Decentralized talent background survey data storage and verification method and system

Abstract

The invention provides a decentralized talent background survey data storage and verification method and system, and relates to the technical field of background survey, and the method comprises the steps: encrypting education, work and skill information submitted by candidates through a national cryptographic algorithm SM2, and generating a data packet with a time sequence weight in combination with an SM3 hash algorithm; constructing a hierarchical time sequence dependency graph by using deep learning to extract associated features; consistency verification of zero-knowledge proof is realized based on the feature fingerprint identification; and a double-layer consensus intelligent contract is adopted for data updating and tracing. According to the method, safe storage, efficient verification and credible tracing of talent background information are realized, data tampering is effectively prevented, and the reliability of background investigation is improved.

Description

Technical Field

[0001] The present invention relates to background investigation technology, and in particular to a decentralized talent background investigation data storage and verification method and system. Background Art

[0002] With the rapid development of the talent market, companies need to conduct a comprehensive verification of candidates' educational background, work experience and professional skills during the recruitment process. Traditional talent background checks mainly rely on manual verification or third-party investigation agencies, which is inefficient and costly. In recent years, with the development of blockchain technology, decentralized data storage and verification solutions have gradually been applied to the field of talent background checks. The current blockchain talent background check system usually uses distributed ledger technology to record candidate information, implements data verification through smart contracts, and uses encryption algorithms to protect data security.

[0003] The existing blockchain talent background investigation system lacks effective management of data temporality and cannot accurately reflect the temporal continuity of candidates’ education and work experience. It is prone to information gaps or time sequence confusion, which affects the accuracy of background investigations.

[0004] Traditional blockchain data storage solutions lack a flexible processing mechanism for data updates. When the stored information needs to be updated or corrected, it is often necessary to rebuild the entire data structure. This method is not only inefficient, but may also lead to confusion in data version management.

[0005] Existing verification mechanisms generally adopt simple consensus algorithms, lack dynamic evaluation of the trust of verification nodes, and are unable to effectively prevent verification behavior by malicious nodes. They also lack a complete data traceability mechanism, making it difficult to ensure the credibility and traceability of verification results. Summary of the invention

[0006] The embodiments of the present invention provide a decentralized talent background investigation data storage and verification method and system, which can solve the problems in the prior art.

[0007] According to a first aspect of the embodiments of the present invention, Provide decentralized talent background check data storage and verification methods, including: Based on the identity authentication layer, the candidate's education experience information, work experience information and skill certification information are received, and the received information is asymmetrically encrypted using the national encryption algorithm SM2 to obtain a first encrypted data set. The first encrypted data set is combined with the SM3 hash algorithm to generate an original data packet with a time series weight; The first encrypted data set is segmented according to the timing weight in the original data packet, and a hierarchical timing dependency graph is constructed. The hierarchical timing dependency graph extracts the timing correlation features between each data segment through a deep learning model to generate a second encrypted data set. The second encrypted data set includes a weighted feature fingerprint identifier and an information segment hash value, and the second encrypted data set is written into the trusted node group of the blockchain network through a three-layer Merkle tree structure; When the trusted node group receives a verification request, it locates the target data location based on the feature fingerprint identifier, extracts the second encrypted data set of the target time period according to the hierarchical temporal dependency graph, and calculates the node real-time trust score at the same time. The node real-time trust score is dynamically updated based on the node's verification history behavior feature vector. When the overall trust of the node group reaches the adaptive threshold, the feature fingerprint identifier is used to start the consistency verification mechanism based on zero-knowledge proof, and the consistency verification mechanism generates a tamper-proof verification certificate containing the temporal integrity proof for the verification result; When the hierarchical timing dependency graph detects a need to update data, the correlation between the update request information and the timing weight in the original data packet is calculated, and the deep learning model evaluates the rationality of the update. After confirmation, the data segment to be updated is located in the second encrypted data set, and the two-layer consensus smart contract is activated to establish a version control chain. Based on the feature fingerprint identification, a traceability network is built in the trusted node group. The updated data is merged with the original timing dependency graph through a three-layer Merkel tree structure to generate a new verification link, thereby achieving traceability of the entire data update process.

[0008] The received information is asymmetrically encrypted by the national encryption algorithm SM2 to obtain a first encrypted data set, and the original data packet with a timing weight is generated by combining the SM3 hash algorithm for the first encrypted data set, including: The received education experience, work experience and skill certification information is asymmetrically encrypted using the national secret algorithm SM2, a base point G and order n are selected based on the SM2 elliptic curve cryptography system to generate a key pair (d, Q), the public key Q is used to perform encryption operations on the standardized information, and a first encrypted data set is generated in combination with a random number k, the first encrypted data set contains encrypted structured information; The first encrypted data set is securely processed in combination with the SM3 hash algorithm. The encrypted data is padded with messages in multiples of 512 bits, and iterative operations are performed through the compression function. At the same time, a timing weight parameter is introduced to calculate the timestamp difference and information correlation between adjacent data blocks, and an original data packet with a timing weight is generated. The original data packet has a timing integrity feature.

[0009] The received education experience, work experience and skill certification information is asymmetrically encrypted using the national secret algorithm SM2, a key pair (d, Q) is generated by selecting the base point G and the order n based on the SM2 elliptic curve cryptography system, and the public key Q is used to perform encryption operations on the standardized information, and the first encrypted data set is generated in combination with the random number k, including: Receive the education experience information, work experience information and skill certification information submitted by the user, classify and encode the education experience information, work experience information and skill certification information respectively, obtain education experience structured data, work experience structured data and skill certification structured data, and integrate the education experience structured data, work experience structured data and skill certification structured data to construct a standardized data template; In the SM2 elliptic curve cryptosystem, a base point G and an order n are selected, a private key d is obtained using a random number generator, a public key Q is generated based on a point multiplication operation between the private key d and the base point G, and a key pair (d, Q) is obtained. The key validity verification algorithm is used to verify the validity of the key pair (d, Q); Generate a random number k, perform a point multiplication operation on the random number k and the base point G to obtain a temporary public key, perform a point multiplication operation on the temporary public key and the public key Q to obtain a session key, use a key derivation function to process the session key to generate an encryption key, use the encryption key to encrypt the standardized data template, and generate a first encrypted data set containing encrypted data and its corresponding timestamp.

[0010] The second encrypted data set of the target time period is extracted according to the hierarchical time dependency graph, and the real-time trust score of the node is calculated at the same time. The real-time trust score of the node is dynamically updated based on the feature vector of the node's verification history behavior. When the overall trust of the node group reaches the adaptive threshold, the feature fingerprint identification is used to start the consistency verification mechanism based on zero-knowledge proof. Based on the consistency verification mechanism, the verification result generates a tamper-proof verification certificate containing the time integrity proof, including: The hierarchical timing dependency graph includes a timing node set and a timing edge set. A second encrypted data set of a target time period is extracted according to the hierarchical timing dependency graph. The second encrypted data set includes encrypted data in the target time period and its corresponding timing dependency relationship. A verification history behavior feature vector is established for the nodes in the timing node set. The node real-time trust score is calculated based on the verification history behavior feature vector. The node real-time trust score is dynamically adjusted as the verification history behavior feature vector is updated. The real-time trust scores of each node in the time series node set are weighted and summed to obtain the overall trust of the node group. The overall trust of the node group is compared with the adaptive threshold. When the overall trust of the node group is greater than the adaptive threshold, a feature fingerprint is generated based on the overall trust of the node group and the adaptive threshold. A prover commitment is constructed based on the characteristic fingerprint identifier, and the prover commitment is combined with the verification rule to generate a zero-knowledge proof. The zero-knowledge proof is used to start the consistency verification mechanism, and the second encrypted data set is checked for consistency and a verification result is generated. A Merkle tree is constructed based on the verification result, and a temporal integrity proof including the Merkle tree root node and the verification path is generated. A tamper-proof verification certificate is generated based on the temporal integrity proof.

[0011] A Merkle tree is constructed based on the verification results, and a temporal integrity proof including the Merkle tree root node and the verification path is generated. The tamper-proof verification certificate is generated based on the temporal integrity proof, including: Perform hash calculation on each data block in the verification result and the corresponding timestamp sequence to obtain the data block hash value, combine the data block hash value with the index information to generate a leaf node; perform pairing calculation on the leaf nodes to obtain the intermediate node hash value, recursively pair the intermediate node hash values ​​until the root node hash value is generated, and construct a Merkle tree structure based on the root node hash value; Extract the verification path node set and path direction sequence in the Merkle tree structure, combine the verification path node set and path direction sequence to generate a verification path, and combine the verification path with the root node hash value to generate a temporal integrity proof; A tamper-proof verification certificate is constructed based on the temporal integrity proof. The tamper-proof verification certificate includes the certificate header information and the certificate body content. The certificate header information and the certificate body content of the tamper-proof verification certificate are hashed to obtain a certificate summary, and the certificate summary is signed with a private key to obtain a digital signature. The digital signature is added to the tamper-proof verification certificate to complete the certificate generation.

[0012] Activate the two-layer consensus smart contract to establish a version control chain, build a traceability network in the trusted node group based on feature fingerprint identification, merge the updated data with the original time-series dependency graph through the three-layer Merkle tree structure, generate a new verification link, and achieve traceability of the entire data update process, including: Construct the main consensus layer contract and the verification consensus layer contract. The main consensus layer contract records data version control information, and the verification consensus layer contract records data update verification information. Activate the main consensus layer contract and the verification consensus layer contract to form a double-layer consensus smart contract and establish a version control chain. Perform hash calculation on the data block to obtain the data block hash value, combine the data block hash value with the timestamp sequence to generate a feature fingerprint identifier, calculate the fingerprint similarity between nodes based on the feature fingerprint identifier, and select trusted nodes based on the fingerprint similarity to build a traceability network. Construct a three-layer Merkel tree structure in the traceability network, write the updated data information into the data layer of the three-layer Merkel tree structure to generate the terminal node, calculate the temporal dependency relationship based on the terminal node to generate the intermediate node and write it into the dependency layer of the three-layer Merkel tree structure; calculate the verification information of the intermediate node to obtain the root node and write it into the verification layer of the three-layer Merkel tree structure, and establish a data update path based on the terminal node, intermediate node and root node in the three-layer Merkel tree structure; The updated data is merged with the original time-series dependency graph through the data update path to generate a new verification link containing node verification records. The new verification link records the complete process of data update. When receiving a data update request, the main consensus layer contract triggers the verification consensus layer contract to perform verification. After the verification passes, the update information is recorded in the version control chain, and the traceability of the entire data update process is achieved based on the new verification link.

[0013] Perform hash calculation on the data block to obtain the data block hash value, combine the data block hash value with the timestamp sequence to generate a feature fingerprint identifier, calculate the fingerprint similarity between nodes based on the feature fingerprint identifier, and select trusted nodes based on the fingerprint similarity to build a traceability network, including: Perform hash calculation on the input data block to obtain the main hash value of the data block, perform MD5 and SHA1 hash calculation on the data block to obtain the auxiliary hash sequence of the data block, combine the main hash value of the data block with the auxiliary hash sequence of the data block and add a random salt value to generate the hash value of the data block; collect the timestamp information of the data block at different processing nodes, arrange the timestamp information according to the time sequence relationship to generate a timestamp sequence, perform encoding processing on the timestamp sequence to obtain a time feature vector, and perform weighted combination operation on the data block hash value and the time feature vector to generate a feature fingerprint identifier; Establish a node feature matrix, the row vector of the node feature matrix represents the node identity of the feature fingerprint, and the column vector represents the different feature dimensions of the feature fingerprint. The distance between different row vectors in the node feature matrix is ​​calculated based on the local sensitive hashing algorithm to obtain the initial similarity; the initial similarity is weightedly fused with the node's historical behavior reputation to obtain the fingerprint similarity between nodes. The historical behavior reputation is calculated based on the node's historical interaction records and verification pass rate; Set the dynamic node similarity threshold, which is adaptively adjusted according to the network scale and security level. Compare the fingerprint similarity with the dynamic node similarity threshold, and select the nodes with fingerprint similarity higher than the dynamic node similarity threshold as trusted nodes. Calculate the trust transfer strength between trusted nodes, which is determined by the direct similarity and indirect similarity between nodes. Build the weighted connection relationship between nodes based on the trust transfer strength. A hierarchical node traceability network is constructed using weighted connection relationships. The hierarchical node traceability network includes a core trusted layer and an extended verification layer. The core trusted layer is composed of nodes with the highest fingerprint similarity, and the extended verification layer is composed of other trusted nodes.

[0014] According to a second aspect of the embodiments of the present invention, An electronic device is provided, comprising: processor; a memory for storing processor-executable instructions; The processor is configured to call instructions stored in the memory to execute the aforementioned method.

[0015] According to a third aspect of the embodiments of the present invention, A computer-readable storage medium is provided, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the aforementioned method is implemented.

[0016] The beneficial effects of this application are as follows: 1. Use the national secret algorithm SM2 for asymmetric encryption and the SM3 hash algorithm to generate original data packets with timing weights, combine the deep learning model to build a hierarchical timing dependency graph, and use a three-layer Merkel tree structure to write data into the blockchain network, thereby achieving secure storage and efficient management of talent background information and effectively preventing data from being tampered with or forged.

[0017] 2. Based on feature fingerprint identification and node real-time trust scoring mechanism, combined with the consistency check of zero-knowledge proof, a tamper-proof verification certificate containing temporal integrity proof is generated, which improves the credibility and verification efficiency of background investigation data and ensures the accuracy and reliability of verification results.

[0018] 3. The rationality of data updates is evaluated through deep learning models, a two-layer consensus smart contract is used to establish a version control chain, and a traceability network is built based on feature fingerprint identification, which realizes the traceability of the entire data update process, ensuring the timeliness of the data and maintaining the transparency and credibility of information updates. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A flowchart of a decentralized talent background investigation data storage and verification method according to an embodiment of the present invention; Figure 2 This is a schematic diagram of verifying accuracy evaluation in an embodiment of the present invention; Figure 3 A schematic diagram showing a comparison between hash calculation performance and feature fingerprint generation efficiency according to an embodiment of the present invention; Figure 4 This is a schematic diagram of node trust evaluation according to an embodiment of the present invention; Figure 5This is a schematic diagram of a timestamp sequence information panel according to an embodiment of the present invention. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] The technical solution of the present invention is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0022] Figure 1 The following is a flowchart of a decentralized talent background investigation data storage and verification method according to an embodiment of the present invention. Figure 1 As shown, the method includes: Based on the identity authentication layer, the candidate's education experience information, work experience information and skill certification information are received, and the received information is asymmetrically encrypted using the national encryption algorithm SM2 to obtain a first encrypted data set. The first encrypted data set is combined with the SM3 hash algorithm to generate an original data packet with a time series weight; The first encrypted data set is segmented according to the timing weight in the original data packet, and a hierarchical timing dependency graph is constructed. The hierarchical timing dependency graph extracts the timing correlation features between each data segment through a deep learning model to generate a second encrypted data set. The second encrypted data set includes a weighted feature fingerprint identifier and an information segment hash value, and the second encrypted data set is written into the trusted node group of the blockchain network through a three-layer Merkle tree structure; When the trusted node group receives a verification request, it locates the target data location based on the feature fingerprint identifier, extracts the second encrypted data set of the target time period according to the hierarchical temporal dependency graph, and calculates the node real-time trust score at the same time. The node real-time trust score is dynamically updated based on the node's verification history behavior feature vector. When the overall trust of the node group reaches the adaptive threshold, the feature fingerprint identifier is used to start the consistency verification mechanism based on zero-knowledge proof, and the consistency verification mechanism generates a tamper-proof verification certificate containing the temporal integrity proof for the verification result; When the hierarchical timing dependency graph detects a need to update data, the correlation between the update request information and the timing weight in the original data packet is calculated, and the deep learning model evaluates the rationality of the update. After confirmation, the data segment to be updated is located in the second encrypted data set, and the two-layer consensus smart contract is activated to establish a version control chain. Based on the feature fingerprint identification, a traceability network is built in the trusted node group. The updated data is merged with the original timing dependency graph through a three-layer Merkel tree structure to generate a new verification link, thereby achieving traceability of the entire data update process.

[0023] In an optional implementation, asymmetric encryption is performed on the received information by using the national encryption algorithm SM2 to obtain a first encrypted data set, and the original data packet with a timing weight is generated for the first encrypted data set in combination with the SM3 hash algorithm, including: The received education experience, work experience and skill certification information is asymmetrically encrypted using the national secret algorithm SM2, a base point G and order n are selected based on the SM2 elliptic curve cryptography system to generate a key pair (d, Q), the public key Q is used to perform encryption operations on the standardized information, and a first encrypted data set is generated in combination with a random number k, the first encrypted data set contains encrypted structured information; The first encrypted data set is securely processed in combination with the SM3 hash algorithm. The encrypted data is padded with messages in multiples of 512 bits, and iterative operations are performed through the compression function. At the same time, a timing weight parameter is introduced to calculate the timestamp difference and information correlation between adjacent data blocks, and an original data packet with a timing weight is generated. The original data packet has a timing integrity feature.

[0024] First, the received education experience, work experience and skill certification information is standardized and preprocessed. The unstructured text information is converted into structured data in a unified format, including key fields such as time, location, institution, major, and position. Data cleaning and format normalization are performed on each field to generate a standardized information data set.

[0025] Then, the key is generated based on the SM2 elliptic curve cryptography. Select the elliptic curve parameters, determine the base point G and the order n. Randomly generate the private key d and calculate the public key Q. Generate a key pair with 256 bits as the security strength. For example, select the elliptic curve parameters as the recommended curve sm2p256v1 to generate a 256-bit private key and a 512-bit public key.

[0026] Then, the standardized information is encrypted using the SM2 public key encryption algorithm. The information is grouped into fixed-length data blocks, each of which is 128 bytes. For each data block, a random number k is generated, and encryption calculation is performed in combination with the public key Q to obtain a ciphertext point. All ciphertext points are combined to form the first encrypted data set. For example, after encrypting an educational experience information "2018-2022 Computer Science Peking University", a 512-byte ciphertext is obtained.

[0027] The first encrypted data set is then processed using the SM3 hash algorithm. First, message padding is performed to pad the data length to an integer multiple of 512 bits. The compression function is then iterated, processing 512-bit data blocks each time. At the same time, a timing weight is introduced to calculate the timestamp difference between adjacent data blocks. For example, if the time difference between two educational experiences is 4 years, the weight is set to 0.8. Finally, a 256-bit hash value is generated.

[0028] Finally, the hash value is combined with the timing weight to generate an original data packet with timing characteristics. The data packet contains information such as encrypted data, hash value, timing weight, etc. The timing integrity verification of data is achieved through timing weight. For example, the generated data packet size is 1024 bytes, containing 512 bytes of ciphertext, 256-bit hash value and 64-bit timing weight.

[0029] SM2 asymmetric encryption is used to achieve secure data storage, protect user privacy information from being leaked, and improve the security of data transmission and storage. Combined with the SM3 hash algorithm, data fingerprints are generated to ensure data integrity, prevent data tampering, and enhance data credibility and anti-counterfeiting capabilities. The timing weight mechanism is introduced to verify the timing integrity of data, effectively prevent data replay attacks, and ensure the timing correctness and consistency of data.

[0030] In an optional implementation, the received education experience, work experience and skill certification information is asymmetrically encrypted using the national secret algorithm SM2, a base point G and an order n are selected based on the SM2 elliptic curve cryptography system to generate a key pair (d, Q), and the public key Q is used to perform encryption operations on the standardized information, and the first encrypted data set is generated in combination with the random number k, including: Receive the education experience information, work experience information and skill certification information submitted by the user, classify and encode the education experience information, work experience information and skill certification information respectively, obtain education experience structured data, work experience structured data and skill certification structured data, and integrate the education experience structured data, work experience structured data and skill certification structured data to construct a standardized data template; In the SM2 elliptic curve cryptosystem, a base point G and an order n are selected, a private key d is obtained using a random number generator, a public key Q is generated based on a point multiplication operation between the private key d and the base point G, and a key pair (d, Q) is obtained. The key validity verification algorithm is used to verify the validity of the key pair (d, Q); Generate a random number k, perform a point multiplication operation on the random number k and the base point G to obtain a temporary public key, perform a point multiplication operation on the temporary public key and the public key Q to obtain a session key, use a key derivation function to process the session key to generate an encryption key, use the encryption key to encrypt the standardized data template, and generate a first encrypted data set containing encrypted data and its corresponding timestamp.

[0031] First, receive the education experience information, work experience information and skill certification information submitted by the user. Classify and encode the education experience information, including fields such as academic level, major category, and graduation school, and generate education experience structured data. Classify and encode the work experience information, including fields such as work unit, position, and years of work, and generate work experience structured data. Classify and encode the skill certification information, including fields such as certificate type, issuing agency, and time of acquisition, and generate skill certification structured data. Integrate the three types of structured data into a standardized data template.

[0032] In the SM2 elliptic curve cryptography system, the coordinates of the base point G are selected as (x, y), and the order n is a 256-bit prime number. A 256-bit random number is generated by a random number generator as the private key d. The private key d is multiplied by the base point G to obtain the coordinates of the public key Q. The generated key pair (d, Q) is validated to verify whether the private key d is in the interval [1, n-1] and whether the public key Q satisfies the elliptic curve equation.

[0033] Generate a random number k, multiply k by the base point G to get a temporary public key (x1, y1). Multiply the temporary public key by the public key Q to get a session key (x2, y2). Use the key derivation function KDF to process the session key coordinate value to generate an encryption key. Use the encryption key to encrypt the standardized data template to obtain ciphertext data. Add a timestamp to the ciphertext data to generate the first encrypted data set.

[0034] For example, the user submits undergraduate education information, including "Computer Science and Technology Major, a certain university, graduated in 2020" and other content. Work experience includes "Software Engineer, a certain company, 3 years of work experience" and other content. Skill certification includes "Senior Programmer Certificate, a certain certification body, obtained in 2022" and other content. After structured processing and encryption, ciphertext data is generated and the timestamp of "2023-06-01 10:00:00" is added.

[0035] This embodiment provides a data encryption method based on the national secret algorithm SM2. In view of the fact that the prior art generally uses a simple symmetric encryption method to process the personal information submitted by users, there are problems such as complex key management and low security, as well as the defects of inconsistent data structure leading to difficulties in subsequent processing.

[0036] This embodiment first classifies and codes the education experience, work experience and skill certification information submitted by the user and structures it to establish a standardized data template, thus solving the problem of inconsistent original data formats and difficulty in unified management and analysis. By establishing a unified data structure template, the standardization and processability of the data are improved.

[0037] Secondly, this embodiment uses the national secret algorithm SM2 for asymmetric encryption, which has higher security than the commonly used symmetric encryption method in the prior art. The key pair is generated through the SM2 elliptic curve cryptography system, random numbers are introduced to increase the encryption strength, point multiplication operations are used to generate session keys, and key derivation functions are combined for key processing, forming a multi-level security protection mechanism. This solution avoids the difficulties of key distribution and management in symmetric encryption, while ensuring the security of the encryption process.

[0038] Again, this embodiment adds timestamp information on the basis of encrypted data to facilitate tracking the timeliness and version management of data. This design not only ensures the traceability of data, but also facilitates the system to maintain and update data.

[0039] Through the above technical solution, this embodiment realizes the standardized processing and secure encrypted storage of user personal information, improves data management efficiency, enhances information security protection capabilities, and provides a good foundation for subsequent data analysis and application. While ensuring data security, this solution also takes into account the practicality and scalability of the system, and has good application value.

[0040] In an optional implementation, a second encrypted data set of a target time period is extracted according to a hierarchical time dependency graph, and a real-time trust score of a node is calculated at the same time. The real-time trust score of a node is dynamically updated based on a feature vector of a node's verification history behavior. When the overall trust of the node group reaches an adaptive threshold, a consistency verification mechanism based on zero-knowledge proof is started using a feature fingerprint identifier. The verification result is generated based on the consistency verification mechanism to generate a tamper-proof verification certificate containing a proof of time integrity, including: The hierarchical timing dependency graph includes a timing node set and a timing edge set. A second encrypted data set of a target time period is extracted according to the hierarchical timing dependency graph. The second encrypted data set includes encrypted data in the target time period and its corresponding timing dependency relationship. A verification history behavior feature vector is established for the nodes in the timing node set. The node real-time trust score is calculated based on the verification history behavior feature vector. The node real-time trust score is dynamically adjusted as the verification history behavior feature vector is updated. The real-time trust scores of each node in the time series node set are weighted and summed to obtain the overall trust of the node group. The overall trust of the node group is compared with the adaptive threshold. When the overall trust of the node group is greater than the adaptive threshold, a feature fingerprint is generated based on the overall trust of the node group and the adaptive threshold. A prover commitment is constructed based on the characteristic fingerprint identifier, and the prover commitment is combined with the verification rule to generate a zero-knowledge proof. The zero-knowledge proof is used to start the consistency verification mechanism, and the second encrypted data set is checked for consistency and a verification result is generated. A Merkle tree is constructed based on the verification result, and a temporal integrity proof including the Merkle tree root node and the verification path is generated. A tamper-proof verification certificate is generated based on the temporal integrity proof.

[0041] The hierarchical temporal dependency graph contains a set of temporal nodes and a set of temporal edges. The temporal nodes represent the time points of data generation, and the temporal edges represent the dependencies between data. First, the hierarchical temporal dependency graph is traversed according to the target time period, and the encrypted data and its corresponding temporal dependencies within the time period are extracted to form a second encrypted data set. Specifically, for each temporal node within the time period, its encrypted data and the dependency information represented by the temporal edges connected to it are obtained.

[0042] A verification history behavior feature vector is established for each time series node. The feature vector contains multi-dimensional features such as node verification accuracy, verification timeliness, and verification consistency. The real-time trust score of the node is calculated based on the feature vector, which is obtained by weighted summation of each dimension of the feature vector. As the node participates in the verification process, the feature vector is dynamically updated to adjust the trust score in real time. For example, if the verification accuracy of a node is 0.95, the verification timeliness is 0.9, and the verification consistency is 0.85, and the weights of each dimension are 0.4, 0.3, and 0.3 respectively, then the real-time trust score of the node is 0.904.

[0043] The real-time trust scores of each node in the time series node set are weighted and summed to obtain the overall trust of the node group. The adaptive threshold is set to 0.85. When the overall trust of the node group is greater than 0.85, a feature fingerprint is generated based on the difference between the two. The feature fingerprint uses the SHA256 hash algorithm, and the input is the difference between the overall trust of the node group and the adaptive threshold.

[0044] Based on the characteristic fingerprint identification, the prover commitment is constructed, and the Pedersen commitment scheme is adopted. The commitment is combined with the preset verification rules to generate a zero-knowledge proof. The consistency verification mechanism is started to verify the data and its dependencies in the second encrypted data set and generate a verification result. The verification result is constructed as a Merkle tree, and a temporal integrity proof containing the root node hash value and the verification path is generated. Finally, a tamper-proof verification certificate is generated based on the temporal integrity proof, and the certificate uses digital signature technology for tamper-proof protection.

[0045] The encrypted data and dependencies of the target time period are extracted through the hierarchical temporal dependency graph, and the trust score is dynamically calculated based on the node verification historical behavior feature vector, realizing the credibility assessment of the data verification process. The overall trust of the node group and the adaptive threshold comparison mechanism are adopted, combined with feature fingerprint identification and zero-knowledge proof to ensure the security and privacy of the verification process. The temporal integrity proof is constructed through the Merkle tree, and the tamper-proof verification certificate is generated to ensure the non-tamperability and traceability of the verification results, and improve the reliability of the entire verification mechanism.

[0046] In an optional implementation, constructing a Merkle tree according to the verification result, generating a temporal integrity proof including a Merkle tree root node and a verification path, and generating a tamper-proof verification certificate based on the temporal integrity proof includes: Perform hash calculation on each data block in the verification result and the corresponding timestamp sequence to obtain the data block hash value, combine the data block hash value with the index information to generate a leaf node; perform pairing calculation on the leaf nodes to obtain the intermediate node hash value, recursively pair the intermediate node hash values ​​until the root node hash value is generated, and construct a Merkle tree structure based on the root node hash value; Extract the verification path node set and path direction sequence in the Merkle tree structure, combine the verification path node set and path direction sequence to generate a verification path, and combine the verification path with the root node hash value to generate a temporal integrity proof; A tamper-proof verification certificate is constructed based on the temporal integrity proof. The tamper-proof verification certificate includes the certificate header information and the certificate body content. The certificate header information and the certificate body content of the tamper-proof verification certificate are hashed to obtain a certificate summary, and the certificate summary is signed with a private key to obtain a digital signature. The digital signature is added to the tamper-proof verification certificate to complete the certificate generation.

[0047] First, the data blocks and timestamp sequences in the verification results are processed. Take a set of verification results containing 10 data blocks as an example. Each data block contains transaction information and a corresponding timestamp. The SHA-256 hash algorithm is used to calculate a 32-byte hash value for each data block, and the hash value is combined with the data block index number to form a leaf node. For example, the hash value of the first data block is "7d8f...3a2b", the index number is "0", and the combined leaf node is "0-7d8f...3a2b".

[0048] Then, all leaf nodes are paired and calculated to generate intermediate nodes. The hash values ​​of two adjacent leaf nodes are concatenated and the SHA-256 hash operation is performed again to obtain the parent node hash value. For example, the first-layer leaf nodes "0-7d8f...3a2b" and "1-9e4c...8f1d" are paired and calculated to obtain the parent node "01-5a3d...2c9e". This recursive calculation is repeated until the root node hash value "root-4b2a...9f7d" is finally obtained.

[0049] Extract verification path information from the Merkle tree structure. On the path from the leaf node to the root node, collect the hash values ​​of the sibling nodes used for verification at each layer, and record the path direction (left 0 right 1). For example, the verification path of the first leaf node contains three node hash values ​​"9e4c...8f1d", "2b7a...4e8c", "8c3f...1a9d", and the direction sequence is "011". Combine the node set and the direction sequence to get the complete verification path.

[0050] Construct a temporal integrity proof based on the root node hash value and the verification path. Pack the root node hash value, verification path node set, and path direction sequence to generate a proof data structure. The proof structure contains: root node hash value "root-4b2a...9f7d", verification path node array, direction sequence "011" and other information.

[0051] Finally, a tamper-proof verification certificate is generated. The certificate header contains basic information such as version number, issuance time, and validity period. The certificate body contains the time integrity proof content. Calculate the SHA-256 hash of the certificate header and body content to get the certificate summary, and use the RSA private key to sign the summary to get a digital signature. Add the signature to the certificate to complete the generation.

[0052] Figure 2 This is a schematic diagram of verifying the accuracy evaluation of an embodiment of the present invention: This figure compares the accuracy performance of three verification schemes under different system loads. The triangle-marked "this technical scheme" represents a hybrid verification mechanism that combines zero-knowledge proof and temporal integrity verification, including a node real-time trust scoring system and adaptive threshold judgment; the circle-marked "basic zero-knowledge proof scheme" represents a verification scheme that only uses the standard zero-knowledge proof protocol; the square-marked "traditional verification scheme" represents a conventional blockchain verification scheme using digital signatures and hash chains. When the system load rate is 0.2, this technical scheme maintains a high accuracy of 99.5%, while the basic zero-knowledge proof scheme is 95.0% and the traditional verification scheme is 97.0%. When the load rate increases to 0.6, this technical scheme still maintains an excellent performance of 98.8%, while the other two schemes drop to 91.0% and 93.0% respectively. In the extreme case of the maximum load rate of 1.0, this technical scheme still maintains a high accuracy of 98.0%, which is significantly better than the 85.0% of the basic zero-knowledge proof scheme and the 87.0% of the traditional verification scheme.

[0053] This embodiment provides a method for generating a temporal integrity proof and tamper-proof verification certificate based on a Merkle tree. In the prior art, simple digital signatures or timestamps are usually used to verify data integrity, which has problems such as low verification efficiency, poor traceability, and difficulty in proving the temporal relationship of data.

[0054] This embodiment first performs hash calculation on the data blocks and timestamp sequences in the verification results, and builds leaf nodes in combination with index information, laying the foundation for the subsequent construction of the Merkle tree. This method not only ensures the privacy of the original data, but also realizes the unique identification of the data, solving the problem of difficult verification of data integrity in traditional methods.

[0055] Secondly, this embodiment constructs a Merkle tree structure and generates the hash values ​​of the intermediate nodes and the root node by recursive pairing calculation. Compared with the single hash chain structure in the prior art, the Merkle tree has higher verification efficiency and stronger scalability. Through the design of the tree structure, the computational complexity of the verification process is greatly reduced, and the system's ability to process large-scale data is improved.

[0056] Again, this embodiment innovatively extracts the verification path information in the Merkle tree, combines the verification path node set with the path direction sequence, and forms a complete temporal integrity proof. This design enables the verification of any node to be performed independently without obtaining a complete data set, significantly improving verification efficiency and flexibility.

[0057] Finally, this embodiment normalizes and encapsulates the proof of time integrity by constructing a tamper-proof verification certificate. By adding certificate header information, calculating the certificate summary and performing digital signature, a complete trust chain is established. This multi-level security mechanism effectively prevents the certificate content from being tampered with and ensures the reliability of the verification results.

[0058] Through the above technical solution, this embodiment realizes efficient verification of data integrity and reliable proof of time series relationship, significantly improving the efficiency and credibility of data verification. This solution not only solves the problems of low efficiency and poor scalability in traditional verification methods, but also provides reliable technical support for long-term storage and verification of data, which has important practical significance.

[0059] In an optional implementation, a two-layer consensus smart contract is activated to establish a version control chain, a traceability network is constructed in a trusted node group based on feature fingerprint identification, and the updated data is merged with the original time-series dependency graph through a three-layer Merkel tree structure to generate a new verification link to achieve traceability of the entire data update process, including: Construct the main consensus layer contract and the verification consensus layer contract. The main consensus layer contract records data version control information, and the verification consensus layer contract records data update verification information. Activate the main consensus layer contract and the verification consensus layer contract to form a double-layer consensus smart contract and establish a version control chain. Perform hash calculation on the data block to obtain the data block hash value, combine the data block hash value with the timestamp sequence to generate a feature fingerprint identifier, calculate the fingerprint similarity between nodes based on the feature fingerprint identifier, and select trusted nodes based on the fingerprint similarity to build a traceability network. Construct a three-layer Merkel tree structure in the traceability network, write the updated data information into the data layer of the three-layer Merkel tree structure to generate the terminal node, calculate the temporal dependency relationship based on the terminal node to generate the intermediate node and write it into the dependency layer of the three-layer Merkel tree structure; calculate the verification information of the intermediate node to obtain the root node and write it into the verification layer of the three-layer Merkel tree structure, and establish a data update path based on the terminal node, intermediate node and root node in the three-layer Merkel tree structure; The updated data is merged with the original time-series dependency graph through the data update path to generate a new verification link containing node verification records. The new verification link records the complete process of data update. When receiving a data update request, the main consensus layer contract triggers the verification consensus layer contract to perform verification. After the verification passes, the update information is recorded in the version control chain, and the traceability of the entire data update process is achieved based on the new verification link.

[0060] First, a two-layer consensus smart contract is constructed. The main consensus layer contract contains fields such as data version number, timestamp, data block hash, etc., which are used to record version control information; the verification consensus layer contract contains fields such as verification rules and verification results, which are used to record data verification information. The two contracts are associated through the contract call interface to form a two-layer consensus structure. For example, when the data version is updated from V1.0 to V1.1, the main contract records the version number V1.1, timestamp 20230601, and data block hash value 0xabc, and the verification contract records the verification rule checksum and the verification result valid.

[0061] Then, a traceability network is constructed based on the characteristic fingerprint. The SHA256 hash value is calculated for each data block, and the hash value is concatenated with the timestamp to generate the characteristic fingerprint. The fingerprint similarity between nodes is calculated, and nodes with a similarity of more than 0.8 are selected as trusted nodes. For example, the fingerprint of node A is "0xabc-20230601", and the fingerprint of node B is "0xabc-20230602". The similarity is 0.9, and node B is selected as a trusted node. Connections are established between trusted nodes to form a traceability network.

[0062] Then, a three-layer Merkle tree is constructed in the traceability network. The data layer stores updated data blocks, such as "block_new"; the dependency layer records the temporal dependency between data blocks, such as "block_new depends on block_old"; and the verification layer stores verification information. Based on this three-layer structure, a data update path is established: the data layer node block_new is associated with the verification layer node through the dependency layer node.

[0063] The update path is merged with the original dependency graph. The original dependency graph records the relationship between data blocks. The new path connects the updated data blocks to the original graph through the verification node to form a new verification link. For example, in the original graph, block_old points to block_old2, and after the update, block_new points to block_old through the verification node, achieving traceability.

[0064] When a data update request is received, the main contract triggers the verification contract to execute rule verification. After the verification is passed, the update information is recorded in the version control chain, and the entire update process can be traced based on the verification link.

[0065] The hierarchical management of data version management and verification is realized through two-layer smart contracts, which improves the maintainability and scalability of the system. Based on feature fingerprints and three-layer Merkel tree structure, a trusted data traceability network is built to ensure the traceability and integrity of the data update process. The verification link fusion mechanism is adopted to realize the association mapping between new and old data, making the data update process transparent and controllable, and enhancing the reliability of the system.

[0066] In an optional implementation, performing hash calculation on a data block to obtain a data block hash value, combining the data block hash value with a timestamp sequence to generate a feature fingerprint identifier, calculating fingerprint similarity between nodes based on the feature fingerprint identifier, and selecting a trusted node to build a traceability network based on the fingerprint similarity includes: Perform hash calculation on the input data block to obtain the main hash value of the data block, perform MD5 and SHA1 hash calculation on the data block to obtain the auxiliary hash sequence of the data block, combine the main hash value of the data block with the auxiliary hash sequence of the data block and add a random salt value to generate the hash value of the data block; collect the timestamp information of the data block at different processing nodes, arrange the timestamp information according to the time sequence relationship to generate a timestamp sequence, perform encoding processing on the timestamp sequence to obtain a time feature vector, and perform weighted combination operation on the data block hash value and the time feature vector to generate a feature fingerprint identifier; Establish a node feature matrix, the row vector of the node feature matrix represents the node identity of the feature fingerprint, and the column vector represents the different feature dimensions of the feature fingerprint. The distance between different row vectors in the node feature matrix is ​​calculated based on the local sensitive hashing algorithm to obtain the initial similarity; the initial similarity is weightedly fused with the node's historical behavior reputation to obtain the fingerprint similarity between nodes. The historical behavior reputation is calculated based on the node's historical interaction records and verification pass rate; Set the dynamic node similarity threshold, which is adaptively adjusted according to the network scale and security level. Compare the fingerprint similarity with the dynamic node similarity threshold, and select the nodes with fingerprint similarity higher than the dynamic node similarity threshold as trusted nodes. Calculate the trust transfer strength between trusted nodes, which is determined by the direct similarity and indirect similarity between nodes. Build the weighted connection relationship between nodes based on the trust transfer strength. A hierarchical node traceability network is constructed using weighted connection relationships. The hierarchical node traceability network includes a core trusted layer and an extended verification layer. The core trusted layer is composed of nodes with the highest fingerprint similarity, and the extended verification layer is composed of other trusted nodes.

[0067] When performing hash calculation on a data block to obtain the data block hash value, the SHA256 algorithm is first used to calculate the input data block to obtain the main hash value, and the MD5 and SHA1 algorithms are used to calculate the auxiliary hash value sequence. For example, for a certain data block, a 64-bit SHA256 main hash value, a 32-bit MD5 hash value, and a 40-bit SHA1 hash value are calculated. These hash values ​​are concatenated in the order of the main hash value first and the auxiliary hash value later, and an 8-bit random salt value is added to finally generate a 144-bit data block hash value.

[0068] Collect the timestamp information of data blocks on different processing nodes, including the creation time, modification time, and access time of the data blocks. Arrange these timestamps in time sequence. For example, the timestamp sequence of a data block on three nodes is "2023-01-01 10:00:00, 2023-01-01 10:05:00, 2023-01-01 10:10:00". Encode the timestamp sequence and convert the time interval into a feature vector. For example, "300, 300" means that the intervals between adjacent timestamps are 300 seconds. Perform a weighted combination of the data block hash value and the time feature vector to generate a feature fingerprint.

[0069] Construct a node feature matrix, where each row of the matrix corresponds to a node's feature fingerprint. Use the local sensitive hashing algorithm to calculate the Hamming distance between different row vectors in the matrix to obtain the initial similarity between nodes. At the same time, count the historical interaction records and verification pass rate of each node to calculate the node's historical behavior reputation. Weight the initial similarity and historical behavior reputation with weights of 0.7 and 0.3 to obtain the final fingerprint similarity between nodes.

[0070] According to the number of nodes in the current network and the security level requirements, the node similarity threshold is dynamically adjusted. For example, in a network with 100 nodes, the similarity threshold is set to 0.8; when the number of nodes increases to 1000, the threshold is increased to 0.85 accordingly. The fingerprint similarity between each node is compared with the threshold, and the nodes with similarity higher than the threshold are selected as trusted nodes. The direct similarity between trusted nodes and the indirect similarity transmitted through other nodes are calculated to determine the trust transmission strength between nodes.

[0071] Finally, a hierarchical node traceability network is constructed, with nodes with the highest fingerprint similarity (such as similarity greater than 0.9) forming the core trust layer, and other trusted nodes (similarity between 0.8-0.9) forming the extended verification layer. Weighted connection relationships are established between nodes based on the strength of trust transfer to form a complete traceability network structure.

[0072] Figure 3 This is a schematic diagram comparing the hash calculation performance and feature fingerprint generation efficiency of an embodiment of the present invention: The figure shows a comparison of three different hash calculation schemes. The "technical scheme" marked with a triangle represents an innovative algorithm that combines the main hash value, auxiliary hash sequence and time feature vector; the "single SHA256 hash scheme" marked with a circle represents the traditional single hash algorithm method; the "basic timestamp concatenation scheme" marked with a square represents a simple timestamp connection scheme. Experimental data show that when processing 2MB of data, this technical scheme only takes 40ms, while the single hash scheme takes 85ms and the basic timestamp scheme takes 120ms. When the data volume increases to 6MB, the processing time of the three schemes is 100ms, 220ms and 320ms respectively. When the maximum test data volume is 10MB, the processing time of this technical scheme is 150ms, which is significantly better than the 320ms of the single hash scheme and the 450ms of the basic timestamp scheme. Especially in the large data processing range of 8MB to 10MB, the processing time of this technical scheme only increases by 30ms, while the other two schemes increase by 50ms and 70ms respectively, which fully reflects the performance advantages and stability of this technical scheme in large-scale data processing scenarios.

[0073] Figure 4 This is a schematic diagram of the data block hash calculation status analysis panel according to an embodiment of the present invention: This panel shows a complete data block hash calculation status, using multiple hash algorithms to ensure data security. The main hash value is calculated using the SHA256 algorithm to obtain "a1b2c3d4...e5f6" (64 bits) as the main identity of the data block; the first set of auxiliary hash values ​​is generated using the MD5 algorithm to generate "g7h8i9j0...k1l2" (32 bits), and the second set of auxiliary hash values ​​is generated using the SHA1 algorithm to obtain "m3n4o5p6...q7r8" (40 bits). The system also introduces an 8-bit random salt value "s9t0u1v2" to enhance the randomness and security of hash calculations. This combination of multiple hash algorithms not only provides a unique identification of data blocks, but also significantly improves the system's anti-attack capabilities and data integrity protection level through the complementary characteristics of different algorithms. The "Recalculate" button in the panel allows administrators to update hash values ​​when needed to ensure the real-time and accuracy of data.

[0074] Figure 5 This is a schematic diagram of a timestamp sequence information panel according to an embodiment of the present invention: The panel records the complete time trajectory information of the data block in the system in detail. Starting from the creation time "2023-01-01 10:00:00", through the modification time "2023-01-01 10:05:00", and finally to the access time "2023-01-0110:10:00", a complete timestamp sequence is formed. The system automatically calculates the time interval vector [300, 300], indicating that the data block maintains a stable 5-minute (300-second) time interval at different processing stages. This regular time interval not only reflects the standardization and normativeness of system processing, but also provides a benchmark reference for the detection of abnormal behavior. The "Update Timestamp" function provided by the panel ensures the real-time update of time information, which helps system maintenance personnel to grasp the processing status and flow process of data blocks in a timely manner.

[0075] This embodiment provides a method for constructing a trusted node traceability network based on feature fingerprint identification. In the prior art, a single hash algorithm is usually used for data integrity verification, and the node credibility is determined by simple threshold judgment, which has problems such as insufficient security, one-sided credibility assessment, and limited traceability.

[0076] This embodiment first calculates the data block features through a multiple hash algorithm, and constructs the data block hash value by combining the main hash value, the auxiliary hash sequence and the random salt value. This multi-dimensional feature extraction method greatly improves the uniqueness and anti-tampering ability of the features compared to the single hash calculation in the prior art. At the same time, by introducing a timestamp sequence and performing feature encoding, the accurate characterization of the timing characteristics of the data processing process is achieved.

[0077] Secondly, this embodiment innovatively introduces the node feature matrix and local sensitive hashing algorithm to evaluate the similarity by calculating the feature distance between nodes. Different from the simple binary judgment method in the prior art, this solution constructs a more comprehensive and objective node credibility evaluation mechanism by weighted fusion of the initial similarity and the node historical behavior reputation.

[0078] Thirdly, this embodiment adopts a dynamically adjusted similarity threshold mechanism to adaptively adjust the judgment criteria according to the network scale and security requirements. This flexible threshold setting method effectively solves the problem that traditional fixed thresholds are difficult to adapt to dynamic changes in the network. By calculating the direct and indirect trust transmission strength between nodes, a more realistic trust propagation model is constructed.

[0079] Finally, this embodiment constructs a hierarchical node traceability network and divides the trusted nodes into a core trusted layer and an extended verification layer. This hierarchical structure design not only improves the reliability of the traceability network, but also enhances the scalability and fault tolerance of the system. Through the weighted connection relationship between nodes, more accurate trust transfer and traceability are achieved.

[0080] Through the above technical solution, this embodiment realizes reliable traceability of the data processing process and accurate evaluation of node credibility, significantly improving the security and reliability of the system. This solution breaks through the technical bottlenecks of single feature extraction, rough credibility evaluation, and limited traceability in traditional methods, and provides strong technical support for building a highly reliable distributed data processing system, which has important application value.

[0081] According to a second aspect of the embodiments of the present invention, An electronic device is provided, comprising: processor; a memory for storing processor-executable instructions; The processor is configured to call instructions stored in the memory to execute the aforementioned method.

[0082] According to a third aspect of the embodiments of the present invention, A computer-readable storage medium is provided, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the aforementioned method is implemented.

[0083] The present invention may be a method, an apparatus, a system and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for executing various aspects of the present invention.

[0084] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A decentralized talent background investigation data storage and verification method, characterized in that: include: Based on the identity authentication layer, the candidate's education experience information, work experience information and skill certification information are received, and the received information is asymmetrically encrypted using the national encryption algorithm SM2 to obtain a first encrypted data set. The first encrypted data set is combined with the SM3 hash algorithm to generate an original data packet with a time series weight; The first encrypted data set is segmented according to the timing weight in the original data packet, and a hierarchical timing dependency graph is constructed. The hierarchical timing dependency graph extracts the timing correlation features between each data segment through a deep learning model to generate a second encrypted data set. The second encrypted data set includes a weighted feature fingerprint identifier and an information segment hash value, and the second encrypted data set is written into the trusted node group of the blockchain network through a three-layer Merkle tree structure; When the trusted node group receives a verification request, it locates the target data location based on the feature fingerprint identifier, extracts the second encrypted data set of the target time period according to the hierarchical temporal dependency graph, and calculates the node real-time trust score at the same time. The node real-time trust score is dynamically updated based on the node's verification history behavior feature vector. When the overall trust of the node group reaches the adaptive threshold, the feature fingerprint identifier is used to start the consistency verification mechanism based on zero-knowledge proof, and the consistency verification mechanism generates a tamper-proof verification certificate containing the temporal integrity proof for the verification result; When the hierarchical timing dependency graph detects a need to update data, the correlation between the update request information and the timing weight in the original data packet is calculated, and the deep learning model evaluates the rationality of the update. After confirmation, the data segment to be updated is located in the second encrypted data set, and the two-layer consensus smart contract is activated to establish a version control chain. Based on the feature fingerprint identification, a traceability network is built in the trusted node group. The updated data is merged with the original timing dependency graph through a three-layer Merkel tree structure to generate a new verification link, thereby achieving traceability of the entire data update process.

2. The method according to claim 1, characterized in that The received information is asymmetrically encrypted by the national encryption algorithm SM2 to obtain a first encrypted data set, and the original data packet with a timing weight is generated by combining the SM3 hash algorithm for the first encrypted data set, including: The received education experience, work experience and skill certification information is asymmetrically encrypted using the national secret algorithm SM2, a base point G and order n are selected based on the SM2 elliptic curve cryptography system to generate a key pair (d, Q), the public key Q is used to perform encryption operations on the standardized information, and a first encrypted data set is generated in combination with a random number k, the first encrypted data set contains encrypted structured information; The first encrypted data set is securely processed in combination with the SM3 hash algorithm. The encrypted data is padded with messages in multiples of 512 bits, and iterative operations are performed through the compression function. At the same time, a timing weight parameter is introduced to calculate the timestamp difference and information correlation between adjacent data blocks, and an original data packet with a timing weight is generated. The original data packet has a timing integrity feature.

3. The method according to claim 2, characterized in that The received education experience, work experience and skill certification information is asymmetrically encrypted using the national secret algorithm SM2, a key pair (d, Q) is generated by selecting the base point G and the order n based on the SM2 elliptic curve cryptography system, and the public key Q is used to perform encryption operations on the standardized information, and the first encrypted data set is generated in combination with the random number k, including: Receive the education experience information, work experience information and skill certification information submitted by the user, classify and encode the education experience information, work experience information and skill certification information respectively, obtain education experience structured data, work experience structured data and skill certification structured data, and integrate the education experience structured data, work experience structured data and skill certification structured data to construct a standardized data template; In the SM2 elliptic curve cryptosystem, a base point G and an order n are selected, a private key d is obtained using a random number generator, a public key Q is generated based on a point multiplication operation between the private key d and the base point G, and a key pair (d, Q) is obtained. The key validity verification algorithm is used to verify the validity of the key pair (d, Q); Generate a random number k, perform a point multiplication operation on the random number k and the base point G to obtain a temporary public key, perform a point multiplication operation on the temporary public key and the public key Q to obtain a session key, use a key derivation function to process the session key to generate an encryption key, use the encryption key to encrypt the standardized data template, and generate a first encrypted data set containing encrypted data and its corresponding timestamp.

4. The method according to claim 1, characterized in that: The second encrypted data set of the target time period is extracted according to the hierarchical time dependency graph, and the real-time trust score of the node is calculated at the same time. The real-time trust score of the node is dynamically updated based on the feature vector of the node's verification history behavior. When the overall trust of the node group reaches the adaptive threshold, the feature fingerprint identification is used to start the consistency verification mechanism based on zero-knowledge proof. Based on the consistency verification mechanism, the verification result generates a tamper-proof verification certificate containing the time integrity proof, including: The hierarchical timing dependency graph includes a timing node set and a timing edge set. A second encrypted data set of a target time period is extracted according to the hierarchical timing dependency graph. The second encrypted data set includes encrypted data in the target time period and its corresponding timing dependency relationship. A verification history behavior feature vector is established for the nodes in the timing node set. The node real-time trust score is calculated based on the verification history behavior feature vector. The node real-time trust score is dynamically adjusted as the verification history behavior feature vector is updated. The real-time trust scores of each node in the time series node set are weighted and summed to obtain the overall trust of the node group. The overall trust of the node group is compared with the adaptive threshold. When the overall trust of the node group is greater than the adaptive threshold, a feature fingerprint is generated based on the overall trust of the node group and the adaptive threshold. A prover commitment is constructed based on the characteristic fingerprint identifier, and the prover commitment is combined with the verification rule to generate a zero-knowledge proof. The zero-knowledge proof is used to start the consistency verification mechanism, and the second encrypted data set is checked for consistency and a verification result is generated. A Merkle tree is constructed based on the verification result, and a temporal integrity proof including the Merkle tree root node and the verification path is generated. A tamper-proof verification certificate is generated based on the temporal integrity proof.

5. The method according to claim 4, characterized in that A Merkle tree is constructed based on the verification results, and a temporal integrity proof including the Merkle tree root node and the verification path is generated. The tamper-proof verification certificate is generated based on the temporal integrity proof, including: Perform hash calculation on each data block in the verification result and the corresponding timestamp sequence to obtain the data block hash value, combine the data block hash value with the index information to generate a leaf node; perform pairing calculation on the leaf nodes to obtain the intermediate node hash value, recursively pair the intermediate node hash values ​​until the root node hash value is generated, and construct a Merkle tree structure based on the root node hash value; Extract the verification path node set and path direction sequence in the Merkle tree structure, combine the verification path node set and path direction sequence to generate a verification path, and combine the verification path with the root node hash value to generate a temporal integrity proof; A tamper-proof verification certificate is constructed based on the temporal integrity proof. The tamper-proof verification certificate includes the certificate header information and the certificate body content. The certificate header information and the certificate body content of the tamper-proof verification certificate are hashed to obtain a certificate summary, and the certificate summary is signed with a private key to obtain a digital signature. The digital signature is added to the tamper-proof verification certificate to complete the certificate generation.

6. The method according to claim 1, characterized in that Activate the two-layer consensus smart contract to establish a version control chain, build a traceability network in the trusted node group based on feature fingerprint identification, merge the updated data with the original time-series dependency graph through the three-layer Merkle tree structure, generate a new verification link, and achieve traceability of the entire data update process, including: Construct the main consensus layer contract and the verification consensus layer contract. The main consensus layer contract records data version control information, and the verification consensus layer contract records data update verification information. Activate the main consensus layer contract and the verification consensus layer contract to form a double-layer consensus smart contract and establish a version control chain. Perform hash calculation on the data block to obtain the data block hash value, combine the data block hash value with the timestamp sequence to generate a feature fingerprint identifier, calculate the fingerprint similarity between nodes based on the feature fingerprint identifier, and select trusted nodes based on the fingerprint similarity to build a traceability network. Construct a three-layer Merkel tree structure in the traceability network, write the updated data information into the data layer of the three-layer Merkel tree structure to generate the terminal node, calculate the temporal dependency relationship based on the terminal node to generate the intermediate node and write it into the dependency layer of the three-layer Merkel tree structure; calculate the verification information of the intermediate node to obtain the root node and write it into the verification layer of the three-layer Merkel tree structure, and establish a data update path based on the terminal node, intermediate node and root node in the three-layer Merkel tree structure; The updated data is merged with the original time-series dependency graph through the data update path to generate a new verification link containing node verification records. The new verification link records the complete process of data update. When receiving a data update request, the main consensus layer contract triggers the verification consensus layer contract to perform verification. After the verification passes, the update information is recorded in the version control chain, and the traceability of the entire data update process is achieved based on the new verification link.

7. The method according to claim 6, characterized in that Perform hash calculation on the data block to obtain the data block hash value, combine the data block hash value with the timestamp sequence to generate a feature fingerprint identifier, calculate the fingerprint similarity between nodes based on the feature fingerprint identifier, and select trusted nodes based on the fingerprint similarity to build a traceability network, including: Perform hash calculation on the input data block to obtain the main hash value of the data block, perform MD5 and SHA1 hash calculation on the data block to obtain the auxiliary hash sequence of the data block, combine the main hash value of the data block with the auxiliary hash sequence of the data block and add a random salt value to generate the hash value of the data block; collect the timestamp information of the data block at different processing nodes, arrange the timestamp information according to the time sequence relationship to generate a timestamp sequence, perform encoding processing on the timestamp sequence to obtain a time feature vector, and perform weighted combination operation on the data block hash value and the time feature vector to generate a feature fingerprint identifier; Establish a node feature matrix, the row vector of the node feature matrix represents the node identity of the feature fingerprint, and the column vector represents the different feature dimensions of the feature fingerprint. The distance between different row vectors in the node feature matrix is ​​calculated based on the local sensitive hashing algorithm to obtain the initial similarity; the initial similarity is weightedly fused with the node's historical behavior reputation to obtain the fingerprint similarity between nodes. The historical behavior reputation is calculated based on the node's historical interaction records and verification pass rate; Set the dynamic node similarity threshold, which is adaptively adjusted according to the network scale and security level. Compare the fingerprint similarity with the dynamic node similarity threshold, and select the nodes with fingerprint similarity higher than the dynamic node similarity threshold as trusted nodes. Calculate the trust transfer strength between trusted nodes, which is determined by the direct similarity and indirect similarity between nodes. Build the weighted connection relationship between nodes based on the trust transfer strength. A hierarchical node traceability network is constructed using weighted connection relationships. The hierarchical node traceability network includes a core trusted layer and an extended verification layer. The core trusted layer is composed of nodes with the highest fingerprint similarity, and the extended verification layer is composed of other trusted nodes.

8. A decentralized talent background investigation data storage and verification system, used to implement the method described in any one of claims 1 to 7, characterized in that: include: The first unit is used to receive the education experience information, work experience information and skill certification information submitted by the candidate based on the identity authentication layer, asymmetrically encrypt the received information through the national encryption algorithm SM2 to obtain a first encrypted data set, and combine the SM3 hash algorithm to generate an original data packet with a time series weight for the first encrypted data set; The second unit is used to segment the first encrypted data set according to the timing weight in the original data packet, construct a hierarchical timing dependency graph, and extract the timing correlation features between each data segment through a deep learning model to generate a second encrypted data set. The second encrypted data set includes a weighted feature fingerprint identifier and an information segment hash value, and writes the second encrypted data set into a trusted node group of the blockchain network through a three-layer Merkle tree structure; The third unit is used to locate the target data location based on the feature fingerprint identifier when the trusted node group receives the verification request, extract the second encrypted data set of the target time period according to the hierarchical time sequence dependency graph, and calculate the node real-time trust score at the same time. The node real-time trust score is dynamically updated based on the node's verification history behavior feature vector. When the overall trust of the node group reaches the adaptive threshold, the feature fingerprint identifier is used to start the consistency verification mechanism based on the zero-knowledge proof, and the verification result is generated based on the consistency verification mechanism. The tamper-proof verification certificate containing the time sequence integrity proof; The fourth unit is used to calculate the correlation between the update request information and the timing weight in the original data packet when the hierarchical timing dependency graph detects a need for data update, and the deep learning model evaluates the rationality of the update. After confirmation, the data segment to be updated is located in the second encrypted data set, and the two-layer consensus smart contract is activated to establish a version control chain. A traceability network is built in the trusted node group based on feature fingerprint identification, and the updated data is merged with the original timing dependency graph through a three-layer Merkel tree structure to generate a new verification link, thereby achieving traceability of the entire data update process.

9. An electronic device, characterized in that: include: processor; a memory for storing processor-executable instructions; The processor is configured to call the instructions stored in the memory to execute the method described in any one of claims 1 to 7.

10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images