Customer information deduplication and storage method and system based on multi-source data fusion

Abstract

The application discloses a customer information deduplication and storage method and system based on multi-source data fusion, and relates to the field of financial business data processing. The method acquires customer identity records, vehicle asset records, contract records and creditor's rights role records, constructs an asset business chain based on the connection relationship among vehicles, contracts, creditor's rights and asset packages, determines a business entry according to the appearance position of the customer identity records in the asset business chain, and determines the role attribution in combination with the creditor's rights role records. Then, the main file merging boundary, the role linking boundary and the conflict suspension boundary are formed according to the role attribution and the asset business chain. When storage, the system executes customer main file merging, business role linking and conflict record suspension processing according to the boundaries. The application can retain the vehicle, contract and creditor's rights role relationship while deduplicating the customer information, and position the conflict to a specific connection link, thereby reducing the risk of false merging and repeated file building.

Description

Technical Field

[0001] This invention relates to the field of financial business data processing technology, and in particular to a method and system for deduplicating and storing customer information based on multi-source data fusion. Background Technology

[0002] In consumer finance, auto finance, and credit asset management businesses, customer information typically comes from multiple source systems, including application systems, contract systems, vehicle mortgage systems, asset package receiving systems, and collection and disposal systems. These different source systems record customer numbers, contract numbers, vehicle identifiers, and creditor roles in inconsistent ways. The same individual may appear as a borrower, co-borrower, guarantor, actual vehicle user, debtor, or asset recipient in different business records.

[0003] Current methods for storing customer information typically rely on identity fields such as name, ID number, mobile phone number, and address for duplicate checks. When two customer records have the same or similar identity fields, the related customer records are merged into a single customer master file. This approach is suitable for handling duplicate basic customer information, but it has significant shortcomings in auto finance and credit asset scenarios.

[0004] Firstly, in the aforementioned scenarios, customer identity records are often linked to vehicles, contracts, claims, and asset packages in multiple layers, with the same individual playing different business roles in different business chains. If customer master files are merged directly based solely on identity fields, it is easy to overwrite the original contractual roles or claims during the merging process, resulting in the loss of vehicle asset relationships and the overwriting of claim roles. Consequently, business role relationships cannot be simultaneously preserved with the deduplicated customer master files.

[0005] Secondly, the existing method only makes duplicate judgments at the customer level and cannot distinguish which vehicle, which contract, or which debt assumption stage the difference between two suspected duplicate records occurs in. Therefore, it is difficult to locate the specific location of the role conflict. It can only take a coarse-grained approach of merging all or not merging all of the suspected duplicate records. It lacks the ability to distinguish between mergeable parts, attachable parts, and parts that need to be suspended.

[0006] Third, for records from different source systems with varying degrees of credibility, the existing methods lack a mechanism to isolate and suspend conflicting records before they are entered into the database. Once a mistaken merger occurs, it is difficult to trace back, which can easily lead to problems such as duplicate filing, mistaken merger, and inability to locate conflicting records.

[0007] Therefore, existing methods struggle to balance deduplication of customer master files and retention of business roles when processing multi-source financial business data. There is an urgent need for a customer information deduplication and data entry solution that can identify asset business chains, locate role conflicts, and form merging boundaries before data entry. Summary of the Invention

[0008] To address the aforementioned problems in existing technologies, this invention provides a method and system for deduplicating and storing customer information based on multi-source data fusion. The purpose is to identify the scope of customer master files, the scope of business roles that can be attached, and the scope of conflicting data to be suspended, based on the business continuity relationships between customer identity records, vehicle asset records, contract records, and creditor role records, before storing multi-source financial business data. This achieves deduplication of customer information while avoiding coverage of vehicle, contract, and creditor role relationships.

[0009] To achieve the above objectives, the first aspect of the present invention provides a method for deduplicating and storing customer information based on multi-source data fusion, comprising: Obtain financial business data to be entered into the database, including customer identity records, vehicle asset records, contract records, and creditor role records; An asset business chain is formed based on the vehicle asset records, the contract records, and the creditor role records. The asset business chain represents the business connection relationship between vehicles, contracts, creditors, and asset packages. Based on the location of the customer identity record in the asset business chain, determine the business entry point corresponding to the customer identity record; Based on the business entry point and the creditor role record, determine the role of the customer identity record in the asset business chain; Based on the role affiliation and the asset business chain, a customer merging boundary is formed, which includes the main file merging boundary, the role attachment boundary, and the conflict suspension boundary. Based on the customer merging boundary, the customer identity records are deduplicated and entered into the database, and the vehicle asset records, contract records, and creditor role records are linked to the customer master file after entry into the database.

[0010] Unlike existing methods that directly merge customer master files based solely on identity fields, this invention first constructs an asset business chain, placing each customer identity record into a business chain comprised of specific vehicles, contracts, claims, or asset packages. Then, based on the position of the customer identity record within the asset business chain and the claim role record, its business entry point and role affiliation are determined. Furthermore, based on role affiliation and the asset business chain, master file merging boundaries, role attachment boundaries, and conflict suspension boundaries are defined. Finally, differentiated deduplication and data entry processing is performed according to these boundaries. Thus, the merging of customer master files and the attachment to business roles are processed separately, and conflicting records are located in specific business chain segments and suspended individually, rather than merging or rejecting all suspected duplicate records indiscriminately.

[0011] Optionally, an asset business chain is formed based on the vehicle asset record, the contract record, and the creditor role record, including: The vehicle asset node is determined according to the vehicle identifier in the vehicle asset record, the contract node is determined according to the contract identifier in the contract record, and the creditor node is determined according to the creditor number in the creditor role record. By connecting the vehicle asset node, the contract node, and the debt node in the order of business occurrence, an asset business chain representing the business succession relationship between vehicles, contracts, debts, and asset packages is obtained.

[0012] Optionally, based on the business entry point and the creditor role record, the role of the customer identity record in the asset business chain is determined, including: Determine the business entry point where the customer identity record first appears in the asset business chain, and read the responsibility identity corresponding to the business entry point from the creditor role record; The business entry point and the liability identity are used together as the role basis for customer identity records, and the customer identity record is determined to belong to the main customer role, guarantor role, joint liability role or asset receiving role based on the role basis.

[0013] Optionally, based on the location of the customer identity record within the asset business chain, the business entry point corresponding to the customer identity record is determined, including: Read the vehicle asset node, contract node, or debt node that the customer identity record is first associated with, determine the business document corresponding to the customer identity record when it enters the asset business chain, and determine the business stage to which the business document belongs as the business entry point corresponding to the customer identity record.

[0014] Optionally, determining the role of customer identity records in the asset business chain based on business entry points and creditor role records also includes: If the business entry point corresponding to the customer identity record is the application entry point or the contract signing entry point, and the debtor role record contains a repayment responsibility identity, then the customer identity record will be determined as the main customer role. If the business entry point corresponding to the customer identity record is the guarantee entry point, and the debtor role record contains a guarantee liability identity, then the customer identity record will be identified as a guarantee role.

[0015] Optionally, based on the role affiliation and the asset business chain, a customer consolidation boundary is formed, including: Based on the identity clues in the customer identity records, customer identity records that are suspected to belong to the same natural person are grouped into the same candidate merging group; Within the candidate merging group, along the asset business chain corresponding to each customer identity record, the role change node where the role affiliation changes is determined; Based on the vehicle asset records, contract records, and creditor role records between adjacent role change nodes, a role succession segment of the candidate merging group is formed. Compare the customer master files and responsible parties pointed to by the different role-receiving sections to identify conflicting receiving sections that prevent the same customer master files from being directly merged. Based on the position of the conflict-handling segment in the asset business chain, the candidate merging group is divided into the main file merging range, the role attachment range, and the conflict suspension range, and the main file merging boundary, the role attachment boundary, and the conflict suspension boundary are formed respectively.

[0016] Optionally, by comparing the customer master files and responsible parties pointed to by different role-specific successor sections, conflicting successor sections that prevent the same customer master file from being directly merged are identified, including: For any role-based segment, determine the starting business entry point, ending business entry point, asset object to be inherited, and responsibility object to be inherited for that role-based segment; Write the asset receiving object and the responsibility receiving object into the chain segment placeholder record corresponding to the role receiving segment; If the chain segment placeholder records of different roles point to the same asset object, and the chain segment placeholder records of different roles occupy the same responsibility object position, then the conflicting undertaking segments are determined in sequence according to the source priority field and the write time field in the chain segment placeholder record. The source priority field is a numerical sorting field, with smaller values ​​indicating higher source trust levels. When the source priority field values ​​of two chain segment placeholder records are different, the role segment corresponding to the chain segment placeholder record with the larger source priority field value is identified as the conflicting role segment. When the source priority field values ​​of two chain segment placeholder records are the same, the role segment corresponding to the chain segment placeholder record with the larger timestamp value in the write time field is identified as the conflicting role segment. When both the source priority field and the write time field are the same, a conflict record to be reviewed is generated.

[0017] Optionally, based on the position of the conflict-receiving segment in the asset business chain, the candidate merging group is divided into a main file merging scope, a role attachment scope, and a conflict suspension scope, including: Customer identity records located before the conflict resolution section and pointing to the same customer master file are included in the master file merging scope; Records located on both sides of the conflict acceptance section and corresponding to independent vehicle asset records, contract records, or creditor role records are included in the role attachment scope; The customer identity record corresponding to the conflict acceptance segment is included in the conflict suspension scope.

[0018] Optionally, based on the customer merging boundary, the customer identity records are deduplicated and entered into the database, and the vehicle asset records, contract records, and creditor role records are linked to the customer master file after entry into the database, including: If the customer merging boundary is the master file merging boundary, then the customer identity record is written into the same customer master file; If the customer merging boundary includes a role attachment boundary, then the vehicle asset record, the contract record, and the debt role record will be attached to the customer master file according to their role affiliation. If the customer merge boundary is a conflict suspension boundary, then the merging of the master files of the customer identity records is suspended, and a pending confirmation entry record is generated.

[0019] A second aspect of the present invention provides a customer information deduplication and database entry system based on multi-source data fusion, comprising: The data acquisition module is used to acquire financial business data to be entered into the database. The financial business data to be entered into the database includes customer identity records, vehicle asset records, contract records, and creditor role records. The asset business chain formation module is used to form an asset business chain based on the vehicle asset record, the contract record, and the creditor role record. The asset business chain represents the business connection relationship between vehicles, contracts, creditors, and asset packages. The business entry point determination module is used to determine the business entry point corresponding to the customer identity record based on the position of the customer identity record in the asset business chain. The role attribution determination module is used to determine the role attribution of the customer identity record in the asset business chain based on the business entry point and the creditor role record; The merge boundary forming module is used to form customer merge boundaries based on the role affiliation and the asset business chain. The customer merge boundaries include the main file merge boundary, the role attachment boundary, and the conflict suspension boundary. The deduplication and entry module is used to deduplicate the customer identity records according to the customer merging boundary, and to link the vehicle asset records, contract records, and creditor role records to the customer master file after entry.

[0020] The beneficial effects of this invention are as follows: it can identify the asset business chain between vehicles, contracts, claims, and asset packages before deduplicating customer information into the database, merge customer master files and separate them from business roles, and locate conflicts to specific business chain segments through role-inherited segments and chain segment placeholder records, suspending only the conflict-inherited segments. This ensures that only one customer master file is retained for the same natural person, while avoiding the loss of vehicle asset relationships and the overwriting of claim roles, and reducing the problems of duplicate file creation, erroneous merging, and inability to locate conflict records. Attached Figure Description

[0021] To more clearly illustrate the technical solutions of the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a flowchart illustrating the method for deduplicating and storing customer information based on multi-source data fusion, as provided in an embodiment of the present invention.

[0023] Figure 2 This is a schematic diagram of the asset business chain structure provided for an embodiment of the present invention.

[0024] Figure 3 This is a schematic diagram of the role-transfer segment and conflict-transfer segment provided in an embodiment of the present invention.

[0025] Figure 4 This is a schematic diagram of the structure of a customer information deduplication and data entry system based on multi-source data fusion, provided in an embodiment of the present invention. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0027] This invention provides a method for deduplicating and storing customer information based on multi-source data fusion. For example... Figure 1 As shown, the method is executed along the main lines of steps S1 to S6, where steps S1 to S6 correspond sequentially to data acquisition, forming an asset business chain, determining the business entry point, determining role affiliation, forming customer merging boundaries, and deduplication and data entry. It should be noted that, for ease of understanding, the following example illustrates the "mixed entry of auto finance application data and credit asset package receipt data." This example involves four records to be entered from four source systems, denoted as record A, record B, record C, and record D. All four records point to a name "Zhang Mou" and a mobile phone number "138". The customer lead is "0001", but its source system, associated asset contract, role field, source priority, and write time are all different.

[0028] Step S1: Obtain financial business data to be entered into the database.

[0029] In step S1, the system receives pending financial business data from multiple business systems, including the application system, contract system, vehicle mortgage system, asset package receiving system, and collection and disposal system. This pending financial business data includes customer identity records, vehicle asset records, contract records, and creditor role records.

[0030] The customer identity record is used to represent basic customer identity clues, and its fields include at least one of customer number, name, fragment of ID number, mobile phone number, address, and source system number. The vehicle asset record is used to represent vehicle asset objects in auto finance business, and its fields include at least one of vehicle identification number, license plate number, mortgage number, and vehicle order number. The contract record is used to represent loan contracts, financing contracts, guarantee contracts, debt transfer contracts, or disposal contracts, and its fields include at least one of contract number, contract type, signing time, and contract status. The debtor role record is used to represent the customer's liability role in the debt or contract, and its values ​​include at least one of borrower, co-borrower, guarantor, debtor, and asset recipient.

[0031] Unlike methods that directly rely on identity fields for deduplication, the system does not immediately determine whether a customer is a duplicate when receiving the above data. Instead, it retains a source identifier and business field for each record, enabling subsequent processing to trace which system the record came from, which business process it corresponds to, and which role it represents.

[0032] Taking this embodiment as an example, the source system, customer lead, associated asset or contract, role field, source priority, and write time of the four records to be added to the database are as follows: Record A comes from the auto finance application system, and the customer lead is "Zhang, 138". "0001" is associated with vehicle VIN001 and contract HT001, with the role field being the borrower, the source priority being high, and the write time being 10:01; Record B comes from the guarantee contract system, is associated with contract HT002, has the role field as guarantor, has a source priority of high, and was written at 10:03. Record C comes from the credit asset package system, is associated with asset package ZCB01 and debt ZQ003, has the role field as debtor, source priority as medium, and was written at 10:05. Record D comes from an external asset package system, is associated with contract HT001, has the role field as debtor, has a low source priority, and was written at 10:07.

[0033] Step S2: Form an asset business chain based on vehicle asset records, contract records, and creditor role records.

[0034] In step S2, the system forms an asset business chain based on vehicle asset records, contract records, and creditor role records. The asset business chain represents the business connection relationship between vehicles, contracts, creditors, and asset packages.

[0035] Specifically, the system determines vehicle asset nodes based on vehicle identifiers in vehicle asset records, contract nodes based on contract identifiers in contract records, and debt nodes based on debt number in debt role records. Then, the system connects the vehicle asset nodes, contract nodes, and debt nodes according to the order of business transactions, resulting in an asset business chain representing the business relationships between vehicles, contracts, debts, and asset packages. In the auto finance scenario, this order of business transactions represents the reference relationship between vehicle orders, financing contracts, mortgage records, and debt records. In the credit asset scenario, the system further determines asset package nodes based on asset package numbers and connects the nodes according to the business order of asset package receipt, debt assumption, contract association, and role assumption.

[0036] Taking this embodiment as an example, such as Figure 2 As shown, the system forms four asset business chains: The business chain L1 is "Vehicle VIN001 → Contract HT001 → Claim ZQ001 → Borrower Role"; The business chain L2 is "Contract HT002 → Guarantee Liability → Guarantor Role"; The business chain L3 is "Asset package ZCB01 → Claim ZQ003 → Debtor role"; The business chain L4 is "Contract HT001 → Debtor Role".

[0037] Once the asset business chain is formed, each customer identity record is no longer regarded as isolated customer data, but is placed into the business chain of specific vehicles, contracts, claims or asset packages, thus providing a basis for subsequent judgments on whether customer records should be merged and whether role relationships should be retained.

[0038] Step S3: Determine the business entry point corresponding to the customer identity record based on its location in the asset business chain.

[0039] In step S3, the system determines the business entry point corresponding to the customer identity record based on its position in the asset business chain. Specifically, the system reads the vehicle asset node, contract node, or debt node that the customer identity record is first associated with, determines the business document corresponding to the customer identity record when it enters the asset business chain, and identifies the business stage to which the business document belongs as the business entry point corresponding to the customer identity record.

[0040] In one implementation, the correspondence between business entry points and business documents is as follows: if a customer's identity record appears for the first time in a loan application or car purchase application, the business entry point is designated as the application entry point; if a customer's identity record appears for the first time in the signing field of a loan contract, financing contract, or borrowing contract, the business entry point is designated as the signing entry point; if a customer's identity record appears for the first time in the guarantee contract or guarantee liability field, the business entry point is designated as the guarantee entry point; if a customer's identity record appears for the first time in an asset package receipt record or debt transfer record, the business entry point is designated as the asset transfer entry point; if a customer's identity record appears for the first time in a collection disposal or asset disposal receipt record, the business entry point is designated as the disposal receipt entry point.

[0041] Taking this embodiment as an example, record A first appears in the signing position of contract HT001, and its business entry point is the signing entry point; record B first appears in the guarantee field of contract HT002, and its business entry point is the guarantee entry point; record C first appears in the receiving record of asset package ZCB01, and its business entry point is the asset transfer entry point; record D first appears in the debtor field of contract HT001, and its business entry point is the asset transfer entry point. Through the business entry points, the system can distinguish whether a customer identity record enters the business chain as a primary customer, or as a guarantor, debtor, actual user, or asset disposal object, thereby avoiding relying solely on identity fields such as name and mobile phone number for merging judgments.

[0042] Step S4: Determine the role of the customer's identity record in the asset business chain based on the business entry point and the creditor role record.

[0043] In step S4, the system determines the role of the customer identity record in the asset business chain based on the business entry point and the creditor role record. Specifically, the system determines the business entry point where the customer identity record first appears in the asset business chain, reads the liability identity corresponding to that business entry point from the creditor role record, uses both the business entry point and the liability identity as the basis for the customer identity record's role, and determines whether the customer identity record belongs to the primary customer role, guarantor role, joint liability role, or asset receiving role based on the role basis.

[0044] In one implementation, the correspondence between roles and role affiliation is as follows: If the business entry point corresponding to the customer identity record is the application entry point or the contract signing entry point, and the debtor role record contains a repayment responsibility identity, then the customer identity record is determined to be the primary customer role; if the business entry point corresponding to the customer identity record is the guarantee entry point, and the debtor role record contains a guarantee responsibility identity, then the customer identity record is determined to be the guarantee role; if the business entry point is the asset transfer entry point, and the debtor role record contains a debtor identity or a debtor successor identity, then the customer identity record is determined to be the debtor role or the asset receiving role; if the business entry point is the disposal receiving entry point, then the customer identity record is determined to be the disposal object or the asset receiving role based on the responsible object in the disposal record.

[0045] Taking this embodiment as an example, record A's business entry point is the contract signing entry point and it has a repayment responsibility identity, so it is identified as the main customer role; record B's business entry point is the guarantee entry point and it has a guarantee responsibility identity, so it is identified as the guarantor role; records C and D's business entry points are asset transfer entry points and they have debtor identities, so they are identified as debtor roles. It should be noted that role assignment is not the final merge result, but is used to form the role succession segment later; even if two customer identity records are considered to be suspected of belonging to the same natural person, the system still retains their role assignment in different asset business chains to avoid covering the original contract roles or creditor responsibilities when merging customer master files.

[0046] Step S5: Based on role affiliation and asset business chain, form customer consolidation boundaries.

[0047] In step S5, the system forms customer merging boundaries based on role affiliation and asset business chain. These customer merging boundaries include master file merging boundaries, role attachment boundaries, and conflict suspension boundaries. To form these customer merging boundaries, step S5 may further include sub-steps S51 to S55 in one implementation.

[0048] S51: Based on the identity clues in the customer identity records, customer identity records that are suspected to belong to the same natural person are grouped into the same candidate merging group.

[0049] The system groups customer identity records that are suspected to belong to the same natural person into the same candidate merging group based on identity clues in the customer identity records. These identity clues include at least one of the following: name, document fragment, mobile phone number, address, source system customer number, vehicle association, or contract association. If two customer identity records have the same name and mobile phone number, or identical document fragments, then the two customer identity records are grouped into the same candidate merging group. A candidate merging group only indicates that the related customer identity records may belong to the same natural person; it does not mean that they can be directly merged. For example, in this embodiment, records A, B, C, and D have the same name (Zhang) and the same mobile phone number (138). (0001), the system will assign the four records to the same candidate merge group, but will not merge them directly at this time.

[0050] In this embodiment, the candidate merge group serves only as the input set for subsequent chain segment comparisons. Its data items include the candidate merge group number, the set of customer identity record numbers, the identity clue matching result, and the matching basis. The identity clue matching result can be generated by at least one rule among the following: matching document number, matching mobile phone number, matching name and vehicle association, and matching source system customer number and contract association. Among these, the matching basis of matching document number is preferentially used to form the candidate merge group, while other matching basis is used to supplement the collection and enter the subsequent role-transfer segment verification.

[0051] S52: Within the candidate merging group, along the asset business chain corresponding to each customer identity record, determine the role change node where the role affiliation changes.

[0052] Within the candidate merging group, the system identifies role change nodes along the asset business chain corresponding to each customer identity record, where role affiliation changes. These role change nodes are the points where a customer transitions from an initiating role to a contractual liability role, or from a contractual liability role to a guarantee liability role, a debt assumption role, or a disposal receiving role. The system compares the role affiliation of customer identity records on adjacent business nodes node by node; when the role affiliation on adjacent business nodes is inconsistent, the business node where the role affiliation changes is identified as the role change node.

[0053] The role change node is used to identify business nodes in the same candidate merging group where the position of the responsible object occupied by the customer's identity record on the asset business chain changes. The position of the responsible object is a preset role slot under the receiving asset object, including the borrower responsibility slot, co-borrower responsibility slot, guarantee responsibility slot, debt responsibility slot, and asset receiving responsibility slot. When adjacent business nodes correspond to different role slots, or adjacent business nodes correspond to different receiving asset objects but have the same responsible object position, the system records the corresponding node as a role change node.

[0054] S53: Based on the vehicle asset records, contract records, and creditor role records between adjacent role change nodes, form the role succession segment of the candidate merging group.

[0055] Taking this embodiment as an example, such as Figure 3As shown, the system first forms a role-based transition segment based on the customer's entry location, the asset they assume, and the liability they assume. Specifically, if record A enters contract HT001 through the signing entry and assumes repayment responsibility, the system forms role-based transition segment D1, where the asset they assume is contract HT001 and the liability they assume is the repayment responsibility position. If record B enters contract HT002 through the guarantee entry and assumes guarantee responsibility, the system forms role-based transition segment D2, where the asset they assume is contract HT002 and the liability they assume is the guarantee responsibility position. If record C enters debt ZQ003 through the asset transfer entry and assumes debt responsibility, the system forms role-based transition segment D3, where the asset they assume is debt ZQ003 and the liability they assume is the debt responsibility position. If record D enters contract HT001 through the asset transfer entry and assumes repayment responsibility, the system forms role-based transition segment D4, where the asset they assume is contract HT001 and the liability they assume is the repayment responsibility position. Through the role handover segment, the system transforms the question of "whether the customer is duplicated" into the question of "which asset objects and responsibility objects the same candidate customer occupies in different business segments".

[0056] S54: Compare the customer master files and responsible objects pointed to by the different role-receiving segments to identify conflicting segments that prevent the same customer master file from being directly merged.

[0057] For any given role-based segment, the system determines the starting business entry point, ending business entry point, inherited asset object, and inherited responsibility object for that segment, and writes the inherited asset object and the inherited responsibility object into the corresponding chain segment placeholder record. The chain segment placeholder record includes at least the role-based segment number, customer identity record, starting business entry point, ending business entry point, inherited asset object, inherited responsibility object, source priority, and write time. The chain segment placeholder record maps the role-based segment to a comparable chain segment occupancy status.

[0058] For ease of comparison, the source priority field, write time field, and responsible object field in the chain segment placeholder record are all stored using structured fields. The source priority field is an integer sorted value; the write time field is a timestamp generated when the system receives the data into the database; the responsible object field consists of the responsible asset object identifier and the position of the responsible object, such as contract HT001-repayment responsibility position. Therefore, the same responsible object position under the same responsible asset object can only correspond to one chain segment placeholder record that can be directly entered into the database.

[0059] The system compares the placeholder records of chain segments for different roles within the candidate merge group. If the placeholder records of chain segments for different roles point to the same asset object, and the placeholder records of chain segments for different roles occupy the same responsibility object position, then a responsibility object position conflict is determined, and the conflicting takeover segment is determined according to the preset order of the source priority field and the write time field.

[0060] In one implementation, the source priority field uses a numerical sorting value, with smaller values ​​indicating higher source trustworthiness. When the source priority field values ​​of two placeholder records are different, the role segment corresponding to the placeholder record with the larger source priority field value is identified as the conflicting segment. When the source priority field values ​​of two placeholder records are the same, the role segment corresponding to the placeholder record with the larger timestamp value in the write time field is identified as the conflicting segment. When both the source priority field and the write time field are the same, the system generates a conflict record to be reviewed and maintains the conflict suspension state of the corresponding role segment until manual review or supplementary vouchers are provided. The source priority field is pre-configured by the business system; for example, it is configured as 1 for the core contract system, 1 for the guarantee contract system, 2 for the credit asset package system, and 3 for the external asset package system. The source priority field can also be generated based on the data source, contract status, business time, and record completeness, but the same sorting configuration should be used for the same batch of inbound tasks.

[0061] In this embodiment, the source priority field value is pre-configured based on the proximity of the source system to the original business document. The auto finance application system and guarantee contract system are directly generated during the customer application, signing, or guarantee business process, and their records originate from the original contract processing. Therefore, the source priority field value for both the auto finance application system and guarantee contract system is configured as 1. The data in the credit asset package system originates from the asset package receiving stage after debt assumption, and it undergoes one business assumption relative to the original application and contract systems. Therefore, the source priority field value for the credit asset package system is configured as 2. The data in the external asset package system originates from the external receiving or import stage, and its records undergo external transmission relative to the original contract processing. Therefore, the source priority field value for the external asset package system is configured as 3. Thus, record A comes from the auto finance application system, corresponding to a source priority field value of 1; record B comes from the guarantee contract system, corresponding to a source priority field value of 1; record C comes from the credit asset package system, corresponding to a source priority field value of 2; and record D comes from the external asset package system, corresponding to a source priority field value of 3. This source priority field value is used to determine the conflicting assumption segment when there is an occupation conflict at the same asset and liability object location.

[0062] Furthermore, such as Figure 3As shown, the system maps role-bearing segments D1 to D4 to chain segment placeholder records P1 to P4 respectively. P1 corresponds to D1, its asset-bearing object is contract HT001, its liability-bearing object is the repayment liability position, the source priority field value is 1, and the write time field value is 10:01; P2 corresponds to D2, its asset-bearing object is contract HT002, its liability-bearing object is the guarantee liability position, the source priority field value is 1, and the write time field value is 10:03; P3 corresponds to D3, its asset-bearing object is debt ZQ003, its liability-bearing object is the debt liability position, the source priority field value is 2, and the write time field value is 10:05; P4 corresponds to D4, its asset-bearing object is contract HT001, its liability-bearing object is the repayment liability position, the source priority field value is 3, and the write time field value is 10:07. After comparing P1 to P4, the system found that both P1 and P4 point to contract HT001 and both occupy the repayment responsibility position, indicating a conflict in the location of the responsible party. Since the source priority field value 3 of P4 is greater than the source priority field value 1 of P1, the system identifies the role succession segment D4 corresponding to P4 as the conflicting succession segment.

[0063] By using placeholder records for chain segments, conflicting segments are used to indicate specific chain segments that cannot be directly merged into the database, rather than rejecting the entire candidate merge group. If two role segments point to different contracts, vehicles, claims, or liability objects, the system allows the client master files to be merged, while attaching the different role relationships separately; only when two role segments point to the same liability object under the same asset object are conflicting segments identified and suspended.

[0064] S55: Based on the position of the conflict-bearing segment in the asset business chain, the candidate merging group is divided into the main file merging range, the role attachment range, and the conflict suspension range, and the main file merging boundary, the role attachment boundary, and the conflict suspension boundary are formed respectively.

[0065] Based on the position of the conflict-receiving segment in the asset business chain, the system divides candidate merging groups into main file merging scope, role attachment scope, and conflict suspension scope, and respectively forms main file merging boundaries, role attachment boundaries, and conflict suspension boundaries. Specifically, the system includes customer identity records located before the conflict-receiving segment and pointing to the same customer main file in the main file merging scope; records located on both sides of the conflict-receiving segment and corresponding to independent vehicle asset records, contract records, or debt role records in the role attachment scope; and customer identity records corresponding to the conflict-receiving segment in the conflict suspension scope.

[0066] When establishing master file merging boundaries, role attachment boundaries, and conflict suspension boundaries, the system uses the location of the receiving asset object and the responsible object corresponding to the conflict acceptance segment as the dividing benchmark. Customer identity records that do not occupy the same responsible object location as the conflict acceptance segment but can point to the same customer master file are included in the master file merging scope; records corresponding to independent receiving asset objects or independent responsible object locations are included in the role attachment scope; records that occupy the same receiving asset object and the same responsible object location as the conflict acceptance segment are included in the conflict suspension scope, and a pending confirmation entry record is generated.

[0067] If there are no conflicting segments in the candidate merge group, the system establishes a master file merge boundary and a role attachment boundary. If there are conflicting segments in the candidate merge group, the system retains the mergeable and attachable portions while establishing a conflict suspension boundary for the conflicting segments. Taking this embodiment as an example, records A, B, and C, located before conflicting segment D4 and pointing to the same customer master file, are included in the master file merge scope. Role attachment segments D1, D2, and D3 correspond to the independent roles of borrower (contract HT001), guarantor (contract HT002), and debtor (creditor ZQ003), respectively, and are included in the role attachment scope. Conflicting segment D4 is included in the conflict suspension scope. Accordingly, the system establishes a master file merge boundary (records A, B, and C), a role attachment boundary (D1, D2, and D3), and a conflict suspension boundary (D4).

[0068] Step S6: Based on the customer merging boundary, deduplicate the customer identity records and add them to the database, and link the vehicle asset records, contract records, and creditor role records to the customer master file after they are added to the database.

[0069] Based on the customer merging boundary formed in step S5, the system deduplicates and stores customer identity records, and links vehicle asset records, contract records, and debtor role records to the customer master file after storage. Specifically, if the customer merging boundary is the master file merging boundary, the system writes customer identity records within the master file merging boundary into the same customer master file; if the customer merging boundary includes a role attachment boundary, the system attaches vehicle asset records, contract records, and debtor role records within the role attachment boundary to the customer master file according to the role affiliation determined in step S4; if the customer merging boundary is a conflict suspension boundary, the system suspends the master file merging of customer identity records within the conflict suspension boundary and generates a record to be confirmed for storage.

[0070] Taking this embodiment as an example, the system writes records A, B, and C within the master file merging boundary into the same customer master file, so that the natural person Zhang retains only one customer master file; it links vehicle VIN001 to Zhang's customer master file; it links contract HT001 to Zhang's customer master file according to the borrower role, and contract HT002 to Zhang's customer master file according to the guarantor role; it links debt ZQ003 to Zhang's customer master file according to the debtor role; for record D and its corresponding role-bearing segment D4 within the conflict suspension boundary, the system pauses master file merging and generates a record to be confirmed for entry. After the entry is completed, the natural person Zhang retains only one customer master file, and the vehicle, contract, debt, and role relationships will not be overwritten due to the customer master file merging.

[0071] In one implementation, pending entry records within the conflict suspension boundary can enter a manual review process, where business personnel confirm the responsible party for the conflict transfer based on the contract status, debt certificate, and asset receipt certificate before executing the entry. In another implementation, pending entry records can also enter a subsequent automatic review process, where the system automatically releases the suspension and completes the entry after obtaining a record with higher source priority or supplementary certificate.

[0072] This invention identifies the asset business chain between vehicles, contracts, receivables, and asset packages before deduplicating customer information into the database. This transforms isolated customer identity determination into an assessment of the asset and liability objects occupied by the customer's identity record within the asset business chain. By determining the business entry point and role affiliation of customer identity records, this invention can distinguish between merging customer master files and associating them with business roles. While merging customer master files, it retains the customer's business roles in different asset business chains, preventing the loss of vehicle asset relationships and the overwriting of receivable roles. Through role-inheritance segments and chain segment placeholder records, this invention can pinpoint conflicts to specific vehicle, contract, or receivable inheritance stages, suspending only the conflicting inheritance segment rather than rejecting the entire candidate merging group. This isolates conflicting records while retaining mergeable and associative portions. Therefore, this invention reduces problems such as duplicate file creation, erroneous merging, loss of vehicle asset relationships, overwriting of receivable roles, and inability to locate conflicting records.

[0073] like Figure 4 As shown, the present invention also provides a customer information deduplication and entry system based on multi-source data fusion, used to execute the above-described customer information deduplication and entry method. The system includes a data acquisition module M1, an asset business chain formation module M2, a business entry point determination module M3, a role attribution determination module M4, a merging boundary formation module M5, and a deduplication and entry module M6.

[0074] The data acquisition module M1 is used to acquire financial business data to be entered into the database. The financial business data to be entered into the database includes customer identity records, vehicle asset records, contract records and creditor role records. Its function corresponds to step S1.

[0075] The asset business chain formation module M2 is used to form an asset business chain based on vehicle asset records, contract records, and creditor role records. The asset business chain represents the business connection relationship between vehicles, contracts, creditors, and asset packages, and its function corresponds to step S2.

[0076] The business entry point determination module M3 is used to determine the business entry point corresponding to the customer identity record based on the position of the customer identity record in the asset business chain. Its function corresponds to step S3.

[0077] The role attribution determination module M4 is used to determine the role attribution of customer identity records in the asset business chain based on business entry and creditor role records. Its function corresponds to step S4.

[0078] The merge boundary formation module M5 is used to form customer merge boundaries, including master file merge boundaries, role attachment boundaries, and conflict suspension boundaries, based on role affiliation and asset business chain. Its function corresponds to step S5.

[0079] The deduplication and entry module M6 is used to deduplicat customer identity records and enter them into the database based on the customer merging boundary. It also links vehicle asset records, contract records, and creditor role records to the customer master file after entry. Its function corresponds to step S6.

[0080] In one implementation, the merge boundary forming module M5 may further include a candidate merge group forming unit, a role-assigning segment forming unit, a chain segment placeholder record generating unit, and a conflict-assigning segment determining unit, which are respectively used to execute the aforementioned sub-steps S51 to S54; the deduplication and entry module M6 may further include a master file merging unit, a role-attaching unit, and a conflict suspension unit, which are respectively used to perform master file writing, role-based attachment, and generation of records to be confirmed for entry. The above modules can be implemented by a processor executing computer programs stored in memory, or by hardware such as a field-programmable gate array or an application-specific integrated circuit; this invention does not limit this.

[0081] The foregoing embodiments illustrate the use of mixed data entry for auto finance applications and credit asset package receipts as an example, but the present invention is not limited thereto. In other embodiments, the asset objects in the asset business chain may include, in addition to vehicle assets, contracts, and receivables, insurance policies, collateral, credit lines, collection cases, or disposal batches. Correspondingly, the business entry points may include, in addition to the application entry point, contract signing entry point, guarantee entry point, asset transfer entry point, and disposal receipt entry point, other business stages corresponding to the aforementioned asset objects.

[0082] In other implementations, the source priority used to determine the conflict resolution segment can be pre-configured by the business system or dynamically determined based on the data source, contract status, business time, and record completeness. The write time used to determine the conflict resolution segment can be replaced by the business occurrence time, contract effective time, or asset receipt time. Records within the conflict suspension range can either enter a manual review process or a subsequent automatic review process. All of the above substitutions fall within the protection scope of this invention.

Claims

1. A method for deduplicating and storing customer information based on multi-source data fusion, characterized in that, include: Obtain financial business data to be entered into the database, including customer identity records, vehicle asset records, contract records, and creditor role records; An asset business chain is formed based on the vehicle asset records, the contract records, and the creditor role records. The asset business chain represents the business connection relationship between vehicles, contracts, creditors, and asset packages. Based on the location of the customer identity record in the asset business chain, determine the business entry point corresponding to the customer identity record; Based on the business entry point and the creditor role record, determine the role of the customer identity record in the asset business chain; Based on the role affiliation and the asset business chain, a customer merging boundary is formed, which includes the main file merging boundary, the role attachment boundary, and the conflict suspension boundary. Based on the customer merging boundary, the customer identity records are deduplicated and entered into the database, and the vehicle asset records, contract records, and creditor role records are linked to the customer master file after entry into the database.

2. The method for deduplicating and storing customer information based on multi-source data fusion according to claim 1, characterized in that, An asset business chain is formed based on the vehicle asset records, the contract records, and the creditor role records, including: The vehicle asset node is determined according to the vehicle identifier in the vehicle asset record, the contract node is determined according to the contract identifier in the contract record, and the creditor node is determined according to the creditor number in the creditor role record. By connecting the vehicle asset node, the contract node, and the debt node in the order of business occurrence, an asset business chain representing the business succession relationship between vehicles, contracts, debts, and asset packages is obtained.

3. The method for deduplicating and storing customer information based on multi-source data fusion according to claim 1, characterized in that, Based on the business entry point and the creditor role record, determine the role of the customer identity record within the asset business chain, including: Determine the business entry point where the customer identity record first appears in the asset business chain, and read the responsibility identity corresponding to the business entry point from the creditor role record; The business entry point and the liability identity are used together as the role basis for customer identity records, and the customer identity record is determined to belong to the main customer role, guarantor role, joint liability role or asset receiving role based on the role basis.

4. The method for deduplicating and storing customer information based on multi-source data fusion according to claim 1, characterized in that, Based on the location of the customer identity record within the asset business chain, the business entry point corresponding to the customer identity record is determined, including: Read the vehicle asset node, contract node, or debt node that the customer identity record is first associated with, determine the business document corresponding to the customer identity record when it enters the asset business chain, and determine the business stage to which the business document belongs as the business entry point corresponding to the customer identity record.

5. The method for deduplicating and storing customer information based on multi-source data fusion according to claim 1, characterized in that, Based on business entry points and creditor role records, determining the role of customer identity records within the asset business chain also includes: If the business entry point corresponding to the customer identity record is the application entry point or the contract signing entry point, and the debtor role record contains a repayment responsibility identity, then the customer identity record will be determined as the main customer role. If the business entry point corresponding to the customer identity record is the guarantee entry point, and the debtor role record contains a guarantee liability identity, then the customer identity record will be identified as a guarantee role.

6. The method for deduplicating and storing customer information based on multi-source data fusion according to claim 1, characterized in that, Based on the aforementioned role affiliation and the aforementioned asset business chain, a customer consolidation boundary is formed, including: Based on the identity clues in the customer identity records, customer identity records that are suspected to belong to the same natural person are grouped into the same candidate merging group; Within the candidate merging group, along the asset business chain corresponding to each customer identity record, the role change node where the role affiliation changes is determined; Based on the vehicle asset records, contract records, and creditor role records between adjacent role change nodes, a role succession segment of the candidate merging group is formed. Compare the customer master files and responsible parties pointed to by the different role-receiving sections to identify conflicting receiving sections that prevent the same customer master files from being directly merged. Based on the position of the conflict-handling segment in the asset business chain, the candidate merging group is divided into the main file merging range, the role attachment range, and the conflict suspension range, and the main file merging boundary, the role attachment boundary, and the conflict suspension boundary are formed respectively.

7. The method for deduplicating and storing customer information based on multi-source data fusion according to claim 6, characterized in that, By comparing the customer master files and responsible parties pointed to by the different role-specific successor sections, conflicting successor sections that prevent the same customer master file from being directly merged are identified, including: For any role-based segment, determine the starting business entry point, ending business entry point, asset object to be inherited, and responsibility object to be inherited for that role-based segment; Write the asset receiving object and the responsibility receiving object into the chain segment placeholder record corresponding to the role receiving segment; If the chain segment placeholder records of different roles point to the same asset object, and the chain segment placeholder records of different roles occupy the same responsibility object position, then the conflicting undertaking segments are determined in sequence according to the source priority field and the write time field in the chain segment placeholder record. The source priority field is a numerical sorting field, with smaller values ​​indicating higher source trust levels. When the source priority field values ​​of two chain segment placeholder records are different, the role segment corresponding to the chain segment placeholder record with the larger source priority field value is identified as the conflicting role segment. When the source priority field values ​​of two chain segment placeholder records are the same, the role segment corresponding to the chain segment placeholder record with the larger timestamp value in the write time field is identified as the conflicting role segment. When both the source priority field and the write time field are the same, a conflict record to be reviewed is generated.

8. The method for deduplicating and storing customer information based on multi-source data fusion according to claim 6, characterized in that, Based on the position of the conflict-handling segment in the asset business chain, the candidate merging group is divided into a main file merging scope, a role-attached scope, and a conflict-suspended scope, including: Customer identity records located before the conflict resolution section and pointing to the same customer master file are included in the master file merging scope; Records located on both sides of the conflict acceptance section and corresponding to independent vehicle asset records, contract records, or creditor role records are included in the role attachment scope; The customer identity record corresponding to the conflict acceptance segment is included in the conflict suspension scope.

9. The method for deduplicating and storing customer information based on multi-source data fusion according to claim 1, characterized in that, Based on the customer merging boundary, the customer identity records are deduplicated and entered into the database, and the vehicle asset records, contract records, and creditor role records are linked to the customer master file after entry into the database, including: If the customer merging boundary is the master file merging boundary, then the customer identity record is written into the same customer master file; If the customer merging boundary includes a role attachment boundary, then the vehicle asset record, the contract record, and the debt role record will be attached to the customer master file according to their role affiliation. If the customer merge boundary is a conflict suspension boundary, then the merging of the master files of the customer identity records is suspended, and a pending confirmation entry record is generated.

10. A customer information deduplication and data entry system based on multi-source data fusion, characterized in that, include: The data acquisition module is used to acquire financial business data to be entered into the database. The financial business data to be entered into the database includes customer identity records, vehicle asset records, contract records, and creditor role records. The asset business chain formation module is used to form an asset business chain based on the vehicle asset record, the contract record, and the creditor role record. The asset business chain represents the business connection relationship between vehicles, contracts, creditors, and asset packages. The business entry point determination module is used to determine the business entry point corresponding to the customer identity record based on the position of the customer identity record in the asset business chain. The role attribution determination module is used to determine the role attribution of the customer identity record in the asset business chain based on the business entry point and the creditor role record; The merge boundary forming module is used to form customer merge boundaries based on the role affiliation and the asset business chain. The customer merge boundaries include main file merge boundaries, role attachment boundaries, and conflict suspension boundaries. The deduplication and entry module is used to deduplicate the customer identity records according to the customer merging boundary, and to link the vehicle asset records, contract records, and creditor role records to the customer master file after entry.

Images