Method and device for pushing personal pension information

Abstract

The invention discloses a personal pension information pushing method and device, and relates to the technical field of artificial intelligence, and the method comprises the steps: obtaining customer information; customer information is encrypted and packaged into a transaction, and the transaction is broadcasted to a block chain network; after the node in the block chain network passes the transaction consensus verification, if the personal pension account opening information of the customer indicates that the customer does not open the personal pension account, inputting the encrypted customer information into a personal pension account opening prediction model, and outputting a personal pension account opening recommendation result; according to the encrypted customer information, matching personal pension information push rule data of a customer group to which the customer belongs; and pushing personal pension account opening recommendation information to the client according to the matched personal pension information pushing rule data. According to the method, the potential customers who intend to open the personal pension account can be mined, the calculation amount is reduced, the occupied space of a memory is reduced, and the efficiency and accuracy of personal pension information pushing are improved.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a method and device for pushing personal pension information. Background Art

[0002] This section is intended to provide a background or context to the embodiments of the invention recited in the claims. No admission is made that the description herein is prior art by inclusion in this section.

[0003] With the popularization of pensions, the supply of various types of personal pension products is becoming increasingly abundant. At present, personal pensions are attracting more and more attention, and the opening of personal pension accounts is on a growing trend, resulting in increasingly fierce competition in personal pension products. Existing technical methods have the problems of large computational complexity, large memory usage and low accuracy, and cannot efficiently and accurately push personal pension information and mine potential customers who are interested in opening personal pension accounts. Therefore, a personal pension account opening recommendation method is urgently needed to solve the above problems. Summary of the invention

[0004] The embodiment of the present invention provides a method for pushing personal pension information, which is used to mine potential customers who intend to open personal pension accounts, reduce the amount of calculation, reduce the memory space occupied, and improve the efficiency and accuracy of pushing personal pension information. The method includes:

[0005] Obtaining customer information, including the customer's personal pension account opening information, identity information, and financial information; the customer's personal pension account opening information indicates whether the customer has opened a personal pension account;

[0006] Encrypt customer information and encapsulate it into transactions, which are then broadcast to the blockchain network;

[0007] After the nodes in the blockchain network have verified the transaction consensus, if the customer's personal pension account opening information indicates that the customer has not opened a personal pension account, the encrypted customer information is input into the personal pension account opening prediction model, and the personal pension account opening recommendation result is output;

[0008] If the result of the recommendation for opening a personal pension account is to recommend opening a personal pension account, the personal pension information push rule data of the customer group to which the customer belongs is matched based on the encrypted customer information;

[0009] Based on the matching personal pension information push rule data, personal pension account opening recommendation information is pushed to customers.

[0010] The embodiment of the present invention also provides a personal pension information push device, which is used to mine potential customers who intend to open a personal pension account, reduce the amount of calculation, reduce the memory space occupied, and improve the efficiency and accuracy of personal pension information push. The device includes:

[0011] The customer information acquisition module is used to acquire customer information, including the customer's personal pension account opening information, identity information, and financial information; the customer's personal pension account opening information indicates whether the customer has opened a personal pension account;

[0012] The customer information encryption module is used to encrypt customer information and encapsulate it into transactions, which are then broadcast to the blockchain network;

[0013] A module for obtaining the recommendation result of opening a personal pension account, which is used to input the encrypted customer information into the prediction model for opening a personal pension account and output the recommendation result of opening a personal pension account after the nodes in the blockchain network have verified the transaction consensus and if the customer's personal pension account opening information indicates that the customer has not opened a personal pension account;

[0014] A personal pension information push rule data acquisition module is used to match the personal pension information push rule data of the customer group to which the customer belongs according to the encrypted customer information if the personal pension account opening recommendation result is a recommendation to open a personal pension account;

[0015] The personal pension account opening recommendation information push module is used to push personal pension account opening recommendation information to customers based on the matching personal pension information push rule data.

[0016] An embodiment of the present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned personal pension information push method when executing the computer program.

[0017] An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned personal pension information pushing method is implemented.

[0018] An embodiment of the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the above-mentioned personal pension information push method is implemented.

[0019] In an embodiment of the present invention, by obtaining customer information, the customer information includes the customer's personal pension account opening information, identity information, and financial information; the customer's personal pension account opening information indicates whether the customer has opened a personal pension account; the customer information is encrypted and packaged into a transaction, and broadcast to the blockchain network; after the nodes in the blockchain network pass the transaction consensus verification, if the customer's personal pension account opening information indicates that the customer has not opened a personal pension account, the encrypted customer information is input into the personal pension account opening prediction model, and the personal pension account opening recommendation result is output; if the personal pension account opening recommendation result is a recommendation to open a personal pension account, according to the encrypted customer information, the personal pension information push rule data of the customer group to which the customer belongs is matched; according to the matched personal pension information push rule data, the personal pension account opening recommendation information is pushed to the customer. In the above process, the embodiment of the present invention mines potential customers who intend to open a personal pension account through the personal pension account opening prediction model, thereby reducing the amount of calculation, reducing the memory occupied space, and improving the efficiency and accuracy of personal pension information push. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. 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. In the drawings:

[0021] Figure 1 This is a flow chart of a method for pushing personal pension information in an embodiment of the present invention;

[0022] Figure 2 A flowchart of the training of a prediction model for opening an individual pension account in an embodiment of the present invention;

[0023] Figure 3 This is a flowchart of using a training data set to train a LightGBM model in an embodiment of the present invention;

[0024] Figure 4 Schematic diagram of a device for pushing personal pension information in an embodiment of the present invention;

[0025] Figure 5 It is a schematic diagram of a specific device for pushing personal pension information in an embodiment of the present invention. DETAILED DESCRIPTION

[0026] To make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings. Here, the exemplary embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0027] Figure 1 This is a flow chart of a method for pushing personal pension information in an embodiment of the present invention, the method comprising:

[0028] Step 101, obtaining customer information, the customer information includes the customer's personal pension account opening information, identity information, and financial information; the customer's personal pension account opening information indicates whether the customer has opened a personal pension account;

[0029] Step 102, encrypt the customer information and encapsulate it into a transaction, and broadcast it to the blockchain network;

[0030] Step 103, after the nodes in the blockchain network have verified the transaction consensus, if the customer's personal pension account opening information indicates that the customer has not opened a personal pension account, the encrypted customer information is input into the personal pension account opening prediction model, and the personal pension account opening recommendation result is output; Step 104, if the personal pension account opening recommendation result is a recommendation to open a personal pension account, the personal pension information push rule data of the customer group to which the customer belongs is matched according to the encrypted customer information; Step 105, according to the matched personal pension information push rule data, the personal pension account opening recommendation information is pushed to the customer.

[0031] Each step is described in detail below.

[0032] In step 101, customer information is obtained, including the customer's personal pension account opening information, identity information, and financial information; the customer's personal pension account opening information indicates whether the customer has opened a personal pension account.

[0033] In a specific embodiment, customer information is obtained from the personal pension account system, and the corresponding customer's identity, finance, employment history, health, behavioral preferences and other data are matched from the following databases.

[0034] Personal pension account system: used to analyze whether the customer has opened a personal pension account in the bank.

[0035] Personal Customer Information System: Obtain the basic information of customers from the personal customer information system, including name, age, gender, educational background, marital status, etc. This information can be used to determine the personal background and characteristics of the customer.

[0036] Financial data system: Obtain information such as the client’s income level, asset status, investment preferences, etc. from the financial data system. This data helps assess the client’s financial capabilities and risk tolerance.

[0037] Employment history system: obtains the client’s employment history information, including occupation type, years of work experience, salary level, etc. This data can be used to determine the client’s career stage and future pension needs.

[0038] Health record system: Understand the customer's health status, chronic disease history, family medical history, etc. These data are important for assessing the customer's health status and future medical expenses.

[0039] Customer behavior preference database: By analyzing customer preferences and behavior data, such as past investment decisions, transaction records, etc., the customer's risk tolerance, investment preferences and goals can be revealed.

[0040] In step 102, the customer information is encrypted and packaged into a transaction, which is broadcast to the blockchain network.

[0041] In one embodiment, the customer information is encrypted and packaged into a transaction, including:

[0042] Use hash algorithm to encrypt customer information data and generate corresponding fingerprint data;

[0043] The encrypted customer information and the corresponding fingerprint data are packaged into a transaction.

[0044] In a specific embodiment, a hash algorithm is used to encrypt the collected customer information data, and the customer information is hashed to obtain a unique data fingerprint. The hash function converts the customer information data into a fixed-length hash value, which is unique and irreversible. The hashed customer information data is encapsulated into a transaction and broadcast to the blockchain network. The nodes in the blockchain network will verify the transaction to ensure the legitimacy and consistency of the transaction. This is achieved through a consensus mechanism. Once a transaction passes the consensus verification, the transaction is packaged into a block and added to the end of the blockchain. Each block contains the hash value of the previous block, forming a growing chain structure.

[0045] In step 103, after the nodes in the blockchain network pass the transaction consensus verification, if the customer's personal pension account opening information indicates that the customer has not opened a personal pension account, the encrypted customer information is input into the personal pension account opening prediction model, and the personal pension account opening recommendation result is output.

[0046] like Figure 2 As shown, in one embodiment, the personal pension account opening prediction model is obtained as follows:

[0047] Step 201, obtaining historical customer information and personal pension account opening information of historical customers;

[0048] Step 202, extracting features from historical customer information and personal pension account opening information of historical customers to obtain feature information, wherein the feature information includes income information, occupation information, and behavior preference information;

[0049] Step 203, constructing sample data based on income information, occupation information and behavior preference information, and generating a training data set and a test data set;

[0050] Step 204, train the LightGBM model using the training data set, and test the trained LightGBM model using the test data set;

[0051] Step 205, when the test passes, the trained LightGBM is determined as the individual pension account opening prediction model.

[0052] In a specific embodiment, useful features are extracted from historical customer information and personal pension account opening information of historical customers, such as income level, asset portfolio, investment preference, etc. The extracted features are then transformed and digitized or standardized to facilitate model training and comparison.

[0053] In one embodiment, before extracting features from historical customer information and personal pension account opening information of historical customers, the method further includes:

[0054] Perform data cleaning, data transformation and data specification processing on historical customer information and historical customers' personal pension account opening information.

[0055] like Figure 3 As shown, in one embodiment, the LightGBM model is trained using a training data set, including:

[0056] Step 301, input the feature information in the training data set into the LightGBM model, and output the predicted probability of opening a personal pension account corresponding to the feature information;

[0057] Step 302, determining the label value of the training data set based on the historical customer's personal pension account opening information;

[0058] Step 303, calculating a minimized log-likelihood loss function based on the predicted probability of opening a personal pension account and the label value of the training data set;

[0059] Step 304, training the LightGBM model according to minimizing the log-likelihood loss function.

[0060] In a specific embodiment, an initial decision tree model is constructed, and the data set is divided into two subsets. A gradient descent algorithm is used to update the prediction value of each tree node by minimizing the log-likelihood loss function.

[0061]

[0062] Among them, L(y,p) represents the loss function, y i is the test set label of the sample (0 or 1), p i is the probability that the model predicts the sample’s intention to open a pension account.

[0063] In one embodiment, a label value of 1 in the training data set represents that the customer intends to open a personal pension account, and a label value of 0 represents that the customer has no intention to open a personal pension account.

[0064] In step 104, if the personal pension account opening recommendation result is a recommendation to open a personal pension account, the personal pension information push rule data of the customer group to which the customer belongs is matched according to the encrypted customer information.

[0065] In one embodiment, if the result of the recommendation for opening a personal pension account is to recommend opening a personal pension account, according to the encrypted customer information, the personal pension information push rule data of the customer group to which the customer belongs is matched, including:

[0066] According to the encrypted customer information, the personal pension information push rule data of the customer group to which the customer belongs is matched in the Drools rule library; the personal pension information push rule data of different customer groups is pre-loaded in the Drools rule library.

[0067] In step 105, based on the matched personal pension information push rule data, personal pension account opening recommendation information is pushed to the customer.

[0068] In a specific embodiment, after a customer opens a personal pension account at a bank, he or she can purchase a variety of personal pension products offered by the bank. However, the personal pension market is diverse and complex, and traditional marketing methods are often based on large-scale advertising, which lacks efficiency, personalization, and accuracy.

[0069] In one embodiment, the method further includes: if the customer's personal pension account opening information indicates that the customer has opened a personal pension account, input the encrypted customer information into the personal pension product recommendation model, and output the personal pension product recommendation result; the personal pension product recommendation model is obtained by training the LightGBM model based on historical customer information and information on pension products purchased by historical customers; and personal pension product information is pushed to the customer based on the matched personal pension information push rule data and the personal pension product recommendation result. Thus, pension products are accurately pushed to suitable customers.

[0070] In one embodiment, the personal pension product recommendation model is obtained as follows:

[0071] Obtain historical customer information and information on pension products purchased by historical customers;

[0072] Extracting features from historical customer information and information on pension products purchased by historical customers to obtain feature information, wherein the feature information includes income information, occupation information, and behavior preference information;

[0073] Construct sample data based on income information, occupation information, and behavioral preference information to generate training data sets and test data sets;

[0074] Use the training data set to train the LightGBM model, and use the test data set to test the trained LightGBM model;

[0075] When the test passes, the trained LightGBM is determined as the personal pension product recommendation model.

[0076] In one embodiment, before extracting features from historical customer information and information on pension products purchased by historical customers, the method further includes:

[0077] Perform data cleaning, data transformation and data reduction on historical customer information and information on pension products purchased by historical customers.

[0078] In one embodiment, the LightGBM model is trained using a training data set, including:

[0079] Input the feature information in the training data set into the LightGBM model, and output the predicted probability of purchasing personal pension products corresponding to the feature information;

[0080] Determine the label value of the training data set based on the pension product information purchased by historical customers;

[0081] Calculate the minimization of the log-likelihood loss function based on the predicted probability of purchasing pension products and the label value of the training data set;

[0082] The LightGBM model is trained by minimizing the log-likelihood loss function.

[0083] The present invention also provides a personal pension information push device, as described in the following embodiments. Since the principle of solving the problem by the device is similar to that of the personal pension information push method, the implementation of the device can refer to the implementation of the personal pension information push method, and the repeated parts will not be repeated. Figure 4 As shown, the device comprises:

[0084] The customer information acquisition module 401 is used to acquire customer information, which includes the customer's personal pension account opening information, identity information, and financial information; the customer's personal pension account opening information indicates whether the customer has opened a personal pension account;

[0085] The customer information encryption module 402 is used to encrypt the customer information and encapsulate it into a transaction, and broadcast it to the blockchain network;

[0086] The individual pension account opening recommendation result acquisition module 403 is used to input the encrypted customer information into the individual pension account opening prediction model after the nodes in the blockchain network have verified the transaction consensus, if the individual pension account opening information of the customer indicates that the customer has not opened an individual pension account, and output the individual pension account opening recommendation result;

[0087] The personal pension information push rule data acquisition module 404 is used to match the personal pension information push rule data of the customer group to which the customer belongs according to the encrypted customer information if the personal pension account opening recommendation result is a recommendation to open a personal pension account. The personal pension account opening recommendation information push module 405 is used to push the personal pension account opening recommendation information to the customer according to the matched personal pension information push rule data.

[0088] like Figure 5 As shown, in one embodiment, it also includes a personal pension account opening prediction model training module 501, which is specifically used to:

[0089] The prediction model for opening individual pension accounts is obtained as follows:

[0090] Obtain historical customer information and personal pension account opening information of historical customers;

[0091] Extracting features from historical customer information and personal pension account opening information of historical customers to obtain feature information, wherein the feature information includes income information, occupation information, and behavior preference information;

[0092] Construct sample data based on income information, occupation information, and behavioral preference information to generate training data sets and test data sets;

[0093] Use the training data set to train the LightGBM model, and use the test data set to test the trained LightGBM model;

[0094] When the test passes, the trained LightGBM is determined as the individual pension account opening prediction model.

[0095] In one embodiment, it further includes an account opening information data preprocessing module, which is specifically used to:

[0096] Perform data cleaning, data transformation and data specification processing on historical customer information and historical customers' personal pension account opening information.

[0097] In one embodiment, the individual pension account opening prediction model training module 501 is specifically used to:

[0098] Input the feature information in the training data set into the LightGBM model, and output the predicted probability of opening a personal pension account corresponding to the feature information;

[0099] Determine the label value of the training data set based on historical customer personal pension account opening information;

[0100] Calculate the minimization of the log-likelihood loss function based on the predicted probability of opening a personal pension account and the label value of the training data set;

[0101] The LightGBM model is trained by minimizing the log-likelihood loss function.

[0102] In one embodiment, a label value of 1 in the training data set represents that the customer intends to open a personal pension account, and a label value of 0 represents that the customer has no intention to open a personal pension account.

[0103] In one embodiment, the customer information encryption module 402 is specifically used to:

[0104] Use hash algorithm to encrypt customer information data and generate corresponding fingerprint data;

[0105] The encrypted customer information and the corresponding fingerprint data are packaged into a transaction.

[0106] In one embodiment, the personal pension information push rule data acquisition module 404 is specifically used to:

[0107] According to the encrypted customer information, the personal pension information push rule data of the customer group to which the customer belongs is matched in the Drools rule library; the personal pension information push rule data of different customer groups is pre-loaded in the Drools rule library.

[0108] In one embodiment, a personal pension product information push module is also included, which is specifically used to:

[0109] If the customer's personal pension account opening information indicates that the customer has opened a personal pension account, the encrypted customer information is input into the personal pension product recommendation model, and the personal pension product recommendation result is output; the personal pension product recommendation model is obtained by training the LightGBM model based on historical customer information and information on pension products purchased by historical customers;

[0110] Personal pension product information is pushed to customers based on the matching personal pension information push rule data and personal pension product recommendation results.

[0111] In one embodiment, the personal pension product information push module is specifically used to:

[0112] The personal pension product recommendation model is obtained as follows:

[0113] Obtain historical customer information and information on pension products purchased by historical customers;

[0114] Extracting features from historical customer information and information on pension products purchased by historical customers to obtain feature information, wherein the feature information includes income information, occupation information, and behavior preference information;

[0115] Construct sample data based on income information, occupation information, and behavioral preference information to generate training data sets and test data sets;

[0116] Use the training data set to train the LightGBM model, and use the test data set to test the trained LightGBM model;

[0117] When the test passes, the trained LightGBM is determined as the personal pension product recommendation model.

[0118] In one embodiment, a purchase product information data preprocessing module is further included, which is specifically used to:

[0119] Perform data cleaning, data transformation and data reduction on historical customer information and information on pension products purchased by historical customers.

[0120] In one embodiment, the personal pension product recommendation model training module includes:

[0121] Input the feature information in the training data set into the LightGBM model, and output the predicted probability of purchasing pension products corresponding to the feature information;

[0122] Determine the label value of the training data set based on the pension product information purchased by historical customers;

[0123] Calculate the minimization of the log-likelihood loss function based on the predicted probability of purchasing pension products and the label value of the training data set;

[0124] The LightGBM model is trained by minimizing the log-likelihood loss function.

[0125] An embodiment of the present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned personal pension information push method when executing the computer program.

[0126] An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned personal pension information pushing method is implemented.

[0127] An embodiment of the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the above-mentioned personal pension information push method is implemented.

[0128] In an embodiment of the present invention, by obtaining customer information, the customer information includes the customer's personal pension account opening information, identity information, and financial information; the customer's personal pension account opening information indicates whether the customer has opened a personal pension account; the customer information is encrypted and packaged into a transaction, and broadcast to the blockchain network; after the nodes in the blockchain network pass the transaction consensus verification, if the customer's personal pension account opening information indicates that the customer has not opened a personal pension account, the encrypted customer information is input into the personal pension account opening prediction model, and the personal pension account opening recommendation result is output; if the personal pension account opening recommendation result is a recommendation to open a personal pension account, according to the encrypted customer information, the personal pension information push rule data of the customer group to which the customer belongs is matched; according to the matched personal pension information push rule data, the personal pension account opening recommendation information is pushed to the customer. In the above process, the embodiment of the present invention mines potential customers who intend to open a personal pension account through the personal pension account opening prediction model, thereby reducing the amount of calculation, reducing the memory occupied space, and improving the efficiency and accuracy of personal pension information push.

[0129] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0130] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0131] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0132] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0133] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for pushing personal pension information, characterized in that: include: Obtaining customer information, including the customer's personal pension account opening information, identity information, and financial information; the customer's personal pension account opening information indicates whether the customer has opened a personal pension account; Encrypt customer information and encapsulate it into transactions, which are then broadcast to the blockchain network; After the nodes in the blockchain network have verified the transaction consensus, if the customer's personal pension account opening information indicates that the customer has not opened a personal pension account, the encrypted customer information is input into the personal pension account opening prediction model, and the personal pension account opening recommendation result is output; If the result of the recommendation for opening a personal pension account is to recommend opening a personal pension account, the personal pension information push rule data of the customer group to which the customer belongs is matched based on the encrypted customer information; Based on the matching personal pension information push rule data, personal pension account opening recommendation information is pushed to customers.

2. The method according to claim 1, characterized in that The prediction model for opening individual pension accounts is obtained as follows: Obtain historical customer information and personal pension account opening information of historical customers; Extracting features from historical customer information and personal pension account opening information of historical customers to obtain feature information, wherein the feature information includes income information, occupation information, and behavior preference information; Construct sample data based on income information, occupation information, and behavioral preference information to generate training data sets and test data sets; Use the training data set to train the LightGBM model, and use the test data set to test the trained LightGBM model; When the test passes, the trained LightGBM is determined as the individual pension account opening prediction model.

3. The method according to claim 1, characterized in that If the result of the recommendation for opening a personal pension account is to recommend opening a personal pension account, the personal pension information push rule data of the customer group to which the customer belongs is matched according to the encrypted customer information, including: According to the encrypted customer information, the personal pension information push rule data of the customer group to which the customer belongs is matched in the Drools rule library; the personal pension information push rule data of different customer groups is pre-loaded in the Drools rule library.

4. The method according to claim 2, characterized in that Use the training data set to train the LightGBM model, including: Input the feature information in the training data set into the LightGBM model, and output the predicted probability of opening a personal pension account corresponding to the feature information; Determine the label value of the training data set based on historical customer personal pension account opening information; Calculate the minimization of the log-likelihood loss function based on the predicted probability of opening a personal pension account and the label value of the training data set; The LightGBM model is trained by minimizing the log-likelihood loss function.

5. The method according to claim 4, characterized in that The label value of the training data set is 1, which means that the customer intends to open a personal pension account, and the label value is 0, which means that the customer has no intention to open a personal pension account.

6. The method according to claim 1, characterized in that Encrypt customer information and package it into transactions, including: Use hash algorithm to encrypt customer information data and generate corresponding fingerprint data; The encrypted customer information and the corresponding fingerprint data are packaged into a transaction.

7. The method according to any one of claims 1 to 6, characterized in that: Also includes: If the customer's personal pension account opening information indicates that the customer has opened a personal pension account, the encrypted customer information is input into the personal pension product recommendation model, and the personal pension product recommendation result is output; the personal pension product recommendation model is obtained by training the LightGBM model based on historical customer information and information on pension products purchased by historical customers; Personal pension product information is pushed to customers based on the matching personal pension information push rule data and personal pension product recommendation results.

8. The method according to claim 7, characterized in that The personal pension product recommendation model is obtained as follows: Obtain historical customer information and information on pension products purchased by historical customers; Extracting features from historical customer information and information on pension products purchased by historical customers to obtain feature information, wherein the feature information includes income information, occupation information, and behavior preference information; Construct sample data based on income information, occupation information, and behavioral preference information to generate training data sets and test data sets; Use the training data set to train the LightGBM model, and use the test data set to test the trained LightGBM model; When the test passes, the trained LightGBM is determined as the personal pension product recommendation model.

9. A personal pension information push device, characterized in that: include: The customer information acquisition module is used to acquire customer information, including the customer's personal pension account opening information, identity information, and financial information; the customer's personal pension account opening information indicates whether the customer has opened a personal pension account; The customer information encryption module is used to encrypt customer information and encapsulate it into transactions, which are then broadcast to the blockchain network; A module for obtaining the recommendation result of opening a personal pension account, which is used to input the encrypted customer information into the prediction model for opening a personal pension account and output the recommendation result of opening a personal pension account after the nodes in the blockchain network have verified the transaction consensus and if the customer's personal pension account opening information indicates that the customer has not opened a personal pension account; The personal pension information push rule data acquisition module is used to match the personal pension information push rule data of the customer group to which the customer belongs based on the encrypted customer information if the personal pension account opening recommendation result is a recommendation to open a personal pension account; the personal pension account opening recommendation information push module is used to push personal pension account opening recommendation information to the customer based on the matched personal pension information push rule data.

10. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 8 is implemented.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images