Intelligent building method and system for future community member architecture

Abstract

The invention relates to the technical field of intelligent community data governance and main identity normalization management, and discloses a future community member architecture intelligent building method and system, and the method comprises the steps: carrying out the field semantic standardization of structured resident data; fusing the unstructured data; generating a ciphertext identifier; executing cross-system identity matching in combination with a field similarity function; and generating a unique main identity file. Compared with the prior art, the technical problems that data of multiple departments are independently managed, field structures are heterogeneous, a unified identity primary key is lacked, and particularly, high-precision identity fusion and normalization management cannot be realized under the complex data conditions of name duplication, different address expressions, field missing or untrusted sources and the like of residents are solved. According to the method, cross-system multi-source data fusion and identity uniqueness determination are realized through semantic hash slot construction, image-text joint recognition, multi-source fuzzy matching and a credible main file generation mechanism, and the uniqueness and credibility of future community resident identity data are improved.

Description

Technical Field

[0001] This invention relates to the field of smart community data governance and unified management of primary identity, and in particular to a method and system for intelligently constructing a future community member structure. Background Technology

[0002] Currently, in the actual operation and governance of future communities, various types of information related to resident identity are distributed across multiple heterogeneous business systems, such as property management systems, street and community grid systems, education service platforms, and primary healthcare systems. These systems are mostly independently deployed and built in a decentralized manner, with significant differences in the data structures, field naming conventions, data update frequencies, and reliable sources used. This makes it difficult to directly establish mapping relationships between records of the same resident identity across different systems.

[0003] Specifically, different systems use different names for basic fields such as "name," "address," and "contact information" (e.g., "head of household name," "patient's name," "student's parent's name"), and field values ​​often differ in their format, writing order, and symbol encoding. Furthermore, some systems record missing or incorrect information, such as missing ID numbers, duplicate phone number registrations, or unclear addresses, making traditional methods relying on single fields for identity verification prone to false or missed matches. Simultaneously, the lack of a unified primary key across systems hinders cross-platform data fusion centered on residents, impacting subsequent efforts to build unified identity profiles, deliver data-driven services accurately, and conduct traceable analysis of individual behavior.

[0004] Some existing practices rely on directly exposing sensitive identifiers (such as ID card numbers and mobile phone numbers) for comparison when attempting to establish identity mapping, which poses a significant risk of privacy leakage and cannot achieve encrypted matching or controllable traceability, making it difficult to meet the high standards required by future communities for identity security management and intelligent integrated governance.

[0005] Therefore, existing technologies cannot fully meet the urgent need to construct a unique and trustworthy "master identity profile" under complex real-world conditions such as multi-source heterogeneity, inconsistent fields, missing data, and ambiguous identities. Traditional field comparison, manual verification, and static primary key strategies are not only inefficient and poorly scalable in practical applications, but also lack a trusted verification mechanism, which can easily lead to information fragmentation, governance gaps, and even security risks.

[0006] Therefore, there is an urgent need for a method that can still achieve semantic understanding, intelligent alignment, and reliable construction of resident identity information in the context of heterogeneous multi-system structures, ambiguous information descriptions, and missing primary keys, so as to improve the overall capabilities and intelligence level of future communities in cross-system data fusion, accurate identity recognition, and digital governance support. Summary of the Invention

[0007] To address the aforementioned technical shortcomings, the purpose of this invention is to propose an intelligent construction method for the future community member architecture. This method aims to solve the technical problems in existing technologies, such as independent management of data from multiple departments, heterogeneous field structures, lack of unified identity primary keys, and especially the inability to achieve high-precision identity fusion and unified management under complex data conditions, such as residents having the same name, inconsistent address descriptions, missing fields, or unreliable sources.

[0008] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: The present invention provides a method for intelligently constructing a future community member structure.

[0009] The intelligent construction method for the future community member structure includes:

[0010] Step S10: Extract structured resident information dataset D from heterogeneous databases in the future community, and introduce domain field concept abstraction rule set. The format regularization function F is used, and a fusion mechanism based on semantic hash mapping and hierarchical field abstraction is employed to perform field normalization and format standardization operations on the structured resident information dataset D, generating a standardized dataset. ;

[0011] Step S20: Based on the standardized dataset Extract the unstructured data M of the corresponding residents, perform image-text joint recognition and semantic mapping enhancement on the unstructured data M, and compare it with a standardized dataset. An embedded field-trusted fusion mechanism is used to merge the data to obtain a fused dataset.

[0012] Step S30: Construct identity keyword combinations for each resident data record in the fused dataset, and perform encryption on the identity keyword combinations using a multi-factor hash compression and symmetric encryption nesting method to obtain the optimized fused dataset;

[0013] Step S40: Perform a cross-system comparison operation combining exact matching and fuzzy matching based on the optimized fusion dataset, and output a set of candidate matching pairs;

[0014] Step S50: Generate a unified master identity file set based on the candidate matching pair set using a version traceable evidence storage mechanism.

[0015] Preferably, in step S10, the structured resident information dataset D includes name information, birth information, address information, and contact information; and a fusion mechanism based on semantic hash mapping and hierarchical field abstraction is used to perform field normalization and format standardization operations on the structured resident information dataset D to generate a standardized dataset. The steps specifically include:

[0016] The original field names representing resident identity information from different heterogeneous databases are encoded using character vectorization, and the semantic hash value of the fields is calculated based on a preset semantic hash function. Multiple fields with semantic similarity not less than a set threshold ε are grouped into a unified hash slot.

[0017] Introducing Domain Field Concept Abstract Rule Set For fields belonging to the same hash slot, establish corresponding high-level semantic concept mapping relationships to form a ternary hierarchical abstract mapping table of fields, concepts and slots;

[0018] Introduce a format regular expression function F, and call at least one of the format regular expression functions F on the original field name to perform format conversion operations, including normalizing the telephone field to 11 digits, uniformly calling the address resolution interface to perform segmented encoding of the address field, and uniformly converting the date field to ISO-8601 format;

[0019] Replace all field names with the corresponding standard slot identifiers, and output a standardized dataset using a unified structured field model. .

[0020] Preferably, in step S20, the unstructured data M includes PDF form image files, table files, and scanned files; based on the standardized dataset... Extract the unstructured data M of the corresponding residents, perform image-text joint recognition and semantic mapping enhancement on the unstructured data M, and compare it with a standardized dataset. The steps for merging the dataset using an embedded field trusted fusion mechanism specifically include:

[0021] Optical character recognition (OCR) is performed on image-type forms in unstructured data M to extract the original text string T. The OCR operation is implemented using a joint network structure of a multi-channel convolutional neural network and a sequence label prediction model.

[0022] The extracted raw text string T is input into a pre-defined named entity recognition model to identify and label the resident's core information fields, which include name, surname, and date of birth; and based on a standardized dataset... The field slot structure establishes field mapping relationships;

[0023] Identify identical fields originating from different field slots, construct field confidence weights by introducing source confidence factors and field value confidence factors, and determine whether identical fields should be written into the fused dataset based on the field confidence weights;

[0024] Fields whose final identified confidence weights are greater than a set threshold will be written into the standardized dataset as embedded completion fields. The updated fused dataset D is generated, and the original extracted text and the confidence weights of the basis fields are retained as the basis for data changes.

[0025] Preferably, step S30, which involves constructing identity keyword combinations for each resident data record in the fused dataset and encrypting these identity keyword combinations using a multi-factor hash compression and symmetric encryption nested method to obtain the optimized fused dataset, specifically includes:

[0026] Extract each resident data record from the fused dataset to construct identity keyword combinations, and concatenate them in a preset order to form an identity keyword combination string K. Input the identity keyword combination string K into a hash function. Perform information compression to obtain an irreversible hash digest h; wherein, the hash function To achieve a cryptographic hash algorithm that is both one-way and collision-resistant;

[0027] Input the irreversible hash digest h into the preset symmetric encryption function. and using the key Nested encryption is performed to generate a ciphertext identity identifier e. The ciphertext identity identifier e is written into the fused dataset as a unique identity index for residents, and any unauthorized entity is prohibited from reversing the original identity keyword combination string K.

[0028] Preferably, step S40, which involves performing a cross-system comparison operation combining exact matching and fuzzy matching based on the optimized fusion dataset to output a set of candidate matching pairs, specifically includes:

[0029] First, perform exact matching based on the encrypted identity identifiers in the optimized fusion dataset to construct an initial set of candidate matching pairs. ;

[0030] For those not in the initial candidate matching pair set For successfully matched resident data record pairs, perform a fuzzy matching comparison operation based on field similarity function fusion to construct a supplementary candidate matching pair set. .

[0031] Preferably, step S40, the fuzzy matching comparison operation based on field similarity function fusion, specifically includes:

[0032] First construct the similarity function ,in, This represents the overall similarity score between the i-th record and the j-th record; This refers to the field value of the k-th field to be compared in the i-th record; This refers to the field value of the k-th field to be compared in the j-th record; Indicates the first A similarity measurement function for each field; Represents the weighting factor of the field, satisfying ;

[0033] Set field matching confidence threshold and field hit count lower limit When resident data records satisfy the similarity function And the number of field similarity matches When the resident record is successfully matched, it is determined that the resident record is a match and added to the candidate match set. ;

[0034] Output the final set of candidate matching pairs. This serves as input for subsequent primary identity profile merging steps and records the similarity function value for each candidate matching pair to support audit traceability and manual review.

[0035] Preferably, in step S50, the step of generating a unified master identity file set based on the candidate matching pair set using a version traceable evidence storage mechanism specifically includes: generating a unified master identity file for the candidate matching pair set, executing a field conflict resolution strategy, and recording the version number by writing the master index hash. and timestamp When a field is updated, the update differences and historical update history are recorded.

[0036] This invention also provides a future community member structure intelligent building system, comprising:

[0037] The field semantic standardization processing module is used to extract a structured resident information dataset D from heterogeneous databases in the future community and introduce a set of domain field concept abstraction rules. The format regularization function F is used, and a fusion mechanism based on semantic hash mapping and hierarchical field abstraction is employed to perform field normalization and format standardization operations on the structured resident information dataset D, generating a standardized dataset. ;

[0038] The unstructured fusion extraction module is used to extract data from standardized datasets. Extract the unstructured data M of the corresponding residents, perform image-text joint recognition and semantic mapping enhancement on the unstructured data M, and compare it with a standardized dataset. An embedded field-trusted fusion mechanism is used to merge the data to obtain a fused dataset.

[0039] The identity keyword encryption and compression module is used to construct identity keyword combinations for each resident data record in the fused dataset, and to perform encryption on the identity keyword combinations using multi-factor hash compression and symmetric encryption nesting method to obtain an optimized fused dataset.

[0040] The cross-system fuzzy comparison module is used to perform cross-system comparison operations that combine exact matching and fuzzy matching based on the optimized fusion dataset, and output a set of candidate matching pairs;

[0041] The master identity profile generation module is used to generate a unified master identity profile set based on the candidate matching pair set using a version traceable evidence storage mechanism.

[0042] The present invention also provides a future community member architecture intelligent building device, comprising: a memory, a processor, and a future community member architecture intelligent building program stored in the memory and executable on the processor. When the future community member architecture intelligent building program is executed by the processor, it implements a future community member architecture intelligent building method.

[0043] The present invention also provides a computer program product, including a future community member architecture intelligent building program, which, when executed by a processor, implements the future community member architecture intelligent building method.

[0044] The beneficial effects of this invention are as follows: Compared with the existing technology, which features independent management of data from multiple departments, heterogeneous field structures, and a lack of unified identity primary keys, especially under complex data conditions such as residents having the same name, inconsistent address descriptions, missing fields, or unreliable sources, making it impossible to achieve high-precision identity fusion and unified management, this application achieves cross-system multi-source data fusion and unique identity determination by constructing semantic hash slots, image and text joint recognition, multi-source fuzzy matching, and a trusted master file generation mechanism, thereby improving the uniqueness and reliability of future community resident identity data. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in 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.

[0046] Figure 1 This is a flowchart illustrating the first embodiment of a method for intelligently constructing a future community member structure according to the present invention.

[0047] Figure 2 This is a schematic diagram of a device for a future community member architecture intelligent construction method according to the present invention. Detailed Implementation

[0048] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0049] Example 1: As Figure 1 The diagram shown is a flowchart of the first embodiment of the intelligent construction method for the future community member architecture of the present invention, which presents the first embodiment of the intelligent construction method for the future community member architecture of the present invention.

[0050] In the first embodiment, the intelligent construction method for the future community member architecture includes:

[0051] Step S10: Extract structured resident information dataset D from heterogeneous databases in the future community, and introduce domain field concept abstraction rule set. The format regularization function F is used, and a fusion mechanism based on semantic hash mapping and hierarchical field abstraction is employed to perform field normalization and format standardization operations on the structured resident information dataset D, generating a standardized dataset. ;

[0052] It should be noted that the structured resident information dataset D in the heterogeneous database includes, but is not limited to, resident information tables in property management systems, household registration information tables in street-level government systems, student parent registration tables in education platforms, and health record tables in primary healthcare systems. Because the naming, format, and options for the same fields vary significantly across systems, direct comparison without a unified field abstraction mechanism can lead to mismatches, duplicate matches, or missing information. The domain field concept abstraction rule set is used to map multiple semantically consistent fields to a unified abstract concept. For example, "owner's name," "guardian's name," and "patient's name" can be uniformly abstracted to the "name" field slot; "building number," "house number," and "residential address" can be uniformly abstracted to the "address" field slot. This rule set can be constructed through a combination of expert definition and pre-trained language model generation to improve the accuracy and adaptability of field abstraction.

[0053] It should be understood that, compared with the existing technology that relies solely on manually configuring field mapping tables or matching fields by exact equality, this invention introduces a semantic hash mapping mechanism, uses models such as BERT and Word2Vec to embed and encode field names, and combines SimHash or MinHash to calculate semantic distance, thereby automatically identifying fields with different names but consistent semantics and classifying them into a unified hash slot, effectively improving the intelligence and automation level of field normalization.

[0054] For example, in a property management system, a record might have fields like "Homeowner's Name," "Contact Information," and "Address," while in an education platform, the same resident's fields might be "Parent's Name," "Phone Number," and "Home Address." This invention can use a semantic hash slot mapping mechanism to group "Homeowner's Name" and "Parent's Name" into the "Name" slot, and "Contact Information" and "Phone Number" into the "Phone Number" slot. Furthermore, it uses regular expressions to standardize phone numbers to 11 digits and addresses to "No. 78, XX Road," thereby generating records with consistent structure and standardized fields. This lays the foundation for subsequent identity aggregation and encrypted matching.

[0055] Step S20: Based on the standardized dataset Extract the unstructured data M of the corresponding residents, perform image-text joint recognition and semantic mapping enhancement on the unstructured data M, and compare it with a standardized dataset. An embedded field-trusted fusion mechanism is used to merge the data to obtain a fused dataset.

[0056] It should be noted that the unstructured data M mainly includes content from PDF household registration certificates, property registration photos, scanned copies of medical paper records, and uploaded images of school registration forms. This data exists in irregular forms such as images, text blocks, and form elements, and cannot be directly merged into a standardized structure using field mapping rules. Therefore, this step introduces image-text joint recognition (Multimodal OCR) and a semantic mapping enhancement mechanism to construct a field-level trusted fusion channel, enabling the structured extraction of unstructured field information and high-confidence semantic alignment with structured data D. The image-text joint recognition process includes not only extracting text content from images based on visual OCR models (such as PaddleOCR and LayoutLMv3), but also recognizing table structures and parsing field relationships. Simultaneously, semantic embedding models such as BERT are used to represent the recognition results as vectors, and embedding matching and semantic fusion are performed with known fields in the standardized data. This mechanism is significantly superior to traditional pure OCR solutions and can handle problems such as missing fields, misalignment, and non-template-based text.

[0057] It should be understood that, compared to existing field merging methods based on "direct coverage" or "priority trusted sources," this invention introduces an embedded trusted fusion mechanism when handling field conflicts. It dynamically evaluates field reliability using multi-factor fusion indicators such as field source confidence and semantic consistency before deciding on the final merged value. This approach improves the robustness and interpretability of field fusion, effectively reducing mis-aggregation and misidentification problems caused by field conflicts in the actual establishment of resident identity files.

[0058] For example, in the experimental sample, the standardized data from a street-level government affairs platform recorded a resident's address as "No. X, Jiangdong South Road," while the scanned copy of the resident's uploaded property certificate showed the address as "Unit X, Building X, Jiangdong South Road." Traditional field merging methods, due to the inconsistency of the fields, easily classify the two as different entities, leading to the risk of splitting them. This invention, however, identifies the nested address relationship after semantic enhancement and, combined with dimensions such as the credibility of the uploaded document and the degree of naming matching, successfully determines that they belong to the same address, achieving field merging and completion. In the experimental sample, the accuracy rate for processing 100 sets of address conflict samples reached 92.7%, significantly better than the 78.4% of the traditional template-based merging method.

[0059] Step S30: Construct identity keyword combinations for each resident data record in the fused dataset, and perform encryption on the identity keyword combinations using a multi-factor hash compression and symmetric encryption nesting method to obtain the optimized fused dataset;

[0060] It should be noted that even after integrating structured and unstructured sources, the fused dataset still suffers from issues such as exposed sensitive fields, data redundancy, and non-unique identity pointers. To improve data security and the stability of the fused identifiers, this step proposes a compression and encryption mechanism centered on "identity keyword combinations." Highly identifiable fields such as name, date of birth, ID number fragments, and address keywords are selected. By constructing a controllable-length compressed digest and employing symmetric encryption, the confidentiality and uniqueness of the data are ensured during transmission, comparison, and deduplication.

[0061] Understandably, unlike traditional methods that concatenate all fields of identity information before hashing, this method constructs a keyword set based on a "multi-factor dynamic combination" mechanism. Each set member is selected based on its confidence level, semantic stability, and privacy sensitivity. The resulting keyword combination no longer relies on the stability and consistency of all fields, but instead constitutes a highly redundancy-tolerant identity representation. This improvement can significantly reduce identity mismatch problems caused by changes in a certain field (such as address migration or contact information changes), thereby enhancing the persistent binding capability of resident identity files.

[0062] It should be understood that the "multi-factor hash compression and symmetric encryption nesting method" adopted in this invention adds a layer of structural nesting and access control compared to traditional single MD5 or SHA256 hash schemes. Its core idea is: first, construct a field vector digest based on the keyword set and perform two-level hash compression; then, combine it with an AES or SM4 symmetric encryption algorithm to generate a ciphertext identifier. During this process, digest length control and encryption key derivation are coupled and bound to the resident's source system, avoiding data replay and forgery attacks across systems. This structure, while maintaining computational efficiency, significantly enhances the feasibility of comparison and security protection capabilities after data anonymization.

[0063] For example, suppose a resident named "Wang Wu" has the keywords "Wang Wu," "1974-05," "3201****0012," and "No. 20, Fenghuang West Street" in the fused dataset. Based on field confidence levels, the first three items are selected to form a keyword combination. This combination is then hashed and compressed to form a 32-bit identity digest code "ae45c12f...". A symmetric key K is then derived from the community key, and the digest code is encrypted to generate a ciphertext identifier E. In subsequent inter-system comparisons, only the consistency of the encrypted digests needs to be compared to confirm whether it is the same resident, without exposing the original information. In a trial of 300 fused records, compared to plaintext field comparisons, this method maintained a comparison accuracy of over 98% while protecting privacy.

[0064] Step S40: Perform a cross-system comparison operation combining exact matching and fuzzy matching based on the optimized fusion dataset, and output a set of candidate matching pairs;

[0065] It should be noted that the "cross-system comparison operation combining exact matching and fuzzy matching" mentioned in this step refers to the following: for each resident identity data in the optimized and fused dataset, field consistency verification is first performed based on its encrypted digest or joint primary key field; if exact matching fails, fuzzy matching logic is activated, and fault-tolerant identification is performed based on the semantic similarity between information such as name, address, and contact information. This process further introduces field-level confidence assessment, field importance weighting, and time update weighting to improve the accuracy and stability of candidate matching pairs.

[0066] Understandably, this step enables the aggregation of resident identity records, originally scattered across property management systems, government platforms, educational data, and medical records, into credible identity pairs through field comparison and structured weighted analysis, thereby forming a set of candidate matching pairs. This method balances data integrity and fault tolerance, and even under conditions of differences in field expression, missing information, or delayed historical data updates across systems, it still maintains high recognition capabilities, enhancing the controllability and continuity of the data governance process.

[0067] It should be understood that, compared to the common "unique field primary key alignment" scheme in existing technologies, this step, through a combined comparison strategy, ensures the consistency of the encrypted digest while allowing fuzzy similarity calculations for misspellings in natural persons' names, synonymous expressions in addresses (such as "Donglu Yihao" and "1hao Donglu"), and missing segments in contact information. This makes the identity determination process more fault-tolerant and practical. It is particularly suitable for governance environments in future communities where population data is multi-sourced, heterogeneous, from untrusted sources, or in inconsistent formats.

[0068] For example, in a pilot "future community," the property management system records a resident's name as "Zhang Mouhong," with the address field "Building 4, Room 302, XX Community," and the contact number "1******34." However, in the education system, the record is "Zhang *hong," with the address "Building 4, Room 302, XX Community," but contact information is missing, only "Guardian: Ms. Zhang" is retained. In the medical system, the resident is registered using the English pinyin "Zhang XH," with the address field recorded as "4-302, XX Community," and the contact number "0-****998." In traditional comparison mechanisms, due to the differences in name spelling, address format, and contact information, the system often fails to recognize these data as belonging to the same resident, leading to duplicate records, file splitting, and duplicate service calls. In this step, effective integration is achieved through the following method:

[0069] Exact matching phase: When the primary key of the field (such as the encrypted document number) is missing or inconsistent, the process automatically enters the fuzzy matching process;

[0070] In the fuzzy matching stage: the name field is compared using both the character edit distance algorithm (edit distance of 2) and the semantic similarity model (Chinese name similarity score of approximately 0.88); the address field is decomposed using structured decomposition (building and house number are consistent) and then cosine similarity is calculated after BERT semantic encoding (similarity score of 0.91).

[0071] Weighted averaging strategy: The similarity scores of the name, address, and contact information fields are assigned weights of 0.4, 0.5, and 0.1, respectively. The fusion confidence score is obtained by weighted averaging. If the score is higher than the set threshold (e.g., 0.85), the individuals are considered to be the same resident.

[0072] Output: A unified master identity file is generated, and each source record and version information are retained for subsequent traceability.

[0073] In a multi-source identity sample containing 500 residents, the fusion method addressed typical issues such as field drift, name ambiguity, and missing contact information, achieving a candidate matching accuracy of 94.6% and a false merging rate of less than 1.3%. This is significantly better than the traditional strong consistency matching method based on unique primary keys (such as ID numbers) (with a success rate of only about 72.1%), verifying the applicability and robustness of the method in multi-source heterogeneous data environments.

[0074] Step S50: Generate a unified master identity file set based on the candidate matching pair set using a version traceable evidence storage mechanism.

[0075] It should be noted that the "version traceability and evidence preservation mechanism" in this step refers to creating a traceable version chain for each fusion result record during the generation of the master identity profile. This involves introducing data change signatures, timestamps, and source system identifier fields to record the fusion path and evolution of each identity field, ensuring that the identity data generation process is auditable, verifiable, and legally sound. This mechanism supports the arbitration, rollback, and responsibility allocation for identity conflicts in community governance.

[0076] Understandably, by generating and managing the master identity file in a structured manner, the problems of "information distortion," "unclear sovereignty," and "difficulty in tracing the source of identity reconstruction" that occurred after merging multiple data sources in the past have been solved. This ensures that when various business systems in the community (such as education, property management, government affairs, and healthcare) access resident identity data, they all point to the same dynamically evolving "master file" with a historical context, thereby improving the efficiency of data collaborative governance and ensuring the credible boundaries of data governance.

[0077] It should be understood that, compared to the traditional method of generating master files by simply using "overwriting field merging" or "last update time priority" mechanisms, this method records the entire process of field fusion through a chain version tracking mechanism. Every change to a field generates an immutable version snapshot, and trusted signatures prevent the risk of data pollution from "low-trust fields overwriting high-trust fields," thereby establishing a "trust anchor" for identity data governance in a multi-entity governance environment.

[0078] Example 2: Furthermore, the present invention provides an intelligent system for building a future community member architecture, employing an intelligent building method for a future community member architecture as described in the above embodiments, which can solve a technical problem related to intelligent building of a future community member architecture. Compared with the prior art, the beneficial effects of the intelligent building system for a future community member architecture provided by the present invention are the same as those of the intelligent building method for a future community member architecture provided in the above embodiments, and other technical features of the intelligent building system for a future community member architecture are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0079] Example 3: This invention provides a smart device for building a future community member structure. Please refer to... Figure 2A future community member architecture intelligent building device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, which are executed by the at least one processor to enable the at least one processor to execute the future community member architecture intelligent building method of Embodiment 1 described above. The future community member architecture intelligent building device in this embodiment may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. This future community member architecture intelligent building device is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this invention. A future community member architecture intelligent building device may include a processing device 1001, which can perform various appropriate actions and processes according to a program stored in a read-only memory 1002 or a program loaded from a storage device 1003 into a random access memory 1004. Random access memory 1004 also stores various programs and data required for the operation of a future community member architecture intelligent building device. Processing device 1001, read-only memory 1002, and random access memory 1004 are interconnected via bus 1005. I / O interface 1006 is also connected to the bus. Typically, the following systems can be connected to I / O interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1003; and communication devices 1009. Communication device 1009 allows a future community member architecture intelligent building device to communicate wirelessly or wiredly with other devices to exchange data. Although a future community member architecture intelligent building device with various systems is shown in the figure, it should be understood that it is not required to implement or possess all the systems shown. More or fewer systems may be implemented alternatively.

[0080] Example 4: This invention also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the intelligent construction method for a future community member architecture as described above. The computer program product provided by this invention can solve the technical problem of intelligent construction of a future community member architecture. Compared with the prior art, the beneficial effects of the computer program product provided by this invention are the same as the beneficial effects of the intelligent construction method for a future community member architecture provided in the above embodiments, and will not be repeated here.

[0081] In particular, according to the embodiments disclosed in this invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this invention.

[0082] It should be understood that the various parts disclosed in this invention can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics may be combined in any suitable manner in one or more embodiments or examples.

[0083] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for intelligently constructing a future community member structure, characterized in that, The methods include: Step S10: Extract structured resident information dataset D from heterogeneous databases in the future community, and introduce domain field concept abstraction rule set. The format regularization function F is used, and a fusion mechanism based on semantic hash mapping and hierarchical field abstraction is employed to perform field normalization and format standardization operations on the structured resident information dataset D, generating a standardized dataset. ; Step S20: Based on the standardized dataset Extract the unstructured data M of the corresponding residents, perform image-text joint recognition and semantic mapping enhancement on the unstructured data M, and compare it with a standardized dataset. An embedded field-trusted fusion mechanism is used to merge the data to obtain a fused dataset. Step S30: Construct identity keyword combinations for each resident data record in the fused dataset, and perform encryption on the identity keyword combinations using a multi-factor hash compression and symmetric encryption nesting method to obtain the optimized fused dataset; Step S40: Perform a cross-system comparison operation combining exact matching and fuzzy matching based on the optimized fusion dataset, and output a set of candidate matching pairs; Step S50: Generate a unified master identity file set based on the candidate matching pair set using a version traceable evidence storage mechanism.

2. The intelligent construction method for future community member architecture as described in claim 1, characterized in that, In step S10, the structured resident information dataset D includes name information, birth information, address information, and contact information. A fusion mechanism based on semantic hash mapping and hierarchical field abstraction is used to perform field normalization and format standardization operations on the structured resident information dataset D, generating a standardized dataset. The steps specifically include: The original field names representing resident identity information from different heterogeneous databases are encoded using character vectorization, and the semantic hash value of the fields is calculated based on a preset semantic hash function. Multiple fields with semantic similarity not less than a set threshold ε are grouped into a unified hash slot. Introducing Domain Field Concept Abstract Rule Set For fields belonging to the same hash slot, establish corresponding high-level semantic concept mapping relationships to form a ternary hierarchical abstract mapping table of fields, concepts and slots; Introduce a format regular expression function F, and call at least one of the format regular expression functions F on the original field name to perform format conversion operations, including normalizing the telephone field to 11 digits, uniformly calling the address resolution interface to perform segmented encoding of the address field, and uniformly converting the date field to ISO-8601 format; Replace all field names with the corresponding standard slot identifiers, and output a standardized dataset using a unified structured field model. .

3. The intelligent construction method for future community member architecture as described in claim 1, characterized in that, In step S20, the unstructured data M includes PDF form image files, table files, and scanned files; based on the standardized dataset... Extract the unstructured data M of the corresponding residents, perform image-text joint recognition and semantic mapping enhancement on the unstructured data M, and compare it with a standardized dataset. The steps for merging the dataset using an embedded field trusted fusion mechanism specifically include: Optical character recognition (OCR) is performed on image-type forms in unstructured data M to extract the original text string T. The OCR operation is implemented using a joint network structure of a multi-channel convolutional neural network and a sequence label prediction model. The extracted raw text string T is input into a pre-defined named entity recognition model to identify and label the resident's core information fields, which include name, surname, and date of birth; and based on a standardized dataset... The field slot structure establishes field mapping relationships; Identify identical fields originating from different field slots, construct field confidence weights by introducing source confidence factors and field value confidence factors, and determine whether identical fields should be written into the fused dataset based on the field confidence weights; Fields whose final identified confidence weights are greater than a set threshold will be written into the standardized dataset as embedded completion fields. The updated fused dataset D is generated, and the original extracted text and the confidence weights of the basis fields are retained as the basis for data changes.

4. The intelligent construction method for future community member architecture as described in claim 1, characterized in that, In step S30, an identity keyword combination is constructed for each resident data record in the fused dataset, and encryption is performed on the identity keyword combination using a multi-factor hash compression and symmetric encryption nested method to obtain the optimized fused dataset. This step specifically includes: Extract each resident data record from the fused dataset to construct identity keyword combinations, and concatenate them in a preset order to form an identity keyword combination string K. Input the identity keyword combination string K into a hash function. Perform information compression to obtain an irreversible hash digest h; wherein, the hash function To achieve a cryptographic hash algorithm that is both one-way and collision-resistant; Input the irreversible hash digest h into the preset symmetric encryption function. and using the key Nested encryption is performed to generate a ciphertext identity identifier e. The ciphertext identity identifier e is written into the fused dataset as a unique identity index for residents, and any unauthorized entity is prohibited from reversing the original identity keyword combination string K.

5. The intelligent construction method for the future community member structure as described in claim 1, characterized in that, Step S40, which involves performing a cross-system comparison operation combining exact matching and fuzzy matching based on the optimized fusion dataset and outputting a set of candidate matching pairs, specifically includes: First, perform exact matching based on the encrypted identity identifiers in the optimized fusion dataset to construct an initial set of candidate matching pairs. ; For those not in the initial candidate matching pair set For successfully matched resident data record pairs, perform a fuzzy matching comparison operation based on field similarity function fusion to construct a supplementary candidate matching pair set. .

6. The intelligent construction method for the future community member structure as described in claim 5, characterized in that, Step S40, the fuzzy matching comparison operation based on field similarity function fusion, specifically includes: First construct the similarity function ,in, This represents the overall similarity score between the i-th record and the j-th record; This refers to the field value of the k-th field to be compared in the i-th record; This refers to the field value of the k-th field to be compared in the j-th record; Indicates the first A similarity measurement function for each field; Represents the weighting factor of the field, satisfying ; Set field matching confidence threshold and field hit count lower limit When resident data records satisfy the similarity function And the number of field similarity matches When the resident record is successfully matched, it is determined that the resident record is a match and added to the candidate match set. ; Output the final set of candidate matching pairs. This serves as input for subsequent primary identity profile merging steps and records the similarity function value for each candidate matching pair to support audit traceability and manual review.

7. The intelligent construction method for future community member architecture as described in claim 1, characterized in that, Step S50, which involves generating a unified master identity file set based on the candidate matching pair set using a version-traceable evidence storage mechanism, specifically includes: generating a unified master identity file for the candidate matching pair set, executing a field conflict resolution strategy, and recording the version number by writing the master index hash. and timestamp When a field is updated, the update differences and historical update history are recorded.

8. A future community member architecture intelligent construction system, applied to the future community member architecture intelligent construction method according to any one of claims 1 to 7, characterized in that, The intelligent system for building the future community member structure includes: The field semantic standardization processing module is used to extract a structured resident information dataset D from heterogeneous databases in the future community and introduce a set of domain field concept abstraction rules. The format regularization function F is used, and a fusion mechanism based on semantic hash mapping and hierarchical field abstraction is employed to perform field normalization and format standardization operations on the structured resident information dataset D, generating a standardized dataset. ; The unstructured fusion extraction module is used to extract data from standardized datasets. Extract the unstructured data M of the corresponding residents, perform image-text joint recognition and semantic mapping enhancement on the unstructured data M, and compare it with a standardized dataset. An embedded field-trusted fusion mechanism is used to merge the data to obtain a fused dataset. The identity keyword encryption and compression module is used to construct identity keyword combinations for each resident data record in the fused dataset, and to perform encryption on the identity keyword combinations using multi-factor hash compression and symmetric encryption nesting method to obtain an optimized fused dataset. The cross-system fuzzy comparison module is used to perform cross-system comparison operations that combine exact matching and fuzzy matching based on the optimized fusion dataset, and output a set of candidate matching pairs; The master identity profile generation module is used to generate a unified master identity profile set based on the candidate matching pair set using a version traceable evidence storage mechanism.

9. A smart device for building a future community member structure, characterized in that, The intelligent building device for the future community member architecture includes: a memory, a processor, and an intelligent building program for the future community member architecture stored in the memory and capable of running on the processor. When the intelligent building program for the future community member architecture is executed by the processor, it implements an intelligent building method for the future community member architecture as described in any one of claims 1 to 7.

10. A computer program product, characterized in that, The computer program product includes a future community member architecture intelligent building program, which, when executed by a processor, implements a future community member architecture intelligent building method according to any one of claims 1 to 7.

Images