Product association rule set generation method, product recommendation method and device

Abstract

The invention provides a product association rule set generation method and an association rule set-based product recommendation method, which can be applied to the technical field of big data and artificial intelligence. The association rule set generation method of the product comprises the following steps: preprocessing historical transaction data of the product to obtain preprocessed transaction data; and performing association rule mining on the preprocessed transaction data and the user information corresponding to the transaction data by using a predefined association rule mining method to obtain an association rule set. The association rule set-based product recommendation method comprises the following steps: classifying target customers to obtain a customer classification result, and generating an initial product recommendation result by utilizing an association rule set based on the customer classification result; and analyzing the initial product recommendation result to obtain an analysis result, and recommending a product to the target customer based on the analysis result and a preset interference condition. The invention further provides an association rule set generation device, a product recommendation device, equipment, a medium and a program product.

Description

Technical Field

[0001] The present disclosure relates to the field of big data and artificial intelligence technology, and in particular to a method for generating an association rule set for a product, a product recommendation method based on the association rule set, a device, a medium, and a program product. Background Art

[0002] Data mining is the process of extracting useful information from a large amount of data, that is, discovering regular, implicit, and potentially useful and ultimately understandable information and knowledge between various events from massive, fuzzy, incomplete, noisy, and random actual application data. Specifically in the field of product management, data mining is a technology that deeply analyzes information related to products: it analyzes and explores a large amount of product transaction data according to the business goals set by product operators to reveal unknown, hidden or verified known rules and further model them, thereby extracting relevant business models that are useful for assisting operators in making decisions, and can predict the purchasing behavior of target customers of various credit levels, and formulate relevant business strategies in advance to guide them.

[0003] However, due to the rapid development of Internet technology, a wide variety of data has been generated in actual business activities. How to mine association rules for the above data and recommend products based on the generated association rules has become a focus of increasing attention for technical personnel. Existing product recommendation solutions have technical problems such as low promotion efficiency and low product recommendation accuracy, which in turn leads to problems such as poor purchasing experience for target customers. Summary of the invention

[0004] In view of the above problems, the present disclosure provides a method for generating an association rule set of products to improve the accuracy of product recommendation, a product recommendation method based on the association rule set, an apparatus, a device, a medium and a program product.

[0005] According to a first aspect of the present disclosure, a method for generating an association rule set for a product is provided, comprising:

[0006] Preprocess the historical transaction data of the product to obtain preprocessed transaction data;

[0007] The pre-processed transaction data and the user information corresponding to the transaction data are subjected to association rule mining by using a pre-defined association rule mining method to obtain an association rule set, wherein the pre-defined association rule mining method utilizes a genetic algorithm.

[0008] According to a second aspect of the present disclosure, a product recommendation method based on an association rule set is provided, comprising:

[0009] Classifying target customers to obtain customer classification results, and based on the customer classification results, generating initial product recommendation results using an association rule set, wherein the association rule set is generated by the association rule set generation method of the above product;

[0010] The initial product recommendation results are analyzed to obtain analysis results, and based on the analysis results and preset interference conditions, products are recommended to target customers.

[0011] According to a third aspect of the present disclosure, there is provided a device for generating an association rule set for a product, comprising:

[0012] A data preprocessing module is used to preprocess the historical transaction data of the product to obtain preprocessed transaction data;

[0013] The association rule mining module is used to perform association rule mining on the preprocessed transaction data and the user information corresponding to the transaction data using a predefined association rule mining method to obtain an association rule set, wherein the predefined association rule mining method uses a genetic algorithm.

[0014] According to a fourth aspect of the present disclosure, a product recommendation device based on an association rule set is provided, comprising:

[0015] A classification and generation module, used to classify target customers, obtain customer classification results, and generate initial product recommendation results based on the customer classification results using an association rule set, wherein the association rule set is generated by the association rule set generation method of the above-mentioned product;

[0016] The analysis and recommendation module is used to analyze the initial product recommendation results, obtain the analysis results, and recommend products to target customers based on the analysis results and preset interference conditions.

[0017] According to a fifth aspect of the present disclosure, there is provided an electronic device, including:

[0018] one or more processors;

[0019] a memory for storing one or more computer programs,

[0020] It is characterized in that one or more processors execute one or more computer programs to implement the steps of the product association rule set generation method and the product recommendation method based on the association rule set.

[0021] According to a sixth aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program or instruction is stored, characterized in that when the computer program or instruction is executed by a processor, the steps of the method for generating an association rule set for the above-mentioned product and the method for recommending products based on the association rule set are implemented.

[0022] According to a seventh aspect of the present disclosure, a computer program product is provided, including a computer program or instructions, characterized in that when the computer program or instructions are executed by a processor, the steps of the method for generating an association rule set for the above-mentioned product and the method for recommending products based on the association rule set are implemented.

[0023] The present disclosure utilizes a genetic algorithm to mine association rules for historical transaction data and generates an association rule set for product recommendation, thereby improving the efficiency of generating the association rule set and being able to mine deeper association rules between products. In addition, the present disclosure utilizes the generated association rule set to recommend products, which can provide target customers with more accurate product recommendation solutions, thereby improving user satisfaction and effectively reducing the operating costs of operators in product promotion. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The above contents and other objects, features and advantages of the present disclosure will become more apparent through the following description of the embodiments of the present disclosure with reference to the accompanying drawings, in which:

[0025] Figure 1 It is an application scenario diagram of a method for generating an association rule set for a product and a method for recommending a product based on an association rule set according to an embodiment of the present disclosure;

[0026] Figure 2 is a flowchart of a method for generating an association rule set for a product according to an embodiment of the present disclosure;

[0027] Figure 3 is a flow chart of an association rule mining process based on a genetic algorithm according to an embodiment of the present disclosure;

[0028] Figure 4 is a flow chart of a product recommendation method based on an association rule set according to an embodiment of the present disclosure;

[0029] Figure 5 is a flow chart of a banking financial product recommendation method based on an association rule set according to an embodiment of the present disclosure;

[0030] Figure 6 is a structural schematic diagram of a device for generating an association rule set for a product according to an embodiment of the present disclosure;

[0031] Figure 7 is a structural schematic diagram of a product recommendation device based on an association rule set according to an embodiment of the present disclosure;

[0032] Figure 8 The present invention is a block diagram of an electronic device suitable for implementing a method for generating a product association rule set and a product recommendation method based on an association rule set according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0033] Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present disclosure. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of the embodiments of the present disclosure. However, it is apparent that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessary confusion of the concepts of the present disclosure.

[0034] The terms used herein are only for describing specific embodiments, and are not intended to limit the present disclosure. The terms "including", "comprising", etc. used herein indicate the existence of the features, steps, operations and / or components, but do not exclude the existence or addition of one or more other features, steps, operations or components. All terms (including technical and scientific terms) used herein have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used here should be interpreted as having a meaning consistent with the context of this specification, and should not be interpreted in an idealized or overly rigid manner. In the case of using a statement similar to "at least one of A, B, and C, etc.", it should generally be interpreted according to the meaning of the statement generally understood by those skilled in the art (for example, "a system having at least one of A, B, and C" should include but is not limited to having A alone, having B alone, having C alone, having A and B, having A and C, having B and C, and / or having A, B, C, etc.).

[0035] In the technical solution of the present disclosure, the user information (including but not limited to user personal information, user image information, user device information, such as location information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved are all 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 the 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. In the scenario of using personal information for automated decision-making, the methods, devices and systems provided in the embodiments of the present disclosure provide users with corresponding operation entrances for users to choose to agree or refuse the automated decision results; if the user chooses to refuse, the expert decision-making process is entered. The expression "automated decision-making" here refers to the activity of automatically analyzing and evaluating personal behavior habits, interests and hobbies or economic, health, credit status, etc. through computer programs, and making decisions. The expression "expert decision-making" here refers to the activity of making decisions by people who specialize in a certain field, have specialized experience, knowledge and skills and have reached a certain level of professionalism.

[0036] With the advancement of random technology, the transaction scale of products is getting larger and the transaction frequency is getting higher, thus generating a large amount of transaction data and other data associated with the transaction data. How to mine the massive transaction data and its related data, and find the association rules between products and between products and customers, so as to recommend products to customers more efficiently and accurately, has become the focus of research by relevant technical personnel. Taking the financial industry as an example, influenced by the development of the Internet, the financial industry has seen more diversified product forms and transaction methods. The traditional marketing methods of banks for financial products can no longer adapt to the pressure brought by the increasingly fast business pace and data volume, and are inefficient and cannot achieve customer psychological satisfaction. At present, there are many problems with the product transaction model of the financial industry: it is time-consuming and laborious to use manual methods to screen and recommend massive products to customers, and the customer response effect is poor; it is impossible to identify customer preferences and recommend products that satisfy customers; it is impossible to conduct precise marketing, and it is more difficult to acquire customers.

[0037] At present, the financial industry, such as the banking industry, has accumulated a considerable amount of business data resources in the background. Most banks use data mining methods to analyze and process a considerable amount of business data resources in the background to provide a basis for the bank's decision-making. Therefore, it is necessary to mine association rules for transaction data and other information related to the banking industry, and make efficient and accurate product recommendations to target customers based on the generated association rules, thereby improving the satisfaction of target customers.

[0038] The embodiments of the present disclosure provide a method for generating an association rule set for a product and a product recommendation method based on the association rule set, which are used to improve the efficiency and accuracy of product recommendation.

[0039] Figure 1 It is an application scenario diagram of a method for generating an association rule set for products and a method for recommending products based on an association rule set according to an embodiment of the present disclosure.

[0040] like Figure 1 As shown, the application scenario 100 according to this embodiment may include the fields of big data and artificial intelligence technology. The network 104 is used to provide a medium for a communication link between the first terminal device 101, the second terminal device 102, the third terminal device 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0041] The user can use the first terminal device 101, the second terminal device 102, and the third terminal device 103 to interact with the server 105 through the network 104 to receive or send messages, etc. Various communication client applications can be installed on the first terminal device 101, the second terminal device 102, and the third terminal device 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only for example).

[0042] The first terminal device 101, the second terminal device 102, and the third terminal device 103 may be various electronic devices having display screens and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, and the like.

[0043] The server 105 may be a server that provides various services, such as a background management server (only as an example) that provides support for websites browsed by users using the first terminal device 101, the second terminal device 102, and the third terminal device 103. The background management server may analyze and process the received data such as user requests, and feed back the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal device.

[0044] It should be noted that the product association rule set generation method and the product recommendation method based on the association rule set provided in the embodiment of the present disclosure can generally be executed by the server 105. Accordingly, the product association rule set generation device and the product recommendation device based on the association rule set provided in the embodiment of the present disclosure can generally be set in the server 105. The product association rule set generation method and the product recommendation method based on the association rule set provided in the embodiment of the present disclosure can also be executed by a server or server cluster that is different from the server 105 and can communicate with the first terminal device 101, the second terminal device 102, the third terminal device 103 and / or the server 105. Accordingly, the product association rule set generation device and the product recommendation device based on the association rule set provided in the embodiment of the present disclosure can also be set in a server or server cluster that is different from the server 105 and can communicate with the first terminal device 101, the second terminal device 102, the third terminal device 103 and / or the server 105.

[0045] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0046] The following will be based on Figure 1 The scene described by Figure 2~Figure 5 The disclosed embodiment method is described in detail.

[0047] Figure 2 The present invention is a flowchart of a method for generating an association rule set for a product according to an embodiment of the present disclosure.

[0048] like Figure 2 As shown, the method for generating an association rule set for the above product includes operations S210 to S220.

[0049] In operation S210, historical transaction data of a product is preprocessed to obtain preprocessed transaction data.

[0050] The purpose of the above operation S210 is to obtain reliable historical transaction data, so as to generate a more accurate association rule set for subsequent mining of product association rules using a genetic algorithm.

[0051] In an embodiment of the present disclosure, before obtaining historical transaction data, the user's consent or authorization may be obtained. For example, before operation S210, a request to obtain user information may be issued to the user. When the user agrees or authorizes that the user information may be obtained, the operation S210 is performed. In an embodiment of the present disclosure, a corresponding operation entry may be provided for the user to choose to agree or reject the automated decision result. That is, before preprocessing / decision-making on the user's historical transaction information, an instruction to agree or reject the processing / decision-making entered by the user through the corresponding operation entry may be obtained. If the user agrees to perform the processing / decision-making, the user information is preprocessed / decided, that is, step S210 is executed. If the user refuses to perform the processing / decision-making, the expert decision-making process is entered.

[0052] In operation S220, association rule mining is performed on the preprocessed transaction data and user information corresponding to the transaction data using a predefined association rule mining method to obtain an association rule set, wherein the predefined association rule mining method uses a genetic algorithm.

[0053] The above operations S210 to S220 involve historical transaction data and user information. The following takes the banking industry as an example to further explain the above historical transaction data and user information in detail.

[0054] Table 1 is a transaction details table, which is used as the source data storage table for association rule mining and is the main table for association rule mining; Table 2 is a user / customer information table, which is used to store the natural and social conditions of users / customers; Table 3 is a business category table, which is used to record detailed business attributes; Table 4 is a transaction association table, which is used to record transaction records of the same customer on the same day at similar times.

[0055] Table 1-Transaction Details

[0056]

[0057] Table 2-Customer Information Table

[0058]

[0059] Table 3-Business Types

[0060]

[0061] Table 4 - Transaction association table

[0062]

[0063] To preprocess the historical transaction data, i.e., the data involved in Tables 1, 3, and 4, it is necessary to first classify the bank financial products, and then purify and integrate the transaction data involved in the above tables so as to facilitate mining using the predefined association rule algorithm. The association rule set involved in the present disclosure can output the financial products that the user or target customer is interested in when the relevant information of the user or target customer is input.

[0064] The present disclosure utilizes a genetic algorithm to mine association rules for historical transaction data and generates an association rule set for product recommendation, thereby improving the efficiency of generating the association rule set and being able to mine deeper association rules between products. In addition, the present disclosure utilizes the generated association rule set to recommend products, which can provide target customers with more accurate product recommendation solutions, thereby improving user satisfaction and effectively reducing the operating costs of operators in product promotion.

[0065] According to an embodiment of the present disclosure, the above-mentioned preprocessing of the historical transaction data of the goods to obtain the preprocessed transaction data includes: extracting the storage format using the historical transaction data stored in the transaction database, and analyzing the extracted storage format to obtain a unified storage format; based on the unified storage format, converting the historical transaction data into a plain text format for storage to obtain a temporary storage table; extracting the common components of the products in the temporary storage table by classifying the products, and determining the data irrelevant to the association rule mining based on the common components; deleting the data irrelevant to the association rule mining operation in the temporary storage table to obtain the preprocessed transaction data.

[0066] The following uses the banking industry as a specific example to further illustrate the preprocessing process of transaction data involved in the present disclosure.

[0067] There are many names of bank financial products, which are updated quickly and have complex business natures. Only by classifying and merging them can we complete association rule mining and business recommendations more efficiently. First, we need to scientifically classify electronic banking products and extract the components of each product. All products are classified according to the current nature of banking business operations. They mainly include wealth management, funds, insurance, deposits, precious metals, credit cards, inclusive finance, and public accounts.

[0068] Bank data involves multiple business platforms and exists in different data formats in multiple databases. Based on product classification, the data is uniformly processed into the same data structure in the same database. Data purification retains the bank's mainstream customer data and removes invalid data. After processing the valid data of different structures from multiple channels separately, import them into the mining table according to unified standards.

[0069] Principle of amount purification for transaction association table data integration: By analyzing historical data, find out the transaction records of the same customer in the same day and similar period, as well as a record in the transaction association table, for easy mining. For example, the following operations can be used for data preprocessing: First, since bank data is not stored in a centralized manner and the data format is inconsistent, extract commonalities, unify the format, establish a temporary data table, and convert multi-data source data from different types of databases into plain text format. Secondly, delete the data in the temporary storage table that is not related to association rule mining, such as: non-product purchase transactions, internal management transactions. Finally, fill the transaction details table and transaction association table with data.

[0070] Through the above historical transaction preprocessing operations, accurate transaction information can be obtained, providing a reliable data basis for subsequent association rule mining.

[0071] According to an embodiment of the present disclosure, the above-mentioned use of the predefined association rule mining method to mine association rules on preprocessed transaction data and user information corresponding to the transaction data to obtain an association rule set includes: determining the target data item used for the association rule mining operation from the preprocessed transaction data and user information; defining the support and credibility on which the association rule mining operation depends based on the target data item, and determining the minimum support threshold and the minimum credibility threshold; performing parameter initialization operation on the genetic algorithm used in the predefined association rule mining method based on the preprocessed transaction data and user information; randomly generating an initial population used by the genetic algorithm, and based on the initial population, using the genetic algorithm to mine association rules to generate initial association rules; screening the initial association rules based on the minimum support threshold and the minimum credibility threshold, and constructing an association rule set with the screened initial association rules; iteratively performing the initial association rule generation operation and the screening operation until the preset conditions are met, terminating the association rule mining operation, and obtaining the association rule set.

[0072] The following is a further detailed description of the association rule mining process based on genetic algorithm involved in the present disclosure through specific embodiments.

[0073] Association rules related definitions: Association rules refer to the rules that exist between sample data items, and also refer to the associations between different data items that appear in the same event. represents all data item sets, where Is a data item. Transaction data item ,in Is a transaction data set data items, and An association rule can be expressed as The implication relationship of , ,and . Association rules involve two important technical concepts: support and credibility.

[0074] Support : It indicates the event In the entire transaction set The prevalence of Zhongyou The record supports a subset of , expressed as shown in formula (1):

[0075] (1).

[0076] Credibility : Indicates the credibility of the conclusion caused by the rule premise, that is, Supported Transaction Sets Among the affairs The transaction also supports transaction set , as shown in formula (2):

[0077] (2).

[0078] By discovering association rules, we can find reliable and representative rules, so we need to give a minimum credibility threshold and a minimum support threshold. The problem of association rule mining evolves into mining all association rules whose support and credibility exceed the specified thresholds.

[0079] According to an embodiment of the present disclosure, the above-mentioned parameter initialization operation of the genetic algorithm used in the predefined association rule mining method includes: initializing the encoding length parameter, initial crossover probability parameter, initial mutation probability parameter, initial population size parameter and iteration round parameter of the genetic algorithm.

[0080] The above parameters are used to determine the initial state of the genetic algorithm, and those skilled in the art can set other parameters and initial values ​​according to their own needs.

[0081] According to an embodiment of the present disclosure, the above-mentioned randomly generated initial population used by the genetic algorithm, and based on the initial population, the genetic algorithm is used to mine association rules, and the generation of initial association rules includes: using support and credibility to set a fitness function suitable for the genetic algorithm, and determining the value range of the fitness function; based on the initial population, the genetic algorithm is used to generate original association rules, and the original association rules are cross-evolution and mutation evolution operations are performed; based on the value range of the fitness function, the evolved original association rules are screened to obtain the initial association rules.

[0082] The generation process of the initial association rules involved in the present disclosure is further described in detail below through specific embodiments.

[0083] The association rule mining method based on genetic algorithm, in terms of genetic coding: when using genetic algorithm for knowledge discovery, it is necessary to solve the chromosome coding problem. Chromosomes are generally represented by strings, and genetic factors are represented by elements at various positions on the string. The present disclosure uses the encoding method of real number arrays, and the number of elements in the array corresponds to the number of fields in the business database, as shown in Table 5. The attribute value of the field is the element value.

[0084] Table 5 - Field and attribute table

[0085]

[0086] The business database in Table 5 can be encoded using a real number array, where A[1] represents field 1, A[2] represents field 2, ..., A[n] represents field n; if i represents the attribute value i of the field, then the element value of the array A[R] is the attribute value of the corresponding field, where the value 0 indicates that this field is irrelevant to other fields. The encoding form of a single entity of the database is shown in Table 6.

[0087] Table 6 - Genetic encoding of association rules

[0088]

[0089] In the design of the fitness function, the design of the fitness function should be considered comprehensively. In this disclosure, the fitness is defined as , which reflects the comprehensive result of credibility and support.

[0090] In the selection operation, the support value determines the accuracy of the association rule. The support shows how representative the rule is in the data group. The higher the support, the more relevant the rule is. Through mining, it is found that some rules have high confidence but low support, which means that the usage rate of the rule is low, so this rule is not applicable to the data group. The rules must be selected to keep the rules with high confidence and support. The selection operation is to avoid data gene loss and improve the global convergence of the calculation results. That is, only the rule data that meets the requirements can survive. When selecting an operation, it is necessary to comprehensively consider the support and credibility of the operation. It can only be performed when both meet certain conditions. Since both support and credibility are expressed in percentage values, .

[0091] According to an embodiment of the present disclosure, the above-mentioned method of generating original association rules based on the initial population using a genetic algorithm and performing crossover evolution and mutation evolution operations on the original association rules includes: classifying the attributes of the initial population, and screening the initial population after attribute classification using the value range of the fitness function to generate original association rules; performing crossover evolution on the original association rules based on adaptive random dynamic adjustment of the crossover probability to obtain the original association rules after crossover evolution; performing mutation evolution on the original association rules after crossover evolution based on adaptive random dynamic adjustment of the mutation probability to obtain the original association rules after mutation evolution.

[0092] The crossover evolution and variation evolution involved in the present disclosure are further described in detail below through specific examples.

[0093] Crossover operator: In the present disclosure, the crossover operator uses single-point crossover, integrates the elite recombination algorithm to select the offspring and the parent after the crossover, and uses the crossover operation to evolve the set of rules.

[0094] Determination of mutation operator: The mutation operator adopted in the present disclosure selects single-point mutation and uses mutation operation to evolve the set of rules.

[0095] Conditions for obtaining ideal rules of genetic algorithms: A new generation of rules is generated through selection operations. After several iterations, when the algorithm meets the termination conditions, a set of ideal rules will be obtained. The generated rules are used to process the business data in the database, delete the records involved in the rules, and repeat the above operations on the remaining data to obtain a new set of rules.

[0096] The confidence and support of the results are used to extract the newly generated rules. The extraction criterion is: output all rules that meet the user-given confidence threshold requirements and discard the rest.

[0097] According to an embodiment of the present disclosure, the above-mentioned preset conditions include an iteration round determined based on the total amount of pre-processed transaction data and user information or an iteration round set by the user.

[0098] The genetic algorithm iterates until all instances are covered or the preset genetic algorithm termination condition is met.

[0099] The termination conditions of the genetic algorithm in the present disclosure are exemplary, and those skilled in the art can set other termination conditions according to their own needs.

[0100] The following pseudo code is used as a specific embodiment and combined with the attached Figure 3 The association rule mining process based on genetic algorithm involved in the present disclosure is further described in detail.

[0101] Figure 3 It is a flowchart of an association rule mining process based on a genetic algorithm according to an embodiment of the present disclosure.

[0102] like Figure 3 As shown, the program pseudo code is as follows:

[0103] Read in data;

[0104] Initialize genetic algorithm parameters;

[0105] Randomly generate the initial population;

[0106] While (the current iteration number is less than the total iteration number)

[0107] {Evaluate the fitness of all genes;

[0108] Replicate a certain number of elites and other individuals selected through the gambling wheel to form a new generation of groups;

[0109] A certain number of chromosomes are randomly selected with the crossover probability to perform the crossover operation;

[0110] A certain number of chromosomes are randomly selected to perform mutation operations based on the mutation probability;

[0111] The current iteration count increases by 1;}

[0112] Evaluate the fitness of all genes; extract rules based on the set support and confidence thresholds.

[0113] Figure 4 The present invention is a flowchart of a product recommendation method based on an association rule set according to an embodiment of the present disclosure.

[0114] like Figure 4 As shown, the above-mentioned product recommendation method based on the association rule set includes operations S410 to S420.

[0115] Operation S410 , classifying target customers to obtain customer classification results, and generating initial product recommendation results based on the customer classification results using an association rule set, wherein the association rule set is generated by the association rule set generation method of the above product.

[0116] In an embodiment of the present disclosure, before obtaining historical transaction data, the user's consent or authorization may be obtained. In an embodiment of the present disclosure, a corresponding operation portal may be provided for the user to choose to agree or reject the automated decision result. That is, before preprocessing / decision-making on the user's historical transaction information, an instruction to agree or reject the processing / decision-making input by the user through the corresponding operation portal may be obtained. If the user agrees to perform the processing / decision-making, the user information is preprocessed / decided, that is, step S410 is executed. If the user refuses to perform the processing / decision-making, the expert decision-making process is entered.

[0117] Operation S420: analyzing the initial product recommendation result to obtain an analysis result, and recommending products to target customers based on the analysis result and a preset interference condition.

[0118] The disclosure overcomes the problem that the traditional way of recommending financial products to customers by banks is time-consuming, labor-intensive, and cannot be accurately marketed, and provides a bank customer product recommendation system that integrates artificial intelligence algorithms. The disclosure first uses the bank's rich data resources such as customer and product information, transaction information, APP behavior information, etc. to group all customers according to the transactions and positions of different customers. Secondly, corresponding recall and sorting strategies are designed for different customer groups. Finally, in the sorting stage, an integrated learning model is selected to further predict and score the recalled financial products, and the TOP10 (adjustable) products with the predicted probability corresponding to each customer are selected as the recommendation results, so as to realize personalized recommendations for each customer. This can achieve accurate marketing of bank financial products and improve business efficiency and business volume.

[0119] According to an embodiment of the present disclosure, the above-mentioned classification of target customers to obtain customer classification results, and based on the customer classification results, using an association rule set to generate initial product recommendation results includes: based on the authorization of the target customers, obtaining the target customer information through transaction channels, and dividing the target customers into a first category of target customers and a second category of target customers; for the first category of target customers, based on the acquired attribute information of the first category of target customers, generating initial product recommendation results for the first category of target customers from the association rule set; for the second category of target customers, determining the products with the largest transaction volume for the second category of target customers based on the acquired historical transaction information of the second category of target customers, and generating initial product recommendation results for the second category of target customers from the association rule set based on the products with the largest transaction volume.

[0120] Classifying target customers and adaptively applying association rule sets based on the classification results can better provide product recommendations for target customers. It should be noted that the above acquisition and processing of target customer personal information has been authorized by the target customer.

[0121] According to an embodiment of the present disclosure, the above-mentioned recommending products to target customers based on analysis results and preset interference conditions includes: when the analysis result is that there is only one product in the initial product recommendation results, recommending to the target customer through the transaction channel; when the analysis result is that there are multiple products in the initial product recommendation results, setting the preset interference condition according to the needs of the product operator, and screening the initial product recommendation results based on the preset interference condition, and recommending the screened products to the target customer through the transaction channel.

[0122] In the above-mentioned embodiments of the present disclosure, if the initial product recommendation results include multiple product recommendations, that is, multiple products that target customers may be interested in, the initial product recommendation results can be further processed according to the interference conditions preset by the operator, and the conditions of concern to the operator are used as interference items to screen and filter the initial product recommendation results, and the screened and filtered products are recommended to the target customers. Taking the banking industry as an example, when the initial product recommendation results include multiple financial products, the financial products can be further screened according to the needs of the bank, for example, popular or high-yield financial products can be recommended to the target customers.

[0123] The following is a specific example of the banking industry, and combined with the attached Figure 5 Tables 7 to 15 further illustrate the process of applying the association rule set to recommend financial products in the present disclosure.

[0124] Figure 5 It is a flowchart of a banking financial product recommendation method based on an association rule set according to an embodiment of the present disclosure.

[0125] Association rule mining is to mine the relationships between some seemingly unrelated data in the banking industry. This patent concept combines data mining technology. On the basis of data collection and organization, it then establishes an association rule mining model based on genetic algorithms, and finally derives relevant rules to help bank branch account managers accurately recommend suitable products to customers.

[0126] In the association rules, the product name uses the classified product code. The product code is divided into three categories according to business risk: G (high risk), Z (low risk), and D (low risk).

[0127] (1) By mining the transaction association table, we can find the purchase associations between different products. That is, we can predict what products the customer will buy based on the products they have currently purchased.

[0128] The rule examples are: buy product G1 => buy product G4; buy product G2 => buy product G4; buy product Z1 => buy product Z3; buy product D2 => buy product D3.

[0129] (2) Predict what products customers will buy based on their natural and social attributes.

[0130] Example of derived rules: Age 20-30 => Purchase product G1; 4000-5000 yuan / month, age 30-40 => Purchase product G2; 3000-4000 yuan / month, male => Purchase product Z1; Age 50-60 => Purchase product D1.

[0131] Based on the mined association rules, combined with the bank's own interest in different products and the customer's credit record, a product recommendation sequence is derived, and then the recommendation of electronic banking products is comprehensively considered.

[0132] Product recommendations are based on the following principles: products with high interest from the bank are given priority, and products are recommended based on the association rules that have been mined.

[0133] In terms of product recommendation association rule mining:

[0134] Data processing and genetic algorithm coding: The database of customer attributes and customer products is aggregated and cleaned, and the attributes (fields) are selected to establish an information database table, as shown in Table 7.

[0135] Table 7-Customer Information Table

[0136]

[0137] Since the present disclosure adopts a real number array encoding method, a real number encoding string is used to represent a rule, and the real number string includes two parts: the premise (the characteristic part of the rule) and the conclusion (the rule category part). For example, 124432->2 means that the attribute value 124432 of fields 1 to 6 can be used to obtain the attribute value 2 of field 7.

[0138] The following maps multiple attributes into Table 8.

[0139] Table 8 - Attribute mapping table

[0140]

[0141] Encode the gender attribute. There are only 2 data. Encode according to the rules in Table 9.

[0142] Table 9- Gender coding table

[0143]

[0144] The age attribute encoding method is shown in Table 10.

[0145] Table 10 - Age coding table

[0146]

[0147] Income is divided into 6 categories and the codes are shown in Table 11.

[0148] Table 11 - Income Code Table

[0149]

[0150] Educational qualifications are divided into four levels, and the codes are shown in Table 12.

[0151] Table 12- Educational qualification coding table

[0152]

[0153] Assets are divided into four levels and the codes are shown in Table 13.

[0154] Table 13-Asset Coding Table

[0155]

[0156] Bank products are divided into 8 levels, and the codes are shown in Table 14.

[0157] Table 14-Bank product coding table

[0158]

[0159] In this way, the attributes in Table 8 are all converted into integer values, and the mapping results of the database table are shown in Table 15.

[0160] Table 15 - Mapping results table

[0161]

[0162] Genetic algorithm is about parameter setting and result analysis. In association rule mining based on genetic algorithm, the parameters that need to be set when the program is running include: coding length L, crossover probability Pc, population size M, termination algebra T, mutation probability Pm, etc.

[0163] (1) Coding length: This disclosure adopts a real number array coding scheme. The coding length is the number of fields to be analyzed. According to data sampling, the value in this example is set to 5. (2) Crossover probability: The crossover operation is the main operator of the algorithm, so the value of the crossover probability should be selected to be larger. However, if the crossover probability is too large, it is easy to have an adverse effect on the evolutionary operation. If the value is small, the speed of generating a new generation of individuals will slow down. Therefore, after comprehensive analysis, this example sets Pc to 0.8. (3) Population size: The population size (M) represents the number of individuals contained in the data population. When the M value is large, the computational efficiency of the genetic algorithm will decrease. If the M value is too small, although it will increase the computational speed, it will reduce the diversity of the population and easily cause the algorithm to mature prematurely. In this example, the population size M is set to 100 based on comprehensive considerations. (4) Mutation probability: The mutation probability Pm also needs to be considered comprehensively. If the value is too large, not only will more new individuals be generated, but also many good patterns will be destroyed, resulting in poor algorithm performance. If the value is too small, the effect of suppressing premature maturation will not be obvious, and the ability of mutation operation to generate new offspring will be weakened. Therefore, it is set to 0.05 in this example. (5) Termination algebra: The termination algebra T is the conditional parameter for terminating the algorithm. The genetic algorithm will stop the operation after reaching the termination algebra and output the best individual in the current population as the optimal solution. In this example, what is sought is a set of rules that meet the user's given threshold, not the optimal solution, so the termination algebra is to terminate the operation when there is no rule less than the user threshold.

[0164] In the process of mining the designed association rule algorithm, the minimum support is set to 0.08 and the minimum confidence is set to 0.6. The final mining results can be obtained according to the above genetic algorithm, as shown in Table 16.

[0165] Table 16 - Association rule mining results

[0166]

[0167] The above genetic algorithm and some of the rules shown in Table 16 are analyzed:

[0168] Rule 1: <25124> = <2> (Support 0.23, Confidence 0.962)

[0169] That is, <Gender: Female, Age: 50-60 years old, Income: 5,000 yuan>, Education: Bachelor's degree, Total assets: 200,000-500,000 yuan <Suitable bank products: Funds; Meaning of the rules: Female customers, aged between 50 and 60, with a salary of around 5,000 yuan, a bachelor's degree, and total assets between 200,000 and 500,000 yuan, the most suitable bank product is a fund.

[0170] Rule 2: <25124> = <4> (Support 0.21, Confidence 0.851)

[0171] That is, <Gender: Female, Age: 50-60 years old, Income: 5,000 yuan>, Education: Bachelor degree, Total assets: 200,000-500,000 yuan <Second most suitable bank product: Time deposit; Meaning of the rule: Female customer, aged between 50-60 years old, salary of around 5,000 yuan, bachelor degree, total assets between 200,000-500,000 yuan, the second most suitable bank product is time deposit.

[0172] Bank account managers can Figure 5 The process shown applies the association rule set and recommends banking financial products to target customers. First, the transaction records in the transaction database are preprocessed to remove useless data and generate a database for association rule mining. Second, the association rule mining algorithm is applied to the preprocessed database to obtain the association rule set and store it in the business association database. Finally, the association rules are applied to recommend banking products based on the attributes of new customers and the consumption records of existing customers, combined with the importance the bank attaches to different products.

[0173] The technical solution disclosed in the present invention uses the association rule discovery and knowledge extraction of genetic algorithms to find valuable rules, and mines the relationship between some seemingly unrelated data in the banking industry. Based on the massive data accumulated in the development of banking business, the present invention uses genetic algorithm association rules to deeply process it, and analyzes it from multiple angles and directions. Through induction, analysis, and processing, it discovers what people don’t know in advance, and finds the inherent relationship and potentially valuable association rules in the data, from which some implicit laws and knowledge are discovered and obtained, and business data are analyzed and counted, and comprehensive calculations are performed, and future business events are inferred and predicted. Analysis. Thereby helping the banking industry to support marketing decisions in the business process. Ultimately, it can meet customer needs and better provide customers with high-quality and convenient services.

[0174] Figure 6 It is a structural diagram of a device for generating an association rule set for a product according to an embodiment of the present disclosure.

[0175] like Figure 6 As shown, the product association rule set generation device 600 includes a data preprocessing module 610 and an association rule mining module 620 .

[0176] The data preprocessing module 610 is used to preprocess the historical transaction data of the product to obtain preprocessed transaction data. In one embodiment, the data preprocessing module 610 can be used to perform the operation S210 described above, which will not be repeated here.

[0177] The association rule mining module 620 is used to perform association rule mining on the preprocessed transaction data and the user information corresponding to the transaction data using a predefined association rule mining method to obtain an association rule set, wherein the predefined association rule mining method uses a genetic algorithm. In one embodiment, the association rule mining module 620 can be used to perform the operation S220 described above, which will not be repeated here.

[0178] Figure 7 It is a structural diagram of a product recommendation device based on an association rule set according to an embodiment of the present disclosure.

[0179] like Figure 7 As shown, the above-mentioned product recommendation device 700 based on association rule sets includes a classification and generation module 710 and an analysis and recommendation module 720 .

[0180] The classification and generation module 710 is used to classify target customers, obtain customer classification results, and generate initial product recommendation results based on the customer classification results using an association rule set, wherein the association rule set is generated by the association rule set generation method of the above-mentioned product; in one embodiment, the classification and generation module 710 can be used to perform the operation S410 described above, which will not be repeated here.

[0181] The analysis and recommendation module 720 is used to analyze the initial product recommendation results, obtain analysis results, and recommend products to target customers based on the analysis results and preset interference conditions. In one embodiment, the analysis and recommendation module 720 can be used to perform the operation S420 described above, which will not be repeated here.

[0182] According to an embodiment of the present disclosure, any multiple modules in the data preprocessing module 610 and the association rule mining module 620 (or the classification and generation module 710 and the analysis and recommendation module 720) can be combined into one module for implementation, or any one of the modules can be split into multiple modules. Alternatively, at least part of the functions of one or more of these modules can be combined with at least part of the functions of other modules and implemented in one module. According to an embodiment of the present disclosure, at least one of the data preprocessing module 610 and the association rule mining module 620 (or the classification and generation module 710 and the analysis and recommendation module 720) can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application specific integrated circuit (ASIC), or can be implemented by hardware or firmware such as any other reasonable way of integrating or packaging the circuit, or implemented in any one of the three implementation methods of software, hardware and firmware or in an appropriate combination of any of them. Alternatively, at least one of the data preprocessing module 610 and the association rule mining module 620 (or the classification and generation module 710 and the analysis and recommendation module 720) may be at least partially implemented as a computer program module, which may perform corresponding functions when executed.

[0183] Figure 8 The present invention is a block diagram of an electronic device suitable for implementing a method for generating a product association rule set and a product recommendation method based on an association rule set according to an embodiment of the present disclosure.

[0184] like Figure 8 As shown, the electronic device 800 according to an embodiment of the present disclosure includes a processor 801, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 802 or a program loaded from a storage part 808 to a random access memory (RAM) 803. The processor 801 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or a related chipset and / or a dedicated microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 801 may also include an onboard memory for caching purposes. The processor 801 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present disclosure.

[0185] In RAM 803, various programs and data required for the operation of electronic device 800 are stored. Processor 801, ROM 802 and RAM 803 are connected to each other via bus 804. Processor 801 performs various operations of the method flow according to the embodiment of the present disclosure by executing the program in ROM 802 and / or RAM 803. It should be noted that the program can also be stored in one or more memories other than ROM 802 and RAM 803. Processor 801 can also perform various operations of the method flow according to the embodiment of the present disclosure by executing the program stored in the one or more memories.

[0186] According to an embodiment of the present disclosure, the electronic device 800 may further include an input / output (I / O) interface 805, which is also connected to the bus 804. The electronic device 800 may further include one or more of the following components connected to the input / output (I / O) interface 805: an input portion 806 including a keyboard, a mouse, etc.; an output portion 807 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage portion 808 including a hard disk, etc.; and a communication portion 809 including a network interface card such as a LAN card, a modem, etc. The communication portion 809 performs communication processing via a network such as the Internet. A drive 810 is also connected to the input / output (I / O) interface 805 as needed. A removable medium 811, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 810 as needed, so that the computer program read therefrom is installed into the storage portion 808 as needed.

[0187] The present disclosure also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or may exist independently without being assembled into the device / apparatus / system. The above computer-readable storage medium carries one or more programs, and when the above one or more programs are executed, the method according to the embodiment of the present disclosure is implemented.

[0188] According to an embodiment of the present disclosure, the computer-readable storage medium may be a non-volatile computer-readable storage medium, for example, it may include but is not limited to: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, an apparatus or a device. For example, according to an embodiment of the present disclosure, the computer-readable storage medium may include the ROM 802 and / or RAM 803 described above and / or one or more memories other than ROM 802 and RAM 803.

[0189] The embodiment of the present disclosure also includes a computer program product, which includes a computer program, and the computer program contains program code for executing the method shown in the flowchart. When the computer program product is run in a computer system, the program code is used to enable the computer system to implement the product association rule set generation method and product recommendation method based on the association rule set provided by the embodiment of the present disclosure.

[0190] The above functions defined in the system / device of the embodiment of the present disclosure are executed when the computer program is executed by the processor 801. According to the embodiment of the present disclosure, the system, device, module, unit, etc. described above can be implemented by a computer program module.

[0191] In one embodiment, the computer program may rely on tangible storage media such as optical storage devices, magnetic storage devices, etc. In another embodiment, the computer program may also be transmitted and distributed in the form of signals on a network medium, and downloaded and installed through the communication part 809, and / or installed from a removable medium 811. The program code contained in the computer program may be transmitted using any appropriate network medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0192] In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 809, and / or installed from the removable medium 811. When the computer program is executed by the processor 801, the above functions defined in the system of the embodiment of the present disclosure are performed. According to the embodiment of the present disclosure, the system, device, means, module, unit, etc. described above can be implemented by a computer program module.

[0193] According to an embodiment of the present disclosure, the program code for executing the computer program provided by the embodiment of the present disclosure can be written in any combination of one or more programming languages. Specifically, these computing programs can be implemented using high-level process and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, Java, C++, python, "C" language or similar programming languages. The program code can be executed entirely on the user computing device, partially on the user device, partially on the remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, using an Internet service provider to connect through the Internet).

[0194] The flow charts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0195] Those skilled in the art will appreciate that the features described in the various embodiments of the present disclosure may be combined and / or combined in a variety of ways, even if such combinations or combinations are not explicitly described in the present disclosure. In particular, without departing from the spirit and teachings of the present disclosure, the features described in the various embodiments of the present disclosure may be combined and / or combined in a variety of ways. All of these combinations and / or combinations fall within the scope of the present disclosure. The embodiments of the present disclosure are described above. However, these embodiments are for illustrative purposes only and are not intended to limit the scope of the present disclosure. Although the embodiments are described above separately, this does not mean that the measures in the various embodiments cannot be used in combination to advantage. Without departing from the scope of the present disclosure, those skilled in the art may make a variety of substitutions and modifications, which should all fall within the scope of the present disclosure.

Claims

1. A method for generating an association rule set for a product, characterized in that: The method comprises: Preprocess the historical transaction data of the product to obtain preprocessed transaction data; The pre-processed transaction data and user information corresponding to the transaction data are subjected to association rule mining by using a pre-defined association rule mining method to obtain an association rule set, wherein the pre-defined association rule mining method uses a genetic algorithm.

2. The method according to claim 1, characterized in that The historical transaction data of the commodity is preprocessed, and the preprocessed transaction data includes: The storage format is extracted using the historical transaction data stored in the transaction database, and the extracted storage format is analyzed to obtain a unified storage format; Based on the unified storage format, the historical transaction data is converted into a plain text format for storage to obtain a temporary storage table; Extracting common components of the products in the temporary storage table by classifying the products, and determining data irrelevant to association rule mining based on the common components; The data irrelevant to the association rule mining operation in the temporary storage table is deleted to obtain the preprocessed transaction data.

3. The method according to claim 1, characterized in that The pre-processed transaction data and the user information corresponding to the transaction data are subjected to association rule mining using a pre-defined association rule mining method, and the obtained association rule set includes: Determining target data items used for performing association rule mining operations from the preprocessed transaction data and the user information; Defining the support and credibility on which the association rule mining operation depends based on the target data item, and determining a minimum support threshold and a minimum credibility threshold; Based on the preprocessed transaction data and the user information, performing a parameter initialization operation on the genetic algorithm used in the predefined association rule mining method; Randomly generate an initial population used by the genetic algorithm, and based on the initial population, use the genetic algorithm to perform association rule mining to generate initial association rules; Screening the initial association rules based on the minimum support threshold and the minimum credibility threshold, and constructing the association rule set with the screened initial association rules; The initial association rule generation operation and the screening operation are iterated until a preset condition is met, and the association rule mining operation is terminated to obtain the association rule set.

4. The method according to claim 3, characterized in that The parameter initialization operation of the genetic algorithm used in the predefined association rule mining method includes: The encoding length parameter, initial crossover probability parameter, initial mutation probability parameter, initial population size parameter and iteration round parameter of the genetic algorithm are initialized.

5. The method according to claim 3, characterized in that: Randomly generate an initial population used by the genetic algorithm, and based on the initial population, use the genetic algorithm to mine association rules, and generate initial association rules including: Using the support and the credibility to set a fitness function applicable to the genetic algorithm, and determining a value range of the fitness function; Based on the initial population, using the genetic algorithm to generate original association rules, and performing crossover evolution and mutation evolution operations on the original association rules; The evolved original association rules are screened based on the value range of the fitness function to obtain the initial association rules.

6. The method according to claim 5, characterized in that Based on the initial population, using the genetic algorithm to generate original association rules, and performing cross evolution and mutation evolution operations on the original association rules include: Classifying the attributes of the initial population, and screening the initial population after the attribute classification using the value range of the fitness function to generate the original association rule; Performing crossover evolution based on adaptive random dynamic adjustment of crossover probability on the original association rules to obtain the original association rules after crossover evolution; The original association rules after the cross evolution are subjected to mutation evolution based on adaptive random dynamic adjustment of mutation probability to obtain the original association rules after mutation evolution.

7. The method according to claim 3, characterized in that The preset conditions include an iteration round determined based on the preprocessed transaction data and the total amount of data of the user information or an iteration round set by the user.

8. A product recommendation method based on association rule sets, characterized in that: The method comprises: Classifying target customers to obtain customer classification results, and based on the customer classification results, generating initial product recommendation results using an association rule set, wherein the association rule set is generated by any one of the methods of claims 1 to 7; The initial product recommendation result is analyzed to obtain an analysis result, and based on the analysis result and a preset interference condition, a product is recommended to the target customer.

9. The method according to claim 8, characterized in that Classifying target customers to obtain customer classification results, and based on the customer classification results, generating initial product recommendation results using an association rule set includes: Based on the authorization of the target customers, obtaining the information of the target customers through the transaction channel, and dividing the target customers into the first category of target customers and the second category of target customers; For the first type of target customers, based on the acquired attribute information of the first type of target customers, generating initial product recommendation results for the first type of target customers from the association rule set; For the second type of target customers, the products with the largest transaction volume for the second type of target customers are determined based on the acquired historical transaction information of the second type of target customers, and initial product recommendation results for the second type of target customers are generated from the association rule set based on the products with the largest transaction volume.

10. The method according to claim 8, characterized in that Based on the analysis results and the preset intervention conditions, the products recommended to the target customers include: If the analysis result shows that there is only one product in the initial product recommendation result, recommending it to the target customer through a transaction channel; When the analysis result shows that there are multiple products in the initial product recommendation results, the preset interference condition is set according to the needs of the product operator, and the initial product recommendation results are screened based on the preset interference condition, and the screened products are recommended to the target customers through transaction channels.

11. A product association rule set mining device, characterized in that: The device comprises: A data preprocessing module is used to preprocess the historical transaction data of the product to obtain preprocessed transaction data; The association rule mining module is used to perform association rule mining on the preprocessed transaction data and user information corresponding to the transaction data using a predefined association rule mining method to obtain an association rule set.

12. A product recommendation device based on an association rule set, characterized in that: The device comprises: A classification and generation module, used to classify target customers, obtain customer classification results, and generate initial product recommendation results based on the customer classification results using an association rule set, wherein the association rule set is mined based on the method described in any one of claims 1 to 7; The analysis and recommendation module is used to analyze the initial product recommendation results, obtain analysis results, and recommend products to the target customers based on the analysis results and preset interference conditions.

13. An electronic device, comprising: one or more processors; a memory for storing one or more computer programs, It is characterized in that the one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1 to 10.

14. A computer-readable storage medium having a computer program or instruction stored thereon, characterized in that: When the computer program or instruction is executed by a processor, the steps of the method according to any one of claims 1 to 10 are implemented.

15. A computer program product comprising a computer program or instructions, characterized in that When the computer program or instruction is executed by a processor, the steps of the method according to any one of claims 1 to 10 are implemented.

Images