Blockchain-based methods, systems, and media for managing digital financial customer records.

Abstract

This invention provides a blockchain-based method, system, and medium for managing digital financial customer files, belonging to the field of file management. It solves the problem of low efficiency in financial customer file management. Specifically, it involves: acquiring customer identity and financial information; designing identity identifiers based on the identity information; constructing a Merkle tree to store financial information and deploying it on the blockchain; classifying customers into high-net-worth and ordinary customers; calculating credit scores for high-net-worth customers and determining loan interest rates for them; analyzing the transaction time, transaction partners, and transaction amounts of ordinary customers to determine their frequently visited shops, further classifying them, and obtaining customer consumption behavior reports; summarizing customer consumption behavior reports and providing feedback; and synchronously updating financial information on the blockchain. This invention improves the security of financial information and enables financial institutions to understand changes in customer consumption in real time by digitizing and storing customer financial information.

Description

Technical Field

[0001] This invention relates to a blockchain-based method, system, and medium for managing digital records of financial clients, and pertains to the field of record management. Background Technology

[0002] Existing methods for managing digital profiles of financial clients have the following shortcomings:

[0003] Storage efficiency issues: Existing management methods result in high redundancy of blockchain operational data. For example, with Hyperledger Fabric, each node needs to store complete blockchain data under its default configuration. When a large bank piloted blockchain file management, the storage requirement of a single node reached the petabyte level, increasing hardware costs.

[0004] Lack of blockchain protection: Existing management methods are limited in function, only recording financial information. In the process of recording high-frequency transactions, the speed of change of financial information is often greater than the data confirmation speed of the blockchain. For example, when a stock exchange tested blockchain stock transaction file management, it found that the confirmation time for a single transaction reached the second level, while the traditional centralized system was only at the millisecond level. This situation not only leads to a waste of blockchain computing resources, but also results in data errors.

[0005] Talent shortage: Blockchain development requires proficiency in languages ​​such as Go and Solidity, making it difficult for traditional financial institution IT teams to quickly adapt; at the same time, blockchain nodes need to run continuously and remain synchronized, resulting in high maintenance costs. Summary of the Invention

[0006] In view of the shortcomings of existing technologies, the purpose of this invention is to provide a blockchain-based method, system and medium for managing digital financial customer files, aiming to solve the problem of low efficiency in financial customer file management.

[0007] To achieve the above objectives, the present invention provides a blockchain-based method for managing digital financial customer records, comprising:

[0008] Obtain customer credit information, eliminate unqualified customers with poor credit, and obtain high-quality customers; obtain the identity and financial information of high-quality customers;

[0009] Design customer identity identifiers based on the identity information of high-quality customers; construct a Merkle tree based on the number of high-quality customers, and create three storage spaces in each node of the Merkle tree to store the identity identifier and encrypted financial information of the high-quality customer corresponding to this node, as well as the identity identifier of the high-quality customer corresponding to the adjacent node or the root node; deploy the Merkle tree on the blockchain to protect and preserve the customer's financial information.

[0010] Based on the account balance in the financial information, customers are divided into high-net-worth customers and ordinary customers; the credit score of high-net-worth customers is calculated, and the loan interest rate for high-net-worth customers is proposed based on the credit score; the transaction time and transaction objects of ordinary customers are analyzed to determine the shops that each ordinary customer frequents; and then, combined with the transaction amount, the number of target customers, potential customers and general customers of each shop that is frequently visited is counted to obtain a customer consumption behavior report.

[0011] Summarize customer consumption behavior reports and provide feedback to users; collect updated financial information from high-performing customers and synchronize it on the blockchain.

[0012] Furthermore, the specific steps for constructing the Merkle tree are as follows:

[0013] Get the number of customers 'na'; obtain the identity information of all customers and generate verifiable credentials to obtain VC. （1） ~VC （na） For VC （1） ~VC （na） Encryption is performed, and then the encrypted verifiable credential is converted into a hash value to obtain the identity identifier hVC. （1） ~hVC （na） ;

[0014] Calculate the depth bh of the Merkle tree: ;

[0015] Construct a complete binary tree of depth bh and number of nodes na, and call it tree A. （1） ;

[0016] In tree A （1） For each node's storage location, three storage spaces are created, denoted as space a, space b, and space c;

[0017] Tree A （1） Starting from the root node, hVC is stored sequentially in the space 'a' corresponding to node 1 to node 'na', from left to right and from top to bottom. （1） ~hVC （na） Tree A is obtained （2） ;

[0018] Define the storage conditions for space b; based on the storage conditions of space b, use tree A... （2） Based on this, hVC （1） ~hVC （na） The data is then stored twice in the space b corresponding to nodes 1 to na, resulting in tree A. （3） ;

[0019] Obtain and encrypt the financial information of customers 1 to na, and obtain the ciphertext emi. （1） ~emi （na） ;

[0020] Tree A （3） Based on this, the ciphertext emi （1） ~emi （na） The Merkle tree is obtained by sequentially storing the data into space c corresponding to nodes 1 to na; the blockchain is deployed in space D, and the Merkle tree is synchronously deployed to the blockchain.

[0021] Furthermore, the storage conditions of space b are as follows:

[0022] Let node ki be a tree A （2） Any node in;

[0023] If ki is odd, then node ki is the right node;

[0024] The neighboring nodes of node ki are nodes (ki-1), and the root node corresponding to node ki is node gki. （1） :gki （1） = [(ki-1) / 2];

[0025] Store the hash value hVC of the (ki-1)th customer in space b of node ki. （ki－1） ;

[0026] Store the hash value hVC of the ki-th customer in space b of node (ki-1). （ki） ;

[0027] Node GKI （1） Space b is divided into space b （1） and space b （2） ;

[0028] At node gki （1） space b （2） In, store in hVC （ki－1） and hVC （ki） ;

[0029] Determine GKI （1） The parity of the space b determines the space. （1） The storage object;

[0030] If gki （1） If it is odd, then in space b （1） Store the (gki) （1） -1) The hVC value of the customer （gki（1）－1） ;

[0031] If gki （1） If it is even, then in space b （1） Store the (gki) （1） +1) Customer's hVC value （gki（1）＋1）;

[0032] If ki is even, then reset the storage conditions of space b.

[0033] Furthermore, the storage conditions for the reset space b are as follows:

[0034] If ki is even, then node ki is the left node;

[0035] The neighboring node of node ki is node (ki+1), and the root node corresponding to node ki is node gki. （2） :gki （2） =ki / 2;

[0036] Store the HVC value of the (ki+1)th customer in space b of node ki. （ki＋1） ;

[0037] Store the HVC value corresponding to the ki-th customer in space b at node (ki+1). （ki） ;

[0038] Node GKI （2） Space b is divided into space b （3） and space b （4） ;

[0039] At node gki （2） space b （4） In, store in hVC （ki＋1） and hVC （ki） ;

[0040] Determine GKI （2） The parity of the space b determines the space. （3） The storage object;

[0041] If gki （2） If it is odd, then in space b （3） Store the (gki) （2） -1) The hVC value of the customer （gki（2）－1） ;

[0042] If gki （2） If it is even, then in space b （3） Store the (gki) （2） +1) Customer's hVC value （gki（2）＋1） .

[0043] Furthermore, the specific steps for deploying cloud servers and encrypting financial information in space D are as follows:

[0044] Create storage space D on the cloud server; in space D, create a blockchain network through the cloud platform console, select the consensus algorithm, number of nodes and data storage method; set the node IP whitelist and firewall rules, write business logic using the Ethereum compatible chain, and complete local testing through Remix or Truffle; upload the contract to the cloud platform blockchain service, call the contract methods through API / SDK to complete the deployment and setup of the blockchain;

[0045] According to hVC （1） ~hVC （na） Use secure communication protocols such as TLS / SSL to obtain the security parameters of the first to the nath client. （1） ~rc （na） ;

[0046] rc （1） ~rc （na） The public key PK is obtained as input to the Setup algorithm. （1） ~PK （na） The master key is denoted as MSK. （1） ~MSK （na） ;

[0047] Public key, master key, and ne （1） Using the KeyGen algorithm, based on PK （1） ~PK （na） and MSK （1） ~MSK （na） Generate the key SK for the first to the nath clients. （1） ~SK （na） , ID number id （1） ~id （na） ;

[0048] Use the RSA algorithm to generate information mi. （1） ~mi （na） The symmetric key is used to obtain SYK. （1） ~SYK （na） ;

[0049] Using the Encrypt algorithm as the encryption algorithm, and employing PK （1） ~PK （na） SK （1） ~SK （na） and SYK （1） ~SYK （na） information mi （1） ~mi （na） Encryption is performed to obtain the ciphertext emi （1） ~emi （na） .

[0050] Further, the specific steps for classifying customers into high-net-worth customers and ordinary customers are as follows:

[0051] Record the account amounts of the customers from the 1st to the na-th as mon （1） ~mon （na） ; Take the maximum value among mon （1） ~mon （na） as mon （max） ;

[0052] Define relation S-1: ;[[ID=第十九]]

[0053] where mon （k） represents the account amount of the k-th customer;

[0054] Take the customers who satisfy relation S-1 as high-net-worth customers, and those who do not satisfy relation S-1 as ordinary customers;

[0055] Count the number nca of high-net-worth customers and the number ncb of ordinary customers;

[0056] Calculate the credit scores cre （1） ~cre （mca） of high-net-worth customers, and formulate the loan interest rates loa （1） ~loa （nca） ;

[0057] Analyze the transaction time and transaction objects of ordinary customers to determine the regular patronage stores of each ordinary customer; then classify ordinary customers in combination with the transaction amount;

[0058] Count the number of target customers, potential customers, and general customers in each regular patronage store.

[0059] Further, the steps for formulating loa （1） are as follows:

[0060] Take the market type to which the 1st high-net-worth customer belongs as the prospective market;

[0061] Take other high-net-worth customers who belong to the prospective market as prospective customers, and count the number ncl of prospective users;

[0062] Obtain the account amounts moy （1） ~moy （ncl） of all prospective customers; calculate the sum sum of mon （1） and moy （1） ~moy （ncl） ;

[0063] Among pr （1） ~pr （mca） and upr（1） ~upr （mca） In the middle, extract the expected profitability po of the first to nth potential users. （1） ~po （ncl） Calculate the quasi-market profitability rate Apr; extract the expected negative interest rate upo. （1） ~upo （ncl） Calculate the quasi-market profitability rate Aupr;

[0064] Calculate the net profit margin AUp in the quasi-market: AUp = Apr - Aupr;

[0065] Calculate the near-market portfolio return REm: Among them, moy （l） This represents the account balance of the l-th prospective customer;

[0066] Calculate the expected profit Qpr of the top high-net-worth client, and calculate the expected profit Qpo of the top ncl prospective client. （1） ~Qpo （ncl） ;

[0067] Qpr, Qpo （1） ~Qpo （ncl） Normalizing to the interval between 0 and 1 yields rpr and rpo. （1） ~rpo （ncl） ;

[0068] Calculate rpr and rpo （1） ~rpo （ncl） The variance SSpr is calculated, and REm is then compared with rpr and rpo in turn. （1） ~rpo （ncl） The covariance SXpr;

[0069] Calculate the beta coefficient β of the quasi-market, β = SXpr / SSpr;

[0070] Calculate the risk premium coefficient δ for the first high-net-worth client, δ = β × (REm - rpr).

[0071] cre （1） Normalizing to the interval between 0 and 1, we get rcre;

[0072] Assuming the bank's initial lending rate is ROA, propose a Loa. （1） :

[0073] loa （1） =roa + (rcre × δ).

[0074] Furthermore, the specific steps for classifying ordinary customers are as follows:

[0075] Define consumption tags: high frequency, low frequency, large amount, small amount;

[0076] Identify the stores frequently visited by the first regular customer and categorize the first regular customer.

[0077] Based on the transaction partner, calculate the total number of shops visited by the first ordinary customer in the past month (shp);

[0078] Count the number of times the first regular customer visited the first to the second shop on day 1: fr （1，1） ~fr （1，shp） Similarly, the number of times the customer visits shops 1 through shp on day dat: fr （dat，1） ~fr （dat，shp） ;

[0079] Constructed matrix B （1） And calculate the covariance matrix C:

[0080] Partitioning matrix C by columns yields matrix Cc. （1） ~Matrix Cc （shp） ;

[0081] Define matrix Cc （1） ~Matrix Cc （shp） weight λ （1） ~λ （shp） and bias μ （1） ~μ （shp） Construct relation S-2: ;

[0082] Calculate λ （1） ~λ （shp） and μ （1） ~μ （shp） The value;

[0083] Calculate the frequency coefficient ba of the first ordinary customer with respect to the first shop. （1） :ba （1） =λ （1） / μ （1） ;

[0084] Similarly, the frequency coefficient ba of the shp-th shop （shp） :ba （shp） =λ （shp） / μ （shp） ;

[0085] Compare ba （1） ~ba （shp） Based on the size, select the shop with the highest frequency coefficient as the shop that the number 1 ordinary customer frequents;

[0086] If the first ordinary customer's consumption tags are: high frequency and large amount, then the first ordinary customer is the target customer of the store that the customer frequents.

[0087] If the consumption tags are: high frequency and small amount or low frequency and large amount, then the first ordinary customer is a potential customer of the store that you frequently visit.

[0088] If the consumption tags are: low frequency and small amount, then the first ordinary customer is a regular customer who frequents the store.

[0089] A blockchain-based digital customer profile management system includes:

[0090] Data acquisition module: used to acquire customer credit information, eliminate unqualified customers with poor credit, and obtain high-quality customers; acquire the identity and financial information of high-quality customers;

[0091] Data storage module: Used to design customer identity identifiers based on the identity information of premium customers; construct a Merkle tree based on the number of premium customers, and create three storage spaces in each node of the Merkle tree to store the identity identifier and encrypted financial information of the premium customer corresponding to this node, as well as the identity identifier of the premium customer corresponding to the adjacent node or the root node; deploy the Merkle tree on the blockchain to protect and preserve the customer's financial information;

[0092] Customer Analysis Module: This module categorizes customers into high-net-worth clients and ordinary clients based on account balances in financial information; calculates credit scores for high-net-worth clients and proposes loan interest rates based on these scores; analyzes transaction times and transaction partners of ordinary clients to determine their frequently visited shops; and, combined with transaction amounts, calculates the number of target customers, potential customers, and general customers for each frequently visited shop to generate a customer consumption behavior report.

[0093] Data update module: used to summarize customer consumption behavior reports and provide feedback to users; collect updated financial information of high-quality customers and update it synchronously on the blockchain.

[0094] A computer-readable storage medium storing a computer program that, when executed by a processor, performs any of the steps in a blockchain-based financial customer digital profile management method.

[0095] Compared with the prior art, the beneficial effects of the present invention are:

[0096] Ensuring data security: The blockchain constructed in this invention uses asymmetric encryption and hash algorithms to encrypt and store customers' financial information; it generates a unique identifier for customers' identity information and puts it on the chain, stores the original data on the cloud platform, and then associates it with transaction records through a Merkle tree structure to synchronously update customers' financial information; this design not only prevents data leakage, but also ensures the security and integrity of customers' financial information through hash verification.

[0097] Preventing data tampering: Each block in this invention contains the hash value of the preceding block, forming an irreversible chain structure. Once a customer's transaction information is recorded in financial information and uploaded to the blockchain, any modification requires recalculating the hash values ​​of all subsequent blocks, resulting in an exponential increase in computational complexity and significantly increasing the cost of tampering. At the same time, the blockchain constructed by this invention also supports hot deployment, further reducing usage pressure.

[0098] Understanding Customer Consumption: After storing customers' financial information, this invention categorizes customers into high-net-worth clients and ordinary clients based on their actual financial situation. It sets loan interest rates for high-net-worth clients and recommends consumption types for ordinary clients, helping financial institutions better understand changes in customer consumption behavior and providing a reference for subsequent financial decisions. Attached Figure Description

[0099] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0100] Figure 1 This is a schematic diagram of the method of the present invention;

[0101] Figure 2 This is a schematic diagram of the system of the present invention;

[0102] Figure 3 This is a schematic diagram of a complete binary tree according to the present invention. Detailed Implementation

[0103] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. Example 1

[0104] Please see Figure 1 Blockchain-based methods for managing digital financial customer records include:

[0105] Step S1: Obtain customer credit information, eliminate unqualified customers with poor credit, and obtain high-quality customers; obtain the identity and financial information of high-quality customers;

[0106] It should be noted that, in this invention, "customer" refers to a customer who uses this invention (a blockchain-based method, system, and medium for managing digital records of financial customers) to manage digital records.

[0107] In this invention, only the "customer" in "step S1 'obtain customer credit information'" refers to all customers, namely, high-quality customers and low-quality customers; the "customer" appearing in subsequent steps S2 to S3 all refer to "high-quality customers". This is hereby explained.

[0108] It should be noted that the term "customer" in claims 2 to 8 refers to "premium customer";

[0109] Step S2: Design customer identity identifiers based on the identity information of premium customers; construct a Merkle tree based on the number of premium customers, and create three storage spaces in each node of the Merkle tree to store the identity identifier and encrypted financial information of the premium customer corresponding to this node, as well as the identity identifier of the premium customer corresponding to the adjacent node or the root node; deploy the Merkle tree on the blockchain to protect and preserve the customer's financial information.

[0110] The specific steps of step S2 are as follows:

[0111] Step S21: Obtain the number of customers na;

[0112] Obtain all customer identity information, use Hyperledger Fabric's identity management components (such as Identity Mixer) to generate verifiable credentials for customers, and obtain VC. （1） ~VC （na） ;

[0113] VC （1） ~VC （na） Stored in the corresponding customer's digital wallet;

[0114] Use AES or RSA algorithms to analyze VC. （1） ~VC （na） Encryption is performed, and then the encrypted verifiable credential is converted into a hash value to obtain the identity identifier hVC. （1） ~hVC （na） ;

[0115] Step S22: Please refer to Figure 3 Construct a Merkle tree to store customer financial information; (It should be noted that in this invention...) Figure 3 (For example demonstration only)

[0116] Calculate the depth bh of the Merkle tree: ;in, If it is a positive integer, round down.

[0117] Use the `build_binary_tree` function to construct a complete binary tree of depth `bh` and number of nodes `na`, which will be tree A. （1） ;

[0118] In tree A （1） For each node's storage location, three storage spaces are created, denoted as space a, space b, and space c;

[0119] Tree A （1） Starting from the root node, hVC is stored sequentially in the space 'a' corresponding to node 1 to node 'na', from left to right and from top to bottom. （1） ~hVC （na） Tree A is obtained （2） ;

[0120] Define the storage conditions for space b;

[0121] Let node ki be a tree A （2） Any node in the array; based on the parity of ki, determine the adjacent nodes and the root node corresponding to node ki, and fill the space b in the adjacent nodes and the root node (corresponding to node ki) with data;

[0122] If ki is odd, it means that node ki is the right node;

[0123] The neighboring nodes of node ki are nodes (ki-1), and the root node corresponding to node ki is node gki. （1） ;gki （1） The formula for calculation is: gki （1） = [(ki-1) / 2];

[0124] Store the HVC value (after conversion of the verifiable credential) corresponding to the (ki-1)th customer in space b of node ki. （ki－1） ;

[0125] Store the HVC value (after the transformation of the verifiable credential) corresponding to the ki-th customer in space b of node (ki-1). （ki） ;

[0126] Node GKI （1） Space b is divided into two storage spaces, resulting in space b （1） and space b （2） ;

[0127] At node gki （1） space b （2） In, store in hVC （ki－1） and hVC （ki） ;

[0128] Determine GKI （1） The parity of the space b determines the space. （1） The storage object;

[0129] If gki （1） If it is odd, then in space b （1） Store the (gki) （1） -1) The hVC value (after verification of the voucher) corresponding to each customer. （gki（1）－1） ;

[0130] If gki （1） If it is even, then in space b （1） Store the (gki) （1） +1) The hVC value (after verification of the voucher) corresponding to the customer. （gki（1）＋1） ;

[0131] If ki is even, then node ki is the left node;

[0132] The neighboring node of node ki is node (ki+1), and the root node corresponding to node ki is node gki. （2） ;gki （2） The formula for calculation is: gki （2） =ki / 2;

[0133] Store the HVC value (after the transformation of the verifiable credential) corresponding to the (ki+1)th customer in space b of node ki. （ki＋1） ;

[0134] Store the HVC value (after the transformation of the verifiable voucher) corresponding to the ki-th customer in space b at node (ki+1). （ki） ;

[0135] Node GKI （2） Space b is divided into two storage spaces, resulting in space b （3） and space b （4） ;

[0136] At node gki （2） space b （4） In, store in hVC （ki＋1） and hVC （ki） ;

[0137] Determine GKI （2） The parity of the space b determines the space. （3） The storage object;

[0138] If gki （2） If it is odd, then in space b （3） Store the (gki) （2）-1) The hVC value (after verification of the voucher) corresponding to each customer. （gki（2）－1） ;

[0139] If gki （2） If it is even, then in space b （3） Store the (gki) （2） +1) The hVC value (after verification of the voucher) corresponding to the customer. （gki（2）＋1） ;

[0140] Based on the storage conditions of space b, using tree A （2） Based on this, hVC （1） ~hVC （na） The data is then stored twice in the space b corresponding to nodes 1 to na, resulting in tree A. （3） ;

[0141] Step S23: Obtain and encrypt the financial information of the first to the nath customers;

[0142] Create a storage space on the cloud server, designated as space D. Within space D, create a blockchain network through the cloud platform console, selecting a consensus algorithm (e.g., PBFT adapted for consortium blockchains), the number of nodes (3-5 nodes initially recommended), and the data storage method (e.g., cloud object storage + local caching). Set up a whitelist of node IPs and firewall rules, write business logic (e.g., product traceability rules) using Solidity (Ethereum compatible chain) or Go (Hyperledger Fabric), and complete local testing using Remix or Truffle. Upload the contract to the cloud platform blockchain service, and call contract methods via API / SDK to complete the deployment and setup of the blockchain.

[0143] Return the blockchain access interface to the first through the nath clients;

[0144] Obtain the financial information of customers 1 to na, and obtain information mi. （1） ~mi （na） ;

[0145] Financial information includes: account balance, transaction time, transaction counterparty, and transaction amount;

[0146] information mi （1） ~mi （na） Encrypt and define the decryption process;

[0147] Encryption process:

[0148] According to hVC （1） ~hVC （na） Use secure communication protocols such as TLS / SSL to obtain the security parameters of the first to the nath client. （1）~rc （na） ;

[0149] rc （1） ~rc （na） The public key PK is obtained as input to the Setup algorithm. （1） ~PK （na） The master key is denoted as MSK. （1） ~MSK （na） ;

[0150] Public key, master key, and ne （1） (i.e., the IP address of the first user) uses the KeyGen algorithm, based on PK. （1） ~PK （na） and MSK （1） ~MSK （na） Generate the key SK for the first to the nath clients. （1） ~SK （na） , ID number id （1） ~id （na） ;

[0151] Use the RSA algorithm to generate information mi. （1） ~mi （na） The symmetric key is used to obtain SYK. （1） ~SYK （na） ;

[0152] Using the Encrypt algorithm as the encryption algorithm, and employing PK （1） ~PK （na） SK （1） ~SK （na） and SYK （1） ~SYK （na） information mi （1） ~mi （na） Encryption is performed to obtain the ciphertext emi （1） ~emi （na） ;

[0153] Tree A （3） Based on this, the ciphertext emi （1） ~emi （na） The Merkle tree is sequentially stored in space c corresponding to nodes 1 to na to obtain the Merkle tree; and the Merkle tree is synchronously deployed to the blockchain in space D.

[0154] Define the decryption process:

[0155] The public key is denoted as PK, the master key as MSK, and the private key as SK;

[0156] Let the public key entered by the visitor be PK`, the private key be SK`, the master key be MSK`, and the ID number be id`;

[0157] Using PK`, MSK`, and SK` as parameters for the Keycheck algorithm, the algorithm checks the format of PK`, MSK`, and SK` against that of PK`. （1） ~PK （na） MSK （1） ~MSK （na） and SK （1） ~SK （na） Are the formats consistent?

[0158] If there is a discrepancy, access will be terminated and an alarm will be triggered immediately.

[0159] If they match, PK`, MSK`, and SK` are used as inputs to the Decrypt algorithm to obtain the symmetric key SYK`; the Decrypt algorithm is then used to output the client IP address iid corresponding to MSK` and SK`.

[0160] Determine if iid and id` are the same;

[0161] If they match, use PK`, SK` and SYK` to decrypt the ciphertext corresponding to iid;

[0162] If there is a discrepancy, access will be terminated and an alarm will be triggered immediately.

[0163] It should be noted that, in this invention, "visitor" refers to a person who accesses "financial information in space D";

[0164] Step S3: Based on the account balance in the financial information, customers are divided into high-net-worth customers and ordinary customers; calculate the credit score of high-net-worth customers and propose the loan interest rate for high-net-worth customers based on the credit score; analyze the transaction time and transaction objects of ordinary customers to determine the frequently visited shops of each ordinary customer; and then, in combination with the transaction amount, count the number of target customers, potential customers and general customers of each frequently visited shop to obtain a customer consumption behavior report.

[0165] The specific steps of step S3 are as follows:

[0166] Step S31: Based on the account balance in the financial information, customers are divided into high-net-worth customers and ordinary customers;

[0167] The account balances of customers 1 through na are denoted as mon. （1） ~mon （na） ; will mon （1） ~mon （na） The maximum value in mon （max） ;

[0168] Define relation S-1: ;

[0169] Among them, mon （k） This represents the account balance of the k-th customer, where k ranges from 1 to na.

[0170] Customers who meet the relationship formula S-1 are classified as high-net-worth customers, while customers who do not meet the relationship formula S-1 are classified as ordinary customers.

[0171] Statistics on the number of high-net-worth clients (nca) and the number of ordinary clients (ncb);

[0172] Step S32: Calculate the credit scores (cre) of the first to nth high-net-worth clients using the Z-Score model. （1） ~cre （mca） The mathematical formula for the Score model is:

[0173] Customer's credit score = (1.2 × ZX) （1） ) + (1.4 × ZX （2） ) + (3.3 × ZX （3） ) + (0.6 × ZX （4） ) + (0.99 × ZX （5） );

[0174] Among them, ZX （1） ZX represents an asset liquidity indicator for high-net-worth clients. （1） = (Current Assets - Current Liabilities) / Total Assets;

[0175] ZX （2） ZX represents the cumulative profitability of high-net-worth clients. （2） = Retained earnings / Total assets;

[0176] ZX （3） ZX represents the operational efficiency of high-net-worth clients. （3） = Earnings Before Interest and Taxes / Total Assets;

[0177] ZX （4） This indicates the capital structure of high-net-worth clients, ZX （4） = Shareholders' equity / Total liabilities;

[0178] ZX （5） ZX represents the asset turnover rate of high-net-worth clients. （5） =Sales revenue / Total assets;

[0179] Obtain the year-end account balances of the top 1 to nca high-net-worth clients over the past three years, calculate the return and negative interest rate for each client over the past three years, and use a Markov chain to predict the expected return rate (pr) of the top 1 to nca high-net-worth clients. （1） ~pr （mca） Expected negative interest rates upr （1）~upr （mca） (If a high-net-worth client's account balance at the end of the previous year is greater than the account balance at the end of the following year, then the negative interest rate for that high-net-worth client is calculated. If the negative interest rate is positive, then the profitability rate of that high-net-worth client is negative.)

[0180] If a high-net-worth client's account balance at the end of the previous year is less than the account balance at the end of the following year, then the high-net-worth client's profitability rate is calculated. A positive profitability rate indicates a negative profitability rate, while a negative profitability rate indicates a negative profitability rate.

[0181] If a high-net-worth client's account balance at the end of the previous year equals their account balance at the end of the following year, both the profitability rate and the negative interest rate will be 0.

[0182] To determine the market type to which high-net-worth clients belong, and to formulate loan interest rates for the top 1 to nca high-net-worth clients;

[0183] Proposed loan interest rate (LOA) for the first high-net-worth client （1） ;

[0184] The market type to which the top high-net-worth client belongs is considered the quasi-market.

[0185] The number of potential users (ncl) is calculated by considering all high-net-worth clients belonging to the target market, excluding the first high-net-worth client.

[0186] Get all potential clients' account balances (moy) （1） ~moy （ncl） ; Calculate mon （1） With moy （1） ~moy （ncl） The sum of (where moy) （1） ~moy （ncl） ∈{mon （1） ~mon （na）});

[0187] In pr （1） ~pr （mca） and upr （1） ~upr （mca） In the middle, extract the expected profitability po of the first to nth potential users. （1） ~po （ncl） And expected negative interest rate upo （1） ~upo （ncl） ;

[0188] Extract pr （1） ,po （1） ~po （ncl） The maximum value po in （max） Minimum value po （min）;

[0189] Calculate the quasi-market profitability rate (Apr): ;

[0190] Extracting UPR （1） 、upo （1） ~upo （ncl） The maximum value of upo in （max） Minimum value upo （min） ;

[0191] Calculate the quasi-market profitability rate Aupr: ;

[0192] Among them, po （l） and upo （l） , where represent the expected profit rate and expected negative interest rate of the l-th prospective user, respectively, and the value of l ranges from 1 to ncl;

[0193] Calculate the net profit margin AUp in the quasi-market: AUp = Apr - Aupr;

[0194] Calculate the near-market portfolio return REm: Among them, moy （l） This represents the account balance of the l-th prospective customer;

[0195] Calculate the expected profit Qpr for the top high-net-worth client: ;

[0196] Repeat the calculation process of Qpr, based on po （1） ~po （ncl） and upo （1） ~upo （ncl） and moy （1） ~moy （ncl） Calculate the expected profit of the first-valued, nth-fold prospect: Qpo （1） ~Qpo （ncl） ;

[0197] The logistic function is used to normalize Qpr to the interval between 0 and 1, resulting in rpr; the logistic function is used to normalize Qpo. （1） ~Qpo （ncl） Normalizing to the interval between 0 and 1, we get rpo. （1） ~rpo （ncl） ;

[0198] Calculate rpr and rpo （1） ~rpo （ncl） The variance SSpr is calculated, and REm is then compared with rpr and rpo in turn. （1） ~rpo （ncl） The covariance SXpr;

[0199] Calculate the beta coefficient β of the quasi-market, β = SXpr / SSpr;

[0200] Calculate the risk premium coefficient δ for the first high-net-worth client, δ = β × (REm - rpr).

[0201] Use the logistic function to calculate Cre （1） Normalizing to the interval between 0 and 1, we get rcre;

[0202] Assuming the bank's initial loan interest rate is ROA, the proposed loan interest rate (LOA) for the first high-net-worth client is... （1） :

[0203] loa （1） =roa + (rcre × δ);

[0204] Repeat Loa （1） The process of formulating the loan interest rate (LOA) for the second to nth highest net worth clients. （2） ~loa （nca） ;

[0205] It should be noted that when performing the relevant process in "step S32", the present invention needs to obtain the relevant authorization from the high-net-worth client; if the authorization from the high-net-worth client is not obtained, then step S32 is performed.

[0206] Step S33: Analyze the transaction time and transaction objects of ordinary customers to determine the stores that each ordinary customer frequents; then, based on the transaction amount, count the number of target customers, potential customers and general customers of each store to obtain a customer consumption behavior report;

[0207] Step S331: Define consumption tags: high frequency, low frequency, large amount, small amount;

[0208] The total number of transactions (i.e., total consumption) of the top 1 to top 20 ordinary users in the past month was calculated to obtain the cn （1） ~cn （ncb） ;

[0209] The maximum single transaction amount of the top 1 to top 10 ordinary users within the past month was calculated to obtain the mmy. （1） ~mmy （ncb） ;

[0210] Extract cn （1） ~cn （ncb） The maximum value cn （max） Minimum value cn （min） Calculate the weighted average of total transactions, acn: ;

[0211] Extract mmy （1） ~mmy （ncb） The maximum value of mmy （max） Minimum value mmy （min） Calculate the weighted average of the total number of transactions, amy: ;

[0212] Among them, cn （v） and mmy （v） , representing the total number of transactions and the maximum single transaction amount of the vth ordinary user in the past month, respectively, with v ranging from 1 to ncb;

[0213] High frequency: The total number of transactions in the past month is greater than that of a regular ACN client;

[0214] Low frequency: Regular customers whose total number of transactions in the past month is less than or equal to that of ACN;

[0215] Large transaction: The largest single transaction amount within the past month is greater than that of Amy's regular customers;

[0216] Small amount: Ordinary customers whose maximum single transaction amount in the past month is less than or equal to Amy's;

[0217] Identify the stores frequently visited by the first regular customer and categorize the first regular customer.

[0218] Based on the transaction partner, calculate the total number of shops visited by the first ordinary customer in the past month (shp);

[0219] Get the number of days in the past month (dat); count the number of times the first regular customer visited the 1st, 2nd, and so on, up to the shp store on the 1st day (within the past month): fr （1，1） ,fr （1，2） ~fr （1，shp） (It should be noted that this invention can count the number of times a regular customer visits a store multiple times; for example, if the first regular customer visited store A 5 times on the first day of the past month, that is, the first regular customer has 5 transaction records at store A on the first day of the past month, then the number of times the first regular customer visited store A on the first day of the past month is 5.)

[0220] Number of times the customer visited the 1st, 2nd, and so on, up to the shp shop on the 2nd day (within the past month): fr （2，1） ,fr （2，2） ~fr （2，shp） ;

[0221] And so on, counting the number of times the customer visited shop number 1, number 2, and so on, up to number shp on day dat (within the past month): fr （dat，1） ,fr （dat，2）~fr （dat，shp） ;

[0222] Construct a matrix B of (dat × shp) （1） : ;

[0223] Calculate matrix B （1） The decentralized matrix is ​​obtained as matrix B. （2） ;

[0224] Calculate matrix B （1） The covariance matrix is ​​obtained, resulting in matrix C: Where * denotes matrix multiplication, and T denotes the transpose of a matrix;

[0225] Partitioning matrix C by columns yields matrix Cc. （1） Matrix Cc （2） ~Matrix Cc （shp） ;

[0226] Define matrix Cc （1） Matrix Cc （2） ~Matrix Cc （shp） weight λ （1） , λ （2） ~λ （shp） and bias μ （1） μ （2） ~μ （shp） Construct relation S-2: ;

[0227] Based on relation S-2, λ is calculated using MATLAB software. （1） , λ （2） ~λ （shp） and μ （1） μ （2） ~μ （shp） The value;

[0228] Calculate the frequency coefficient ba of the first ordinary customer with respect to the first shop. （1） :ba （1） =λ （1） / μ （1） ;

[0229] The frequency coefficient of the shp-th shop ba （2） :ba （2） =λ （2） / μ （2） ;

[0230] And so on, the frequency coefficient ba of the th shop (shp) （shp） :ba （shp） =λ （shp） / μ （shp）;

[0231] Compare ba （1） ~ba （shp） Based on the size, select the shop with the highest frequency coefficient as the shop that the number 1 ordinary customer frequents;

[0232] If the consumption tags of the first ordinary customer are: high frequency and large amount, then the first ordinary customer is the target customer of the frequently visited store (corresponding to the first ordinary customer);

[0233] If the consumption tags of the first ordinary customer are: high frequency and small amount or low frequency and large amount, then the first ordinary customer is a potential customer of the store that the first ordinary customer corresponds to.

[0234] If the consumption tags of the first ordinary customer are: low frequency and small amount, then the first ordinary customer is (corresponding to the first ordinary customer) a general customer who frequently visits the store;

[0235] Repeat the same process of identifying the most frequented shops of the first regular customer and classifying the first regular customer, and then identify and classify the most frequented shops of the second to nb regular customers.

[0236] The system analyzes the shops frequently visited by the top 1 to top 20 ordinary customers (NCB), the number and information of target customers, potential customers and general customers of each shop, and summarizes the loan interest rates of the top 1 to top 20 high-net-worth customers to obtain a customer consumption behavior report.

[0237] It should be noted that "step S3" in this invention is executed by default at 24:00 every night, while at other times the functions of updating and saving financial information in steps S1, S2 and S4 are executed.

[0238] Because blockchain data confirmation takes a long time, the data upload function of blockchain may be limited when customers engage in high-frequency trading, especially securities trading, i.e., the rate of change of financial information is greater than the data confirmation speed of blockchain. When the blockchain constructed by this invention encounters the situation where "the rate of change of financial information is greater than the data confirmation speed of blockchain", the on-chain function of financial information in steps S1, S2 and S4 is temporarily stopped. Only the update time of financial information and the updated financial data are acquired and temporarily stored, and the financial information analysis function in step S3 is executed. When "the rate of change of financial information is less than or equal to the data confirmation speed of blockchain", the updated financial data is then uploaded to the blockchain to prevent the blockchain from malfunctioning.

[0239] Step S4: Compile customer consumption behavior reports and provide feedback to users; collect updated financial information of high-performing customers and synchronize it on the blockchain;

[0240] It should be noted that the "user" in step S4 of this invention refers to a financial institution user who uses this invention (a blockchain-based method, system, and medium for managing digital files of financial customers) to manage digital files of customers. Example 2

[0241] Please see Figure 2 A blockchain-based digital record management system for financial customers includes:

[0242] Data acquisition module: used to acquire customer credit information, eliminate unqualified customers with poor credit, and obtain high-quality customers; acquire the identity and financial information of high-quality customers;

[0243] Data storage module: Used to design customer identity identifiers based on the identity information of premium customers; construct a Merkle tree based on the number of premium customers, and create three storage spaces in each node of the Merkle tree to store the identity identifier and encrypted financial information of the premium customer corresponding to this node, as well as the identity identifier of the premium customer corresponding to the adjacent node or the root node; deploy the Merkle tree on the blockchain to protect and preserve the customer's financial information;

[0244] Customer Analysis Module: This module categorizes customers into high-net-worth clients and ordinary clients based on account balances in financial information; calculates credit scores for high-net-worth clients and proposes loan interest rates based on these scores; analyzes transaction times and transaction partners of ordinary clients to determine their frequently visited shops; and, combined with transaction amounts, calculates the number of target customers, potential customers, and general customers for each frequently visited shop to generate a customer consumption behavior report.

[0245] Data update module: used to summarize customer consumption behavior reports and provide feedback to users; collect updated financial information of high-performing customers and synchronize it on the blockchain; Example 3

[0246] A computer-readable storage medium stores a computer program thereon. When executed by a processor, the computer program performs the steps of a self-test method for intelligent shared charging piles as described in any of the above embodiments. Through the above technical solution, when the computer program is executed by a processor, it performs the method in any optional implementation of the above embodiments to achieve the following functions:

[0247] The system filters, collects, and processes financial information from high-net-worth clients, stores the encrypted financial information in a Merkle tree, and deploys it on a blockchain. It analyzes this information to categorize high-net-worth clients into high-net-worth and ordinary clients, analyzes their spending habits, sets loan interest rates for high-net-worth clients, and identifies the stores frequented by ordinary clients. This helps users understand the real-time changes and spending potential of high-net-worth clients.

[0248] The above formulas are all dimensionless calculations. The formulas are derived from software simulations using a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation. For example, there are weighting coefficients and proportional coefficients. The values ​​set are to quantify each parameter to obtain a specific value, which is convenient for subsequent comparison. The values ​​of the weighting coefficients and proportional coefficients are only required to not affect the proportional relationship between the parameters and the quantified values.

[0249] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for managing a digital archive of a financial customer based on a blockchain, characterized by, The archive management method comprises: Discard unqualified inferior customers to obtain high-quality customers; obtain the identity information and financial information of the high-quality customers, and the financial information comprises: account amount, transaction time, transaction object, and transaction amount; Design an identity label for the customers according to the identity information of the high-quality customers; construct a Merkle tree according to the number of the high-quality customers; and deploy the Merkle tree on a blockchain to protect and save the financial information of the customers; The specific steps of constructing the Merkle tree are as follows: Obtain the number of customers na; obtain the identity information of all customers and generate verifiable credentials, obtain VC （1） ~ VC （na） ; encrypt VC （1） ~ VC （na） , and then convert the encrypted verifiable credentials into a hash value to obtain an identity identifier hVC （1） ~ hVC （na） ; Calculate the depth bh of the Merkle tree: ； Construct a complete binary tree of depth bh and na nodes as tree A （1） ; At tree A （1） At each node, three storage spaces are created at the storage location of the node, denoted as space a, space b and space c; Tree A （1） From the root node, in the order from left to right and from top to bottom, hVC is stored in the space a corresponding to node 1 to node na in turn （1） ~ hVC （na） , and tree A is obtained （2） ; Define the storage condition of space b; according to the storage condition of space b, take tree A （2） as the basis, hVC （1） ~ hVC （na） Secondary storage node 1 to node na corresponding to space b, get tree A （3） ; obtaining and encrypting the financial information of the first to nth customers to obtain ciphertext emi （1） ~ emi （na） ; Based on tree A （3） The ciphertext emi （1） ~ emi （na） In turn, the space c corresponding to node 1 to node na is stored, and a Merkle tree is obtained; a storage space D is created in the cloud server, a blockchain is deployed in the storage space D, and the Merkle tree is synchronized and deployed to the blockchain; The storage condition of the space b is as follows: Let node ki be any node in tree A （2） ; If ki is an odd number, then the node ki is a right node; The adjacent node of the node ki is the node (ki-1), and the root node corresponding to the node ki is the node gki （1） : gki （1） = [(ki-1) / 2]; storing the hash value hVC of the (ki-1)th customer in the space b of the node ki （ki－1） ; storing the hash value hVC of the kth customer in the space b of the node (ki-1) （ki） ; The node gki （1） divides the space b into the space b （1） and the space b （2） ; At node gki （1） In the space b （2） of hVC （ki－1） and hVC （ki） ; determining the parity of gki （1） to determine the storage object of space b （1） ; If gki （1） is odd, then the hash value hVC （1） of the (gki （1） -1)th customer is stored in the space b （gki（1）－1） . If gki （1） is even, then the hash value hVC （1） of the (gki （1） +1)th customer is stored in the space b （gki（1）＋1） . If ki is an even number, then reset the storage condition of the space b; According to the account amount in the financial information, divide the customers into high net worth customers and ordinary customers; calculate the credit score of the high net worth customers, and determine the loan interest rate of the high net worth customers according to the credit score; analyze the transaction time and transaction object of the ordinary customers to determine the frequently-visited shops of each ordinary customer; then, in combination with the transaction amount, count the number of target customers, potential customers, and general customers of each frequently-visited shop to obtain a customer consumption behavior report; Summarize the customer consumption behavior report and feed back to the user; collect the updated financial information of the high-quality customers and synchronously update on the blockchain. 2.The blockchain-based financial customer digitalization archive management method of claim 1, wherein, The reset storage condition of the space b is as follows: If ki is an even number, then the node ki is a left node; The adjacent node of the node ki is the node (ki+1), and the root node corresponding to the node ki is the node gki （2） : gki （2） = ki / 2; storing the hash value hVC of the (ki+1)th customer in the space b of the node ki （ki＋1） ; storing the hash value hVC corresponding to the ki-th customer in the space b of the node (ki+1) （ki） ; The node gki （2） divides the space b into the space b （3） and the space b （4） ; At node gki （2） space b （4） In, store in hVC （ki＋1） and hVC （ki） ; determining the parity of gki （2） to determine the storage object of space b （3） ; If gki （2） is odd, then the hash value hVC （3） of the (gki （2） - 1)th customer is stored in space b （gki（2）－1） ; If gki （2） is even, then the hash value hVC （3） of the (gki （2） +1)th customer is stored in the space b （gki（2）＋1） . 3.The blockchain-based financial customer digitalization archive management method of claim 1, wherein, The specific steps of deploying the blockchain in the storage space D and obtaining and encrypting the financial information of the first to nth customers are as follows: In the storage space D, create a blockchain network through a cloud platform console, select a consensus algorithm, a number of nodes, and a data storage mode; set a node IP white list and a firewall rule, use an Ethereum compatible chain to write business logic, and complete local testing through Remix or Truffle; upload the contract to a cloud platform blockchain service, call the contract method through an API / SDK, and complete the deployment and setting of the blockchain; According to hVC （1） ~ hVC （na） , using TLS / SSL secure communication protocol to obtain the security parameters of the first to the nth customers, rc （1） ~ rc （na） ; rc （1） ~ rc （na） As input to the Setup algorithm, one gets a public key PK （1） ~ PK （na） , and a master secret key, denoted MSK （1） ~ MSK （na） ; The public key, the master key, and the ne （1） Using the KeyGen algorithm, the PK （1） ~ PK （na） and the MSK （1） ~ MSK （na） Generate the first to the na bit of the customer's key SK （1） ~ SK （na） , ID number id （1） ~ id （na） ; ne （1） represents the IP address of the 1st client; Using the RSA algorithm, the information mi is generated （1） ~ mi （na） The symmetric key SYK is obtained by using the symmetric key algorithm （1） ~ SYK （na） ; wherein the information mi （1） ~ mi （na） represents the financial information of the first to the nth customer. Encrypt algorithm as the encryption algorithm, using PK （1） ~ PK （na） , SK （1） ~ SK （na） and SYK （1） ~ SYK （na） , encrypt the information mi （1） ~ mi （na） , to get the ciphertext emi （1） ~ emi （na） .

4. The blockchain-based financial customer digitization archive management method of claim 3, wherein, The specific steps of dividing the customers into high net worth customers and ordinary customers are as follows: mon = max (mon1, mon2,... monn) （1） mon = max (mon1, mon2,... monn) （na） mon = max (mon1, mon2,... monn) （1） mon = max (mon1, mon2,... monn) （na） mon = max (mon1, mon2,... monn) （max） mon = max (mon1, mon2,... monn) Definition relation S-1: ; mon （k） represents the account balance of the kth customer; Customers satisfying the relationship S-1 are high net worth customers, and customers not satisfying the relationship S-1 are ordinary customers; Count the number nca of high net worth customers and the number ncb of ordinary customers; calculating a credit score cre for a high net worth client （1） ~ cre （nca） and setting a loan rate loa for the high net worth client as a function of the credit score （1） ~ loa （nca） ; Analyze the transaction time and transaction object of the ordinary customers to determine the frequently-visited shops of each ordinary customer; then, in combination with the transaction amount, classify the ordinary customers; Count the number of target customers, potential customers, and general customers of each frequently-visited shop. 5.The blockchain-based financial customer digitalization archive management method of claim 4, wherein, loa （1） The drafting steps are as follows: The market type to which the first high net worth customer belongs is a quasi-market; High net worth customers belonging to the quasi-market are quasi-customers, and the number of quasi-customers is ncl; get the account amount moy of all the prospective customers （1） ~ moy （ncl） ; compute mon （1） with moy （1） ~ moy （ncl） the sum sum; In pr （1） ~pr （nca） and upr （1） ~upr （nca） , the expected profit rate po （1） ~po （ncl） of the 1st to nth cl level customers is extracted, and the total profit rate Apr of the quasi market is calculated; the expected negative profit rate upo （1） ~upo （ncl） is extracted, and the total profit rate Aupr of the quasi market is calculated; wherein pr （1） ~pr （nca） and upr （1） ~upr （nca） respectively represent the expected profit rate and the expected negative profit rate of the 1st to nth ca high net worth customers; Calculate the net interest rate AUp of the quasi-market: AUp = Apr - Aupr; Calculate the combined yield rate REm of the quasi-market: ； wherein moy （l） represents the account balance of the lth level customer; Calculating the expected profit Qpr of the 1st high net worth customer, calculating the expected profit Qpo of the 1st to nthcl standard customers （1） ~ Qpo （ncl） ; Qpr, Qpo （1） ~ Qpo （ncl） rpr, rpo （1） ~ rpo （ncl） ; rpr, rpo （1） ~rpo （ncl） SSpr, the covariance SXpr of REm with rpr, rpo （1） ~rpo （ncl） ; Calculate the beta coefficient β of the quasi-market: β = SXpr / SSpr; Calculate the risk premium coefficient δ of the first high net worth customer: δ = β × (REm - rpr); cre （1） normalized to the interval 0 to 1, resulting in rcre; Let roa be the initial lending rate of the bank, and loa be the proposed lending rate （1） : loa （1） = roa + (rcre x delta). 6.The blockchain-based financial customer digitization archival management method of claim 4, wherein, The specific steps of classifying the ordinary customers are as follows: Define consumption tags: high frequency, low frequency, large amount, and small amount; Determine the frequently-visited shop of the first ordinary customer and classify the first ordinary customer; According to the transaction object, the total number of shops visited by the first ordinary customer in the past month is counted shp; The number of times the 1st general customer visits the 1st to shpth shop on the 1st day: fr （1，1） ~ fr （1，shp） ; Similarly, the number of times the 1st general customer visits the 1st to shpth shop on the datth day: fr （dat，1） ~ fr （dat，shp） ; Constructing matrix B （1） and compute the covariance matrix C: ; where matrix B （2） is the decentered matrix of matrix B （1） ; Matrix B （1） The mathematical expression for B is: ; Splitting the matrix C by columns results in the matrix Cc （1） ~ matrix Cc （shp） ; Definition of matrix Cc （1） Matrix Cc （shp） Weight λ （1） λ （shp） Bias μ （1） μ （shp） The relationship S-2 is constructed: ； Computing the values of λ （1） ~ λ （shp） and μ （1） ~ μ （shp） ; Calculate the frequency coefficient ba of the 1st ordinary customer about the 1st shop （1） : ba （1） = λ （1） / μ （1） ; similarly, calculate the frequency coefficient ba of the shpth shop （shp） ; Comparing ba （1） ~ ba （shp） The size of the largest frequency factor, the shop, as the first ordinary customer's regular shop; If the consumption label of the first ordinary customer is: high frequency and large amount, the first ordinary customer is the target customer of the frequently visited shop; If the consumption label is: high frequency and small amount or low frequency and large amount, the first ordinary customer is the potential customer of the frequently visited shop; If the consumption label is: low frequency and small amount, the first ordinary customer is the general customer of the frequently visited shop.

7. A blockchain-based financial customer digital archive management system, applied to the blockchain-based financial customer digital archive management method of any one of claims 1-6, characterized in that, The archive management system comprises: A data acquisition module: used to acquire customer credit information, eliminate substandard customers with unqualified credit, and obtain high-quality customers; acquire the identity information and financial information of high-quality customers; A data storage module: used to design the identity of customers according to the identity information of high-quality customers; construct a Merkle tree according to the number of high-quality customers, deploy the Merkle tree on a blockchain, and protect and save the financial information of customers; A customer analysis module: used to divide customers into high net worth customers and ordinary customers according to the account balance in the financial information; calculate the credit score of high net worth customers, and determine the loan interest rate of high net worth customers according to the credit score; analyze the transaction time and transaction object of ordinary customers to determine the frequently visited shop of each ordinary customer; then, combined with the transaction amount, count the number of target customers, potential customers and general customers of each frequently visited shop to obtain a customer consumption behavior report; A data update module: used to summarize the customer consumption behavior report and feed back to the user; collect the updated financial information of high-quality customers and synchronize the update on the blockchain.

8. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to run the steps in the method of any one of claims 1-6.

Images