model, clusters the reference user features, which can improve the accuracy of clustering the reference user features and greatly cluster related and important features into the same cluster cluster.

[0050] Furthermore, each reference user cluster obtained can be processed based on the first machine learning model, that is, the feature importance of each reference user cluster is ranked by the first machine learning model, so as to determine the target user feature; illustratively, the target user feature can be the user feature in the most important reference user cluster in the ranking result.

[0051] Step 140: Determine associated users associated with each target user feature, and determine each associated user as a target user to be recommended for the target financial product.

[0052] Optionally, in this embodiment, after determining the target user characteristics, associated users associated with the target user characteristics can be further determined among all users in the financial system, for example, users whose similarity with the target user characteristics is greater than a set similarity threshold. Further, these associated users can be determined as target users to be recommended for the target financial product.

[0053] The technical solution of this embodiment, by responding to a recommendation instruction of a target financial product, determines each reference financial product associated with the target financial product and determines the reference product features of each reference financial product; by determining the product features of the reference financial products associated with the target financial product, and based on the product features of the reference financial products and the features of users who purchase the reference financial products; further, a first machine learning model can be determined based on the reference product features and the reference user features, and the importance of each reference user feature can be ranked; each reference user feature is clustered to obtain each reference user cluster cluster, and each reference user cluster cluster is processed based on the first machine learning model to obtain target user features; associated users associated with each target user feature are determined, and each associated user is determined as a target user to be recommended for the target financial product, which solves the problem of user loss due to customer group matching errors, can accurately determine users who have a real intention to purchase the target financial product, and provide help to improve the purchase rate of the target financial product.

[0054] Example 2

[0055] Figure 2 This is a flow chart of a method for determining a target user according to the second embodiment of the present invention. This embodiment is a further refinement of the above technical solution. The technical solution in this embodiment can be combined with various optional solutions in one or more of the above embodiments. Figure 2 As shown, the method includes:

[0056] Step 210: Acquire attribute information of the target financial product; perform feature representation on each attribute information to obtain a first product feature.

[0057] Optionally, in this embodiment, after receiving the recommendation instruction of the target financial product, attribute information of the target financial product can be further obtained. For example, the product manual of the target financial product can be identified to obtain the attribute information of the target financial product; wherein, the attribute information of the target financial product may include at least one of the following: product name, product type, affiliated institution, risk assessment, liquidity, term and terms, user attention and exposure.

[0058] Furthermore, each attribute information may be represented by a feature to obtain the first product feature; in an optional implementation of this embodiment, each attribute information may be represented in the form of a feature vector, and further, each feature vector may be concatenated to obtain the first product feature.

[0059] The solution of this embodiment can quickly obtain the product characteristics of the target financial product by obtaining the attribute information of the target amount product and obtaining the first product characteristics based on each attribute information, and can greatly characterize the target financial product through the product characteristics.

[0060] Step 220: Determine reference financial products associated with the target financial product based on the first product characteristics; obtain attribute information of each reference financial product, and perform feature representation on each attribute information to obtain characteristics of each reference product.

[0061] Optionally, in this embodiment, after determining the first product feature of the target financial product, reference financial products associated with the target financial product can be further determined based on the first product feature; illustratively, the first product feature can be compared with product features or attribute information of other financial products. When the similarity between the two is greater than a set threshold (for example, 0.8, 0.9 or 0.99, etc.), the financial product can be determined as a reference financial product.

[0062] Furthermore, the attribute information of each reference financial product can be obtained separately, and each attribute information can be represented to obtain the reference product characteristics of each reference financial product; illustratively, each attribute information can be represented in the form of a feature vector, and further, each feature vector can be spliced ​​to obtain the reference product characteristics.

[0063] The advantage of this setting is that it can quickly determine the reference financial products associated with the target financial product and determine the product characteristics of each reference financial product, which helps in the subsequent determination of the first machine learning model.

[0064] Step 230: Determine reference user characteristics of historical purchasing users of each reference financial product, and determine a first machine learning model based on each reference product characteristic and each reference user characteristic.

[0065] Step 240: Cluster the reference user features to obtain reference user clusters, and process the reference user clusters based on the first machine learning model to obtain target user features.

[0066] Step 250: Determine associated users associated with each target user feature, and determine each associated user as a target user to be recommended for the target financial product.

[0067] Optionally, in this embodiment, determining the associated users associated with each of the target user characteristics and determining each of the associated users as the target users to be recommended for the target financial product may include: determining the user characteristics of all users, and determining the consistency between each of the user characteristics and the target user characteristics; when it is determined that the target consistency meets the preset consistency requirements, determining the target user corresponding to the target consistency as the target user to be recommended for the target financial product.

[0068] Among them, all users can refer to all registered users of the financial system where the target financial product is launched.

[0069] Optionally, in this embodiment, after determining the target user characteristics, the user characteristics of all users in the financial system can be further determined; further, the similarity between the user characteristics of all users and the characteristics of each reference user can be calculated separately; when the similarity between the user characteristics of the first user and the characteristics of the first reference user is greater than a set similarity threshold (for example, 0.98), the first user can be determined as the user to be recommended for the target financial product.

[0070] It can be understood that in this embodiment, each reference user can be directly determined as a user to be recommended for the target financial product; further, each user related to each reference user in the entire user base can also be determined as a user to be recommended for the target financial product. This can increase the recommended customer base for the target financial product and help improve the purchase rate of the target financial product.

[0071] The solution of this embodiment, after determining the target user characteristics, can further determine the user characteristics of all users and determine the consistency of each user characteristic with the target user characteristics; when it is determined that the target consistency meets the preset consistency requirements, the target user corresponding to the target consistency is determined as the target user to be recommended for the target financial product. This can accurately determine the target user to be recommended for the target financial product, and improve the recommended customer base of the target financial product based on the reference user, thereby helping to increase the purchase rate of the target financial product.

[0072] Example 3

[0073] Figure 3 This is a flow chart of a method for determining a target user according to the third embodiment of the present invention. This embodiment is a further refinement of the above technical solution. The technical solution in this embodiment can be combined with various optional solutions in one or more of the above embodiments. Figure 3 As shown, the method includes:

[0074] Step 310: In response to the recommendation instruction for the target financial product, determine each reference financial product associated with the target financial product, and determine a reference product feature of each reference financial product.

[0075] Step 320: Determine each reference user who purchased each reference financial product within a preset time period; extract user features of each reference user respectively to obtain each reference user feature; input each reference product feature and each reference user feature into a preset machine learning model for pre-training, and obtain the first machine learning model when the iteration stop condition is met.

[0076] The preset time period may be the past month, the past three months, or the past six months, etc., which is not limited in this embodiment.

[0077] Optionally, in this embodiment, after determining each reference financial product, each reference user who purchased each reference financial product within a preset time period can be further determined; for example, if user A purchased reference financial product C yesterday, then user A can be determined as a reference user.

[0078] Furthermore, user features of each reference user are extracted. For example, user information of each reference user (for example, registration information, contact information, account opening time or name, etc.) can be obtained. Furthermore, feature representation can be performed on the obtained user information, and the information after feature representation can be spliced ​​to obtain the features of each reference user.

[0079] Furthermore, the determined reference product features and reference user features can be input into the target machine learning model for pre-training, and the first machine learning model can be obtained when the iteration stop condition is met.

[0080] Among them, the target machine learning model can be a random forest model, an extreme random tree model, a gradient boosting tree model or a deep forest model, etc., which is not limited in this embodiment.

[0081] In this embodiment, the determined reference product features and reference user features can be input into the random forest model for iterative calculation. When the number of iterations reaches the set number of iterations (for example, 10,000 times, 50,000 times, or 100,000 times, etc.), the first machine learning model is obtained.

[0082] The advantage of this setting is that before clustering the features of each reference user, pre-training the first machine learning model can ensure that the feature ranking obtained by the first machine learning model is consistent with the historical experience of fund product sales, and can avoid feature analysis bias affecting the analysis results.

[0083] Step 330: Determine a preset cluster number range for the clustering cluster, and cluster each of the reference user features based on each preset cluster number within the preset cluster number range to obtain each clustering result; determine a clustering quality index for each of the clustering results, and draw an index curve based on each of the clustering quality indexes; and determine a target cluster number based on the index curve.

[0084] The preset number of clusters may range from 1 to 10, 1 to 15, or 1 to 20, etc., which is not limited in this embodiment.

[0085] In this embodiment, the preset cluster number range of the clustering cluster can be randomly determined, and each reference user feature can be further clustered based on each preset cluster number within the preset cluster number range, thereby obtaining each clustering result; illustratively, if the preset cluster number range is 1-10, then each reference user feature can be clustered according to 1 preset cluster number, 2 preset cluster numbers, ..., 10 preset cluster numbers, thereby obtaining a clustering result.

[0086] It can be understood that the obtained clustering results include 1 cluster, 2 clusters, ..., 10 clusters respectively.

[0087] Furthermore, the clustering quality index of each clustering result can be determined respectively, wherein the clustering quality index can be the intra-cluster sum of squares of each clustering result. Furthermore, an index curve can be drawn based on the clustering quality index respectively; wherein the horizontal coordinate of the index curve can be a preset number of clusters (1-10), and the vertical coordinate can be the clustering quality index; further, the point in the index curve where the rate of decline of the index slows down significantly can be used as the optimal cluster number point, that is, the target cluster number.

[0088] It is understandable that directly clustering the characteristics of each reference user can only clearly show that the data of the same cluster is similar, and the data of different clusters are quite different. In order to completely get rid of the influence of expert judgment on the results, the solution of this embodiment calculates the intra-cluster sum of squares under different clusters based on the clustering algorithm, and generates a curve of the relationship between the number of clusters and the sum of squares. The inflection point of the curve is the optimal cluster number point, which can accurately obtain the target number of clusters and provide a basis for the subsequent accurate determination of the users to be recommended.

[0089] Step 340: Cluster the reference user features based on the target number of clusters to obtain reference user clusters; input each reference user cluster into the first machine learning model to obtain a feature importance ranking of each reference user cluster; and determine the target user features based on the feature importance ranking.

[0090] Optionally, in this embodiment, after determining the target number of clusters (for example, 8), each reference user feature can be further clustered based on the target number of clusters to obtain each reference user cluster; it can be understood that if the target number of clusters is 8, then the number of reference user clusters obtained by clustering is 8.

[0091] Furthermore, each reference user cluster can be input into the first machine learning model obtained by the above pre-training, and the feature importance of each reference user cluster can be ranked by the first machine learning model. Furthermore, the target user features can be determined based on the importance ranking results. For example, if the ranking results from high to low importance are: the sixth reference cluster, the seventh reference cluster, the eighth reference cluster, the fifth reference cluster, the fourth reference cluster, the third reference cluster, the second reference cluster, and the first reference cluster, then the reference user features in the sixth cluster can be determined as the target user features; the reference user features in the sixth reference cluster, the seventh reference cluster, and the eighth reference cluster can also be determined as the target user features.

[0092] In the specific implementation, an appropriate number of features can be taken as target user features based on business needs (if too many features are taken, it may result in a smaller number of customers in the final marketing customer base, so it is necessary to extract the most important features based on actual business needs). Furthermore, they can be mapped with product features to generate a mapping table.

[0093] Step 350: Determine associated users associated with each target user feature, and determine each associated user as a target user to be recommended for the target financial product.

[0094] In an optional implementation of this embodiment, after determining the users to be recommended to the target financial product, a marketing plan for a template financial product can be further generated based on the users to be recommended to the target financial product, and the target amount product can be promoted based on the marketing plan.

[0095] The solution of the embodiment of the present invention, after obtaining the target number of clusters, can cluster the reference user features based on the target number of clusters to obtain each reference user cluster; each reference user cluster is input into the first machine learning model respectively to obtain the feature importance ranking of each reference user cluster; the target user features are determined according to the feature importance ranking, and the number of target user features can be accurately determined based on business needs, providing assistance for the subsequent precision marketing of target financial products.

[0096] In order to better understand the method for determining the target user involved in this embodiment, Figure 4is a flowchart of another method for determining a target user according to the third embodiment of the present invention, referring to Figure 4 , which mainly include:

[0097] Step 410: Prepare feature data.

[0098] In this embodiment, it may include reference product features and reference user features.

[0099] Step 420: train a random forest model using the feature data;

[0100] Step 430: Determine the number of clusters.

[0101] Step 440: Cluster and segment the customer groups;

[0102] Step 450: sort the features based on the random forest model;

[0103] Step 460: Extract features to form a mapping.

[0104] It can be understood that in this embodiment, after determining the feature mapping of financial product A and the purchasing population, in the subsequent marketing process, if financial product A or a financial product with the same features is marketed again, the user feature values ​​in the mapping table can be directly taken, and data can be extracted from all users based on the features. Users with the same features become the best marketing customer group for the product.

[0105] In a specific example of this embodiment, a financial institution wants to promote financial product A. Due to limited marketing resources and time, it hopes to find the best audience group for financial product A. The specific operations can be as follows:

[0106] The characteristics of financial product A are extracted in the system, and financial product B with the same characteristics is determined, such as index-type funds with low visit volume and low volatility; further, the characteristics of the historical purchasing population of financial product B are determined; the random forest model is pre-trained based on the characteristics of financial product B and the characteristics of the purchasing population; further, the optimal number of clusters can be determined, such as 5 categories; the purchasing customers of financial product B are divided into 5 categories through the clustering algorithm; the characteristics of the 5 categories of customers are calculated using the random forest model respectively, and the characteristic rankings of the 5 categories of customers are obtained respectively; further, the number of features to be extracted can be determined in combination with marketing resources, and a mapping relationship can be formed with the characteristics of financial product B, such as: Product: index fund, low visit volume, low volatility; User: aggressive risk preference, assets greater than 100,000, preference for purchasing funds; finally, a marketing strategy for financial product A can be formed, and recommendations can be made to fund holding customers with assets greater than 100,000 and aggressive risk preferences.

[0107] The solution of the embodiment of the present invention can use an algorithm to analyze historical data to provide a future marketing customer group mapping, thereby freeing up operational manpower to the greatest extent and reducing the difficulty of data analysis; it can effectively solve the analysis bias caused by the lack of correlation between features caused by extracting feature values ​​through expert judgment.

[0108] Example 4

[0109] Figure 5 FIG. 1 is a schematic diagram of a target user determination device according to a fourth embodiment of the present invention. Figure 5 As shown, the apparatus includes: a first determination module 510 , a second determination module 520 , a clustering module 530 and a third determination module 540 .

[0110] The first determining module 510 is configured to determine, in response to a recommendation instruction for a target financial product, reference financial products associated with the target financial product and reference product characteristics of each reference financial product;

[0111] A second determination module 520 is configured to determine reference user characteristics of historical purchasers of each reference financial product, and determine a first machine learning model based on the reference product characteristics and the reference user characteristics;

[0112] A clustering module 530 is configured to cluster the reference user features to obtain reference user clusters, and process the reference user clusters based on the first machine learning model to obtain target user features;

[0113] The third determining module 540 is configured to determine associated users associated with each target user feature, and determine each associated user as a target user to be recommended for the target financial product.

[0114] The solution of this embodiment is to determine, by a first determination module, each reference financial product associated with the target financial product in response to a recommendation instruction for the target financial product, and determine the reference product characteristics of each reference financial product; determine, by a second determination module, the reference user characteristics of historical purchasing users of each reference financial product, and determine a first machine learning model based on the reference product characteristics and the reference user characteristics; cluster the reference user characteristics by a clustering module to obtain reference user clusters, and process the reference user clusters based on the first machine learning model to obtain target user characteristics; and determine, by a third determination module, associated users associated with each target user characteristic, and determine each associated user as a target user to be recommended for the target financial product. This solves the problem of user loss due to customer group matching errors, can accurately determine users who have a real intention to purchase the target financial product, and help improve the purchase rate of the target financial product.

[0115] In an optional implementation of this embodiment, the target user determination device further includes: a first product feature determination module, configured to:

[0116] Obtaining attribute information of the target financial product;

[0117] Performing feature representation on each of the attribute information to obtain a first product feature;

[0118] The attribute information includes at least one of the following: product name, product type, affiliated institution, economic characteristics, risk assessment, liquidity, term and terms, user attention and exposure.

[0119] In an optional implementation of this embodiment, the first determining module 510 is specifically configured to:

[0120] determining reference financial products associated with the target financial product based on the first product feature;

[0121] Attribute information of each reference financial product is obtained respectively, and characteristic representation is performed on each attribute information to obtain characteristics of each reference product.

[0122] In an optional implementation of this embodiment, the second determining module 520 is specifically configured to:

[0123] Determining reference users who purchased the reference financial products within a preset time period;

[0124] Extracting user features of each reference user respectively to obtain each reference user feature;

[0125] Input each of the reference product features and each of the reference user features into the target machine learning model for pre-training, and obtain the first machine learning model when the iteration stop condition is met.

[0126] In an optional implementation of this embodiment, the target user determination device further includes: a target cluster number determination module, configured to:

[0127] Determining a preset cluster number range for clustering clusters, and clustering each of the reference user features based on each preset cluster number within the preset cluster number range to obtain each clustering result;

[0128] Determining a clustering quality index for each clustering result, and drawing an index curve based on each clustering quality index;

[0129] The target number of clusters is determined based on the indicator curve.

[0130] In an optional implementation of this embodiment, the clustering module 530 is specifically configured to

[0131] Clustering the reference user features based on the target number of clusters to obtain reference user clusters;

[0132] Inputting each of the reference user clusters into the first machine learning model to obtain a feature importance ranking of each of the reference user clusters;

[0133] The target user features are determined according to the feature importance ranking.

[0134] In an optional implementation of this embodiment, the third determining module 540 is specifically configured to:

[0135] Determining user characteristics of all users and determining consistency between the user characteristics and the target user characteristics;

[0136] When it is determined that the target consistency meets the preset consistency requirement, the target user corresponding to the target consistency is determined as the target user to be recommended for the target financial product.

[0137] The target user determination device provided in the embodiment of the present invention can execute the target user determination method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0138] In the technical solutions of the embodiments of the present invention, the attribute information of the financial products involved, as well as the collection, storage, use, processing, transmission, provision and disclosure of user information (such as identity information, credit information, etc.) all comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0139] Example 5

[0140] Figure 6 A schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0141] like Figure 6As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., which is communicatively connected to the at least one processor 11. The memory stores a computer program that can be executed by the at least one processor. The processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. Various programs and data required for the operation of the electronic device 10 can also be stored in the RAM 13. The processor 11, ROM 12, and RAM 13 are connected to each other via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0142] Multiple components in the electronic device 10 are connected to the I / O interface 15, including an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a magnetic disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0143] The processor 11 may be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors that run machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as a target user determination method, which may include: in response to a recommendation instruction for a target financial product, determining each reference financial product associated with the target financial product and determining reference product features for each reference financial product; determining reference user features of historical purchasers of each reference financial product, and determining a first machine learning model based on each reference product feature and each reference user feature; clustering each reference user feature to obtain each reference user cluster cluster, and processing each reference user cluster cluster based on the first machine learning model to obtain target user features; determining associated users associated with each target user feature, and determining each associated user as a target user to be recommended for the target financial product.

[0144] In some embodiments, the target user determination method may be implemented as a computer program that is tangibly contained in a computer-readable storage medium, such as the storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the target user determination method described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to perform the target user determination method in any other appropriate manner (e.g., by means of firmware).

[0145] Various embodiments of the systems and techniques described 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), system-on-chip systems (SOCs), programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0146] Computer programs for implementing 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 the computer program is executed by the processor, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The computer program may be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0147] In the context of the present invention, computer-readable storage media can be tangible media that can contain or store a computer program for use with an instruction execution system, device or equipment or used in combination with an instruction execution system, device or equipment. Computer-readable storage media can include but are not limited to electronic, magnetic, optical, electromagnetic, infrared or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, computer-readable storage media can be machine-readable signal media. More specific examples of machine-readable storage media can include electrical connections based on one or more lines, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0148] 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 can provide input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0149] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0150] A computing system may include clients and servers. The clients and servers are typically remote from each other and typically interact via a communication network. This client-server relationship arises through computer programs running on the respective computers, creating a client-server relationship. The server may be a cloud server, also known as a cloud computing server or cloud host. This server is a hosting product within the cloud computing service ecosystem that addresses the management difficulties and limited scalability of traditional physical hosting and VPS services.

[0151] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in the present invention can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved. This is not limited herein.

[0152] The above specific embodiments do not limit the scope of protection of the present invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention are intended to be included within the scope of protection of the present invention.

[0153] An embodiment of the present invention further provides a computer program product, including a computer program, which, when executed by a processor, implements the database detection method provided in any embodiment of the present application.

[0154] The computer program product may be implemented by writing computer program code for performing the operations of the present invention in one or more programming languages, or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0155] It should be noted that in the embodiments of the present application, certain software, components, models and other existing solutions in the industry may be mentioned. They should be regarded as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of the present application, but it does not mean that the applicant has or will necessarily use the solution.

[0156] Note that the above are only preferred embodiments of the present invention and the technical principles employed. Those skilled in the art will appreciate that the present invention is not limited to the specific embodiments herein, and that various obvious changes, readjustments, and substitutions are possible for those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments and may include many other equivalent embodiments without departing from the scope of the present invention. The scope of the present invention is determined by the scope of the appended claims.

Claims

1. A method for determining target users, characterized in that: include: In response to a recommendation instruction for a target financial product, determining each reference financial product associated with the target financial product, and determining a reference product feature of each reference financial product; Determining reference user characteristics of historical purchasing users of each reference financial product, and determining a first machine learning model based on each reference product characteristic and each reference user characteristic; Clustering the reference user features to obtain reference user clusters, and processing the reference user clusters based on the first machine learning model to obtain target user features; Determine associated users associated with each target user feature, and determine each associated user as a target user to be recommended for the target financial product.

2. The method for determining a target user according to claim 1, wherein: Before determining each reference financial product associated with the target financial product, the method further includes: Obtaining attribute information of the target financial product; Performing feature representation on each of the attribute information to obtain a first product feature; The attribute information includes at least one of the following: product name, product type, affiliated institution, risk assessment, liquidity, term and terms, user attention and exposure.

3. The method for determining a target user according to claim 2, wherein: The determining of each reference financial product associated with the target financial product and determining a reference product feature of each reference financial product includes: determining reference financial products associated with the target financial product based on the first product feature; Attribute information of each reference financial product is obtained respectively, and characteristic representation is performed on each attribute information to obtain characteristics of each reference product.

4. The method for determining a target user according to claim 1, wherein: Determining reference user characteristics of historical purchasers of each reference financial product, and determining a first machine learning model based on each reference product characteristic and each reference user characteristic, includes: Determining reference users who purchased the reference financial products within a preset time period; Extracting user features of each reference user respectively to obtain each reference user feature; Input each of the reference product features and each of the reference user features into the target machine learning model for pre-training, and obtain the first machine learning model when the iteration stop condition is met.

5. The method for determining a target user according to claim 1, wherein: Before clustering the reference user features to obtain reference user clusters, the method further includes: Determining a preset cluster number range for clustering clusters, and clustering each of the reference user features based on each preset cluster number within the preset cluster number range to obtain each clustering result; Determining a clustering quality index for each clustering result, and drawing an index curve based on each clustering quality index; The target number of clusters is determined based on the indicator curve.

6. The method for determining a target user according to claim 5, wherein: The clustering of the reference user features to obtain reference user clusters, and processing the reference user clusters based on the first machine learning model to obtain target user features includes: Clustering the reference user features based on the target number of clusters to obtain reference user clusters; Inputting each of the reference user clusters into the first machine learning model to obtain a feature importance ranking of each of the reference user clusters; The target user features are determined according to the feature importance ranking.

7. The method for determining a target user according to claim 1, wherein: The determining of associated users associated with each target user characteristic, and determining each associated user as a target user to be recommended for the target financial product, includes: Determining user characteristics of all users and determining consistency between the user characteristics and the target user characteristics; When it is determined that the target consistency meets the preset consistency requirement, the target user corresponding to the target consistency is determined as the target user to be recommended for the target financial product.

8. A device for determining a target user, characterized in that: include: a first determining module configured to determine, in response to a recommendation instruction for a target financial product, reference financial products associated with the target financial product and reference product features of each reference financial product; A second determination module is configured to determine reference user characteristics of historical purchasers of each reference financial product, and determine a first machine learning model based on the characteristics of each reference product and the characteristics of each reference user; a clustering module, configured to cluster the reference user features to obtain reference user clusters, and process the reference user clusters based on the first machine learning model to obtain target user features; The third determining module is configured to determine associated users associated with each target user feature, and determine each associated user as a target user to be recommended for the target financial product.

9. An electronic device, characterized in that: The electronic device comprises: 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 is executed by the at least one processor to enable the at least one processor to perform the target user determination method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for determining a target user according to any one of claims 1 to 7 when executed.

11. A computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements the method for determining a target user according to any one of claims 1 to 7.

Images