Multi-party marketing data privacy collaboration method and system based on cross-chain technology

Abstract

The invention discloses a multi-party marketing data privacy collaboration method based on a cross-chain technology, relates to the technical field of data security and privacy computing, and aims to solve the technical problems that a current centralized data collaboration scheme has trust dependence, security risks and efficiency bottlenecks and is difficult to meet cross-mechanism data security fusion and efficient collaboration requirements. The method is implemented on the basis of a multi-party marketing data privacy collaboration system based on the cross-chain technology, and the multi-party marketing data privacy collaboration system based on the cross-chain technology comprises a data provider node cluster: deploying a client with a feature engineering module and a homomorphic encryption engine, the cross-chain relay network is used for generating encrypted feature data and initiating cross-chain asset registration, the cross-chain relay network comprises a verification node pool and a cross-chain protocol gateway, and the verification node pool verifies a cross-chain transaction through a consensus mechanism. The method has the advantages of high credibility, improved privacy level and stable collaboration efficiency.

Description

Technical Field

[0001] This invention relates to the field of data security and privacy computing technology, and more specifically, to a multi-party marketing data privacy collaboration method and system based on cross-chain technology. Background Technology

[0002] With the development of the digital economy, collaborative analysis of marketing data across institutions and industries has become crucial for enhancing business intelligence. However, due to strict restrictions on data privacy, trade secrets, and laws and regulations (such as GDPR and data security laws), "data silos" are formed, making it difficult to securely integrate and extract value from data from multiple parties under the premise of mutual trust.

[0003] Traditional data collaboration solutions often rely on a trusted third-party central server. Each participant must upload raw or encrypted data to this central server, which then handles all data processing and analysis. Data providers must unconditionally trust the central server and rely on it to protect their data from misuse or leakage. This contradicts the current decentralized business development logic and fails to meet the trust requirements of multi-party collaboration. If the central server suffers a hacker attack or system failure, the privacy data of all participants will face the risk of centralized leakage, posing a serious threat to data security. All data processing and computational pressure is concentrated on the central server. When faced with high-concurrency collaborative computing needs for large-scale marketing data, processing delays and system lags are likely to occur, making it impossible to guarantee collaboration efficiency. Therefore, we propose a multi-party marketing data privacy collaboration method based on cross-chain technology. Summary of the Invention

[0004] The purpose of this invention is to provide a multi-party marketing data privacy collaboration method based on cross-chain technology, so as to solve the technical problems of current centralized data collaboration solutions, which have trust dependence, security risks and efficiency bottlenecks, and are difficult to meet the needs of cross-institutional data security integration and efficient collaboration.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a multi-party marketing data privacy collaboration method based on cross-chain technology, wherein the method is implemented based on a multi-party marketing data privacy collaboration system based on cross-chain technology, and the multi-party marketing data privacy collaboration method based on cross-chain technology includes;

[0006] The data provider node locally extracts features from the raw marketing data to generate feature vectors, and then encrypts the feature vectors using a homomorphic encryption algorithm to obtain encrypted feature data.

[0007] The data provider node registers encrypted feature data and data description information including data feature type, statistical distribution label and industry traceability identifier as encrypted assets to the source blockchain through the first cross-chain transaction;

[0008] The cross-chain intermediary service listens to transaction events on the source blockchain. After verifying the legality of the first cross-chain transaction with zero-knowledge proof, it relays the encrypted feature data and data description information to the target collaborative blockchain according to the cross-chain protocol.

[0009] The data user node submits a collaborative computing request to the target side collaborative blockchain. The request includes the preset collaborative computing logic to be invoked and the cryptographic asset identifiers participating in the computing.

[0010] The target side's collaborative blockchain smart contract responds to the request, calls the preset homomorphic operation module to perform collaborative computation on multi-source encrypted feature data, and generates encrypted computation results;

[0011] The smart contract stores the encrypted computation results on the target side's collaborative blockchain and triggers a token settlement mechanism based on preset rules;

[0012] The data user node obtains the encrypted computation results from the target side collaborative blockchain, and decrypts the complete private key by combining the private key components securely distributed by multiple parties, thus obtaining readable collaborative analysis results.

[0013] The multi-party marketing data privacy collaboration method based on cross-chain technology includes the following steps:

[0014] S1: Key System Initialization and Distribution

[0015] Operating entities: Key management center, data provider node, and data user node;

[0016] Process description:

[0017] The key management center generates globally homomorphic cryptographic public keys using the Secure Multi-Party Computation (SMPC) protocol. and N private key components The threshold condition K≥N (K is the minimum number of private key components required for decryption) is satisfied.

[0018] Global public key The broadcast is sent to all data provider nodes, and the private key component is distributed to N data user nodes via a secure channel. Data provider nodes verify the validity of the public key and store it locally, while data user nodes confirm the integrity of the private key component.

[0019] S2: Raw Data Feature Extraction and Encryption

[0020] Operating entity: Data provider node;

[0021] Process description:

[0022] The data provider performs feature engineering on the raw marketing data (such as user behavior, purchase records, etc.) locally:

[0023] Cleaning and noise reduction: Remove outliers and missing values;

[0024] Feature transformation: Converting unstructured data into structured features (such as One-Hot encoding, standardization).

[0025] Generate m-dimensional feature vectors:

[0026]

[0027] Use global public key Perform homomorphic encryption on the feature vector:

[0028] Encryption function Satisfies additive homomorphism:

[0029]

[0030] Generate encrypted feature data:

[0031]

[0032] Generate data description information M, including:

[0033] Feature metadata: feature type (continuous / discrete), dimension range;

[0034] Quality indicators: data completeness score q∈[0,1], sample size n;

[0035] Security identifiers: industry traceability code, usage permission attribute set;

[0036] S3: Source Blockchain Asset Registration

[0037] Operating entities: data provider nodes and source-side blockchain;

[0038] Process description:

[0039] Data provider builds the first cross-chain transaction Transaction Structure: =(sender address, , (Digital signature, timestamp);

[0040] Digital signatures use the ECDSA algorithm:

[0041]

[0042] Private key provider submits On the source blockchain, nodes verify the legality of transactions:

[0043] Verify signature validity:

[0044]

[0045] Verify data format compliance (feature dimensions, completeness of descriptive information);

[0046] Once verification is successful, the source blockchain generates a unique identifier for the encrypted asset:

[0047] ( ∥Block Height∥ (hash);

[0048] Will Linked storage completes asset on-chain recording;

[0049] S4: Cross-chain data relay and verification

[0050] Operating entities: cross-chain intermediary services, target-side collaborative blockchain;

[0051] Process description:

[0052] Cross-chain intermediary services monitor source blockchain transaction events and capture newly registered crypto assets:

[0053] Obtain the latest block from the source chain through the block header synchronization mechanism;

[0054] Parse the block transaction list and extract transaction records containing crypto assets;

[0055] Execute the cross-chain verification process:

[0056] Zero-knowledge proof verification: generating proof of transaction validity This proves that the transaction has not been tampered with and complies with the rules;

[0057] Consensus cluster verification: M verification nodes use the PBFT algorithm to verify the proof. Consensus must be reached, and the consensus result requires the agreement of ≥2 / 3M nodes. Block hash verification: Verify the source chain block header hash and transaction Merkel proof to confirm the authenticity of the data source.

[0058] Data format conversion and relay:

[0059] The cross-chain protocol gateway converts the source chain asset format into a target chain-compatible format (preserving the ciphertext state) to generate a second cross-chain transaction. , including the converted Will Submit to the target-side collaborative blockchain to complete cross-chain asset registration;

[0060] S5: Collaborative Computing Request Submission and Permission Verification

[0061] Operating entities: data user nodes and target-side collaborative blockchain smart contracts;

[0062] Process description:

[0063] Data users can query available encrypted assets:

[0064] Submit query conditions through the smart contract's attribute encryption query interface;

[0065] Provide attribute proof θ to obtain a list of data description information that meets the criteria;

[0066] Construct a collaborative computing request R:

[0067] Request content:

[0068] R=(requester address, (Deposit amount, signature)

[0069] in Preset collaborative computing logic identifiers (such as "logistic regression training" and "user similarity calculation");

[0070] Smart contract permission verification:

[0071] Verify whether the requester has the corresponding computing permissions (match the permission identifier in the data description information);

[0072] Lock the requested amount of tokens as a deposit to ensure that computing fees are paid;

[0073] Verify the status of the crypto assets involved in the computation (whether they are available or locked);

[0074] S6: Ciphertext Domain Collaborative Computation Execution

[0075] Operating entities: target-side collaborative blockchain smart contracts and computing engines;

[0076] Process description:

[0077] Smart contracts call pre-defined computational logic Load the corresponding encrypted feature data:

[0078] Retrieve from on-chain storage:

[0079]

[0080] Initialize the computing engine (homogeneous operation module) and allocate on-chain computing resources;

[0081] Performing encrypted domain collaborative computation (taking a typical scenario as an example):

[0082] Scenario 1: Distributed Model Training

[0083] Initialize model parameters

[0084] Iterative execution:

[0085] Calculate the encryption gradient:

[0086]

[0087] Update encryption parameters:

[0088] ( (learning rate)

[0089] Scenario 2: Feature Similarity Calculation

[0090] Calculate the encrypted cosine similarity:

[0091] Generate encrypted calculation results

[0092]

[0093] This includes the calculation output, intermediate process hash, and calculation timestamp;

[0094] S7: Result Notarization and Automatic Token Settlement

[0095] Operating entity: Target-side collaborative blockchain smart contract;

[0096] Process description:

[0097] Evidence of calculation results:

[0098] Result hash generated:

[0099] =SHA256( / / Block hash / / / Request ID);

[0100] Will and Write it to the blockchain, linking the requester's address with the computation task ID;

[0101] Record the audit log of the calculation process (participating assets, calculation steps, resource consumption);

[0102] Token settlement mechanism execution:

[0103] Calculate the contribution weight of each data provider:

[0104]

[0105] ( (The feature importance coefficient is preset by the smart contract).

[0106] Distribute tokens:

[0107] Data provider i obtains: =(1−α)·deposit· ;

[0108] Cross-chain service fee: Fee = α·deposit (α∈[0.01,0.05]);

[0109] The smart contract automatically executes token transfers to unlock unused deposits;

[0110] S8: Decryption and Acquisition of Calculation Results

[0111] Operating entity: Data user node;

[0112] Process description:

[0113] Data users obtain encrypted computation results from the target-side blockchain. :

[0114] Query related transactions by transaction hash or task ID. Verification result hash Consistency with on-chain records ensures that the data has not been tampered with.

[0115] Private key reconstruction and decryption:

[0116] Collect private key components {sk1,...,skK} from at least K data users;

[0117] Reconstruct the complete private key using Lagrange interpolation:

[0118]

[0119] Decrypt using the private key:

[0120] Obtain readable collaborative analysis results;

[0121] Result verification: The reasonableness of the decryption result was verified locally.

[0122] This invention replaces the traditional central server with cross-chain technology and blockchain smart contracts. Data, after encryption, is directly processed on-chain by smart contracts through collaborative computation, with rules that are open, transparent, and immutable. Parties no longer need to trust any single institution, but instead rely on automated execution logic guaranteed by code and mathematical algorithms. This perfectly aligns with the trend of decentralized business development and solves the fundamental trust problem in multi-party collaboration.

[0123] Preferably, the homomorphic encryption algorithm employs a partially homomorphic encryption or fully homomorphic encryption mechanism, and is executed before encryption:

[0124] A trusted initialization node generates a global public key and N private key components through a secret sharing protocol, where N is the number of data user nodes;

[0125] The global public key is broadcast to all data provider nodes, and the N private key components are distributed to the corresponding N data user nodes through a secure channel.

[0126] Any data user node needs to collect at least K private key components (K≤N) to reconstruct the complete private key. The value of K is preset by the initialization node.

[0127] Preferably, the preset collaborative computing logic includes:

[0128] The distributed gradient descent training module based on encrypted feature vectors supports parameter updates for linear regression, logistic regression, or neural network models in encrypted form.

[0129] The encrypted statistical analysis module for multi-source data can perform weighted summation, covariance matrix calculation, and quantile statistics in the encrypted domain;

[0130] The feature similarity calculation module can calculate the cosine similarity or Euclidean distance between encrypted vectors without revealing the original features.

[0131] Preferably, the data description information further includes a data quality score and a usage permission identifier, and the method further includes:

[0132] The smart contract of the target-side collaborative blockchain provides an attribute-encrypted query interface, and data user nodes obtain data description information that meets the conditions by submitting attribute proofs;

[0133] The smart contract automatically verifies the calling permissions of the data user based on the usage permission identifier in the data description information, and only allows collaborative computing requests that meet the permission requirements to be executed.

[0134] Preferably, the token settlement mechanism includes:

[0135] The smart contract locks a preset number of tokens in the data-using node when a collaborative computing request is submitted;

[0136] The contribution weights of each crypto asset involved in the calculation are determined by a combination of data feature importance, data volume, and quality score.

[0137] After the calculation is completed, tokens are automatically distributed to the cross-chain accounts of the corresponding data provider nodes according to the contribution weight, and a preset proportion of cross-chain service fees are deducted.

[0138] Preferably, the cross-chain intermediary service adopts a hybrid cross-chain architecture, including:

[0139] A consensus cluster consisting of M verification nodes uses a practical Byzantine fault-tolerant algorithm to reach consensus on cross-chain messages.

[0140] The cross-chain verification module verifies the legitimacy of the source of encrypted feature data by verifying the block header hash and transaction Merkel proof of the source blockchain.

[0141] The data conversion module converts the encrypted asset format of the source blockchain into a format compatible with the target collaborative blockchain, while maintaining the encrypted state.

[0142] Preferably, it includes:

[0143] Data provider node cluster: Deploys clients with feature engineering modules and homomorphic encryption engines to generate encrypted feature data and initiate cross-chain asset registration;

[0144] Cross-chain relay network: includes a validator node pool and a cross-chain protocol gateway. The validator node pool verifies cross-chain transactions through a consensus mechanism, and the cross-chain protocol gateway realizes the conversion of encrypted data formats between different blockchains.

[0145] Target-side collaborative blockchain: Deploy a smart contract system that includes a computing engine, a permission management module, and a token settlement module, wherein the computing engine supports collaborative computation of the encrypted field;

[0146] Data user node cluster: Deploy clients with private key management modules and decryption engines to submit computation requests, reconstruct private keys, and decrypt computation results;

[0147] Key Management Center: Generates and distributes encrypted key pairs through a secure multi-party computation protocol, and supports distributed storage and reconstruction of private key components.

[0148] Preferably, the clients of the data provider node and the data user node further integrate:

[0149] The off-chain pre-computation module is used to perform some homomorphic operations locally, reducing on-chain computation overhead.

[0150] The transaction monitoring module tracks the status of cross-chain assets and the progress of collaborative computing tasks in real time.

[0151] The smart contract interface adaptation layer supports standardized interaction with smart contracts on different blockchain platforms.

[0152] Preferably, the smart contract system of the target-side collaborative blockchain further includes:

[0153] The computation task scheduling module dynamically allocates on-chain computing resources based on data size and computational complexity.

[0154] The result storage module uses a combination of encrypted hash and timestamp to store the calculation results in an anti-tampering manner.

[0155] The audit log module records all calls to encrypted assets and the calculation process, supporting compliance audits by regulatory nodes.

[0156] Compared with the prior art, the beneficial effects of the present invention are:

[0157] 1. This invention replaces the traditional central server with cross-chain technology and blockchain smart contracts. Data is encrypted and directly computed on-chain by smart contracts, with rules that are open, transparent, and immutable. Parties do not need to trust any single institution, but instead trust the automated execution logic guaranteed by code and mathematical algorithms. This perfectly aligns with the trend of decentralized business development and solves the fundamental trust problem in multi-party collaboration.

[0158] 2. This invention also ensures that the data provider performs feature extraction and homomorphic encryption locally, preventing the original data from ever leaving the local machine. The encrypted data remains encrypted throughout cross-chain transmission, on-chain storage, and the entire collaborative computation process. The data user ultimately obtains the decrypted aggregated analysis results, not the original data from any participating party. This model fundamentally cuts off the path of data leakage or misuse during circulation and use, significantly improving privacy protection.

[0159] 3. This invention also distributes computational tasks across multiple nodes in the target blockchain network, rather than centralizing them on a single server. This distributed computing architecture not only avoids single-point performance bottlenecks and better handles the high-concurrency computing demands of large-scale data, ensuring collaborative efficiency, but also prevents partial system failures from paralyzing the entire service, significantly improving the availability and robustness of the entire collaborative system.

[0160] 4. This invention also utilizes a token settlement mechanism built into smart contracts to automatically and accurately distribute revenue to data providers based on contribution indicators such as data quality and feature importance. This process is transparent and requires no human intervention, incentivizing the provision of high-quality data while ensuring contributors receive fair compensation, thereby constructing a healthy and sustainable ecosystem for the circulation of data elements. Attached Figure Description

[0161] Figure 1 This is a flowchart illustrating the data preparation and registration process of the present invention.

[0162] Figure 2 This is a flowchart of the cross-chain collaborative computing process of the present invention;

[0163] Figure 3 This is a flowchart illustrating the process of obtaining and decrypting the results of this invention. Detailed Implementation

[0164] To facilitate understanding of the technical solution of the present invention by those skilled in the art, the technical solution of the present invention will now be further described in conjunction with the accompanying drawings.

[0165] Example 1

[0166] like Figures 1 to 3 As shown, the multi-party marketing data privacy collaboration method based on cross-chain technology provided by this invention includes: The data provider node locally extracts features from the original marketing data to generate feature vectors, and encrypts the feature vectors using a homomorphic encryption algorithm to obtain encrypted feature data; the data provider does not directly upload the original user data, but instead generates feature vectors locally using feature engineering methods. The data provider generates a public-private key pair for the homomorphic encryption algorithm. Subsequently, the feature vectors are encrypted using the public key. The result is encrypted feature data, which is a string of ciphertext that cannot be directly deciphered. The core point is that subsequent calculations can be performed directly on this ciphertext.

[0167] Data provider nodes register encrypted characteristic data, along with data description information including data characteristic types, statistical distribution tags, and industry traceability identifiers, as encrypted assets to the source blockchain through the first cross-chain transaction. This specifies the meaning of each dimension in the vector, enabling data users to understand the significance of the data. It describes the overall statistical characteristics of the data, helping data users assess data quality and applicability without revealing specific data. It uniquely identifies the enterprise, industry, or dataset from which the data originates, used for data traceability and rights verification.

[0168] The cross-chain intermediary service monitors transaction events on the source blockchain. After verifying the legitimacy of the first cross-chain transaction using zero-knowledge proofs, it relays the encrypted feature data and data description information to the target collaborative blockchain according to the cross-chain protocol. The cross-chain intermediary service continuously monitors all events related to "crypto asset registration" on the source blockchain. When it detects the first cross-chain transaction, the intermediary service does not directly trust it but verifies its legitimacy using zero-knowledge proofs. For example, it verifies that the transaction was indeed initiated by a legitimate data provider node. The encrypted data in the transaction is in the correct format and matches the description information. The data provider does indeed own the data it claims to have (without disclosing the data itself). After successful verification, the cross-chain intermediary service securely "relays" or "maps" the encrypted feature data and its description information to the target collaborative blockchain based on a predefined cross-chain protocol (such as based on a relay chain, hash time lock, etc.).

[0169] The data-using node submits a collaborative computation request to the target-side collaborative blockchain. This request includes the pre-defined collaborative computation logic to be invoked and the identifiers of the cryptographic assets participating in the computation; that is, the computation function to be executed, such as "F(Data A, Data B) = Weighted Average Purchasing Power" or "G(Data A, Data C) = Cosine Similarity". This logic is pre-deployed on the target chain in the form of smart contracts; specifying which data providers' cryptographic data needs to be used (identified by their unique IDs on the chain).

[0170] The target side's collaborative blockchain smart contract responds to the request, calls the preset homomorphic operation module to perform collaborative computation on multi-source encrypted feature data, and generates encrypted computation results;

[0171] The smart contract stores the encrypted computation result on the target-side collaborative blockchain and triggers a token settlement mechanism based on preset rules. The smart contract retrieves encrypted feature data (still in ciphertext) provided by multiple parties from the blockchain based on the asset identifier. The contract calls a pre-built homomorphic operation module to directly execute the computation logic specified in the request on the ciphertext; the smart contract records the encrypted computation result as a new data asset on the target-side blockchain, completing the notarization.

[0172] The data user node obtains the encrypted computation result from the target side's collaborative blockchain, reconstructs the complete private key by combining the private key components securely distributed by multiple parties, and then decrypts it to obtain a readable collaborative analysis result. The complete decryption private key is not held by any single institution. During scheme initialization, the master private key is split into multiple private key components through secure multi-party computation (MPC) or secret sharing technology, and distributed to different authoritative parties or the data provider itself. The data user initiates requests to these nodes holding private key components, and each node uses its own component to partially decrypt the encrypted result, and then returns the partially decrypted result to the data user.

[0173] The multi-party marketing data privacy collaboration method based on cross-chain technology includes the following steps:

[0174] S1: Key System Initialization and Distribution

[0175] Operating entities: Key management center, data provider node, and data user node;

[0176] Process description:

[0177] The key management center generates globally homomorphic cryptographic public keys using the Secure Multi-Party Computation (SMPC) protocol. and N private key components The threshold condition K≥N (K is the minimum number of private key components required for decryption) is satisfied.

[0178] Global public key The broadcast is sent to all data provider nodes, and the private key component is distributed to N data user nodes via a secure channel. Data provider nodes verify the validity of the public key and store it locally, while data user nodes confirm the integrity of the private key component.

[0179] S2: Raw Data Feature Extraction and Encryption

[0180] Operating entity: Data provider node;

[0181] Process description:

[0182] The data provider performs feature engineering on the raw marketing data (such as user behavior, purchase records, etc.) locally:

[0183] Cleaning and noise reduction: Remove outliers and missing values;

[0184] Feature transformation: Converting unstructured data into structured features (such as One-Hot encoding, standardization).

[0185] Generate m-dimensional feature vectors:

[0186]

[0187] Use global public key Perform homomorphic encryption on the feature vector:

[0188] Encryption function Satisfies additive homomorphism:

[0189]

[0190] Generate encrypted feature data:

[0191]

[0192] Generate data description information M, including:

[0193] Feature metadata: feature type (continuous / discrete), dimension range;

[0194] Quality indicators: data completeness score q∈[0,1], sample size n;

[0195] Security identifiers: industry traceability code, usage permission attribute set;

[0196] S3: Source Blockchain Asset Registration

[0197] Operating entities: data provider nodes and source-side blockchain;

[0198] Process description:

[0199] Data provider builds the first cross-chain transaction Transaction Structure: =(sender address, , (Digital signature, timestamp);

[0200] Digital signatures use the ECDSA algorithm:

[0201]

[0202] Private key provider submits On the source blockchain, nodes verify the legality of transactions:

[0203] Verify signature validity:

[0204]

[0205] Verify data format compliance (feature dimensions, completeness of descriptive information);

[0206] Once verification is successful, the source blockchain generates a unique identifier for the encrypted asset:

[0207] ( ∥Block Height∥ (hash);

[0208] Will Linked storage completes asset on-chain recording;

[0209] S4: Cross-chain data relay and verification

[0210] Operating entities: cross-chain intermediary services, target-side collaborative blockchain;

[0211] Process description:

[0212] Cross-chain intermediary services monitor source blockchain transaction events and capture newly registered crypto assets:

[0213] Obtain the latest block from the source chain through the block header synchronization mechanism;

[0214] Parse the block transaction list and extract transaction records containing crypto assets;

[0215] Execute the cross-chain verification process:

[0216] Zero-knowledge proof verification: generating proof of transaction validity This proves that the transaction has not been tampered with and complies with the rules;

[0217] Consensus cluster verification: M verification nodes use the PBFT algorithm to verify the proof. Consensus must be reached, and the consensus result requires the agreement of ≥2 / 3M nodes. Block hash verification: Verify the source chain block header hash and transaction Merkel proof to confirm the authenticity of the data source.

[0218] Data format conversion and relay:

[0219] The cross-chain protocol gateway converts the source chain asset format into a target chain-compatible format (preserving the ciphertext state) to generate a second cross-chain transaction. , including the converted Will Submit to the target-side collaborative blockchain to complete cross-chain asset registration;

[0220] S5: Collaborative Computing Request Submission and Permission Verification

[0221] Operating entities: data user nodes and target-side collaborative blockchain smart contracts;

[0222] Process description:

[0223] Data users can query available encrypted assets:

[0224] Submit query conditions through the smart contract's attribute encryption query interface;

[0225] Provide attribute proof θ to obtain a list of data description information that meets the criteria;

[0226] Construct a collaborative computing request R:

[0227] Request content:

[0228] R=(requester address, (Deposit amount, signature)

[0229] in Preset collaborative computing logic identifiers (such as "logistic regression training" and "user similarity calculation");

[0230] Smart contract permission verification:

[0231] Verify whether the requester has the corresponding computing permissions (match the permission identifier in the data description information);

[0232] Lock the requested amount of tokens as a deposit to ensure that computing fees are paid;

[0233] Verify the status of the crypto assets involved in the computation (whether they are available or locked);

[0234] S6: Ciphertext Domain Collaborative Computation Execution

[0235] Operating entities: target-side collaborative blockchain smart contracts and computing engines;

[0236] Process description:

[0237] Smart contracts call pre-defined computational logic Load the corresponding encrypted feature data:

[0238] Retrieve from on-chain storage:

[0239]

[0240] Initialize the computing engine (homogeneous operation module) and allocate on-chain computing resources;

[0241] Performing encrypted domain collaborative computation (taking a typical scenario as an example):

[0242] Scenario 1: Distributed Model Training

[0243] Initialize model parameters

[0244] Iterative execution:

[0245] Calculate the encryption gradient:

[0246]

[0247] Update encryption parameters:

[0248] ( (learning rate)

[0249] Scenario 2: Feature Similarity Calculation

[0250] Calculate the encrypted cosine similarity:

[0251] Generate encrypted calculation results

[0252]

[0253] This includes the calculation output, intermediate process hash, and calculation timestamp;

[0254] S7: Result Notarization and Automatic Token Settlement

[0255] Operating entity: Target-side collaborative blockchain smart contract;

[0256] Process description:

[0257] Evidence of calculation results:

[0258] Result hash generated:

[0259] =SHA256( / / Block hash / / / Request ID);

[0260] Will and Write it to the blockchain, linking the requester's address with the computation task ID;

[0261] Record the audit log of the calculation process (participating assets, calculation steps, resource consumption);

[0262] Token settlement mechanism execution:

[0263] Calculate the contribution weight of each data provider:

[0264]

[0265] ( (The feature importance coefficient is preset by the smart contract).

[0266] Distribute tokens:

[0267] Data provider i obtains: =(1−α)·deposit· ;

[0268] Cross-chain service fee: Fee = α·deposit (α∈[0.01,0.05]);

[0269] The smart contract automatically executes token transfers to unlock unused deposits;

[0270] S8: Decryption and Acquisition of Calculation Results

[0271] Operating entity: Data user node;

[0272] Process description:

[0273] Data users obtain encrypted computation results from the target-side blockchain. :

[0274] Query related transactions by transaction hash or task ID. Verification result hash Consistency with on-chain records ensures that the data has not been tampered with.

[0275] Private key reconstruction and decryption:

[0276] Collect private key components {sk1,...,skK} from at least K data users;

[0277] Reconstruct the complete private key using Lagrange interpolation:

[0278]

[0279] Decrypt using the private key:

[0280] Obtain readable collaborative analysis results;

[0281] Result verification: The reasonableness of the decryption result was verified locally.

[0282] In embodiments of the present invention, the homomorphic encryption algorithm employs a partially homomorphic encryption or fully homomorphic encryption mechanism, and is executed before encryption:

[0283] A trusted initialization node generates a global public key and N private key components through a secret sharing protocol, where N is the number of data user nodes;

[0284] The global public key is broadcast to all data provider nodes, and the N private key components are distributed to the corresponding N data user nodes through a secure channel.

[0285] Any data user node needs to collect at least K private key components (K≤N) to reconstruct the complete private key. The value of K is preset by the initialization node.

[0286] In an embodiment of the present invention, the preset collaborative computing logic includes:

[0287] A distributed gradient descent training module based on encrypted feature vectors supports parameter updates for linear regression, logistic regression, or neural network models in encrypted form; this is the core of federated learning. Multiple data providers each hold encrypted local data features.

[0288] The encrypted statistical analysis module for multi-source data can perform weighted summation, covariance matrix calculation, and quantile statistics in the encrypted domain; it supports basic joint data analysis to analyze the correlation between multiple encrypted feature variables, and is the foundation for many advanced statistical analyses (such as principal component analysis, PCA). Matrix operations are implemented through homomorphic addition and multiplication.

[0289] The feature similarity calculation module can calculate the cosine similarity or Euclidean distance between encrypted vectors without revealing the original features; it is often used to measure the similarity of user interests and document content.

[0290] In embodiments of the present invention, the data description information further includes a data quality score and a usage permission identifier, and the method further includes:

[0291] The smart contract of the target-side collaborative blockchain provides an attribute-based encrypted query interface. Data-using nodes obtain data description information that meets the conditions by submitting attribute proofs. A numerical value, set by the system algorithm or the data provider itself, reflects the data's completeness, accuracy, and freshness. This will become an important basis for the "contribution weight" in subsequent settlements, incentivizing providers to share high-quality data.

[0292] The smart contract automatically verifies the data user's access permissions based on the access permission identifiers in the data description information, allowing only collaborative computing requests that meet the permission requirements to execute. The data user cannot directly browse all data description information. It needs to submit an "attribute proof" to the smart contract, demonstrating that it meets certain conditions. After verifying the proof, the contract's query interface only returns data description information whose access permission identifiers match these attributes.

[0293] In an embodiment of the present invention, the token settlement mechanism includes:

[0294] When a collaborative computation request is submitted, the smart contract locks a preset amount of tokens in the data-using node; when submitting a request, the data user must lock sufficient tokens as a "prepayment" and "margin." This ensures the reliability of payments, prevents situations where payments cannot be made after computation is completed, and also gives the smart contract the ability to automatically deduct payments.

[0295] The contribution weights of each crypto asset involved in the calculation are determined by a combination of data feature importance, data volume, and quality score. In tasks such as model training, the importance of a feature can be determined by analyzing its contribution to the final model.

[0296] After calculation, tokens are automatically distributed to the cross-chain accounts of the corresponding data provider nodes according to the contribution weight, and a preset proportion of cross-chain service fees is deducted. After calculation, the smart contract, acting as a "trusted intermediary," immediately executes liquidation: Total Fee = Locked Tokens - Cross-chain Service Fee. Then, the total fee is allocated to the cross-chain accounts of each provider according to their contribution weight.

[0297] In an embodiment of the present invention, the cross-chain intermediary service adopts a hybrid cross-chain architecture, including:

[0298] A consensus cluster consisting of M verification nodes uses a practical Byzantine fault-tolerant algorithm to reach consensus on cross-chain messages.

[0299] The cross-chain verification module verifies the legitimacy of the source of encrypted data by verifying the block header hash and transaction Merkle proof of the source blockchain; this is a lightweight client-based verification principle. The intermediary service does not need to trust the source blockchain, but rather mathematically and trustlessly confirms the legitimacy of the "first cross-chain transaction" by verifying its block header hash (proving the blockchain state) and transaction Merkle proof (proving that a transaction is indeed included in a block).

[0300] The data conversion module converts the encrypted asset format of the source blockchain into a format compatible with the target blockchain while preserving the encrypted state. Different blockchains have different data standards. This module is responsible for converting the "encrypted asset" data structure of the source chain into an equivalent data structure that the target chain can recognize and process.

[0301] Example 2

[0302] This invention provides a multi-party marketing data privacy collaboration system based on cross-chain technology, according to Embodiment 1, comprising:

[0303] Data provider node cluster: Deploys clients with feature engineering modules and homomorphic encryption engines to generate encrypted feature data and initiate cross-chain asset registration; transforms raw data (user behavior logs, transaction records) into standardized, high-value feature vectors. Supports common data preprocessing and feature transformation algorithms. Integrates mainstream homomorphic encryption libraries such as Paillier and CKKS for efficient encryption and decryption operations.

[0304] Cross-chain relay network: This includes a validator node pool and a cross-chain protocol gateway. The validator node pool verifies cross-chain transactions through a consensus mechanism, while the cross-chain protocol gateway performs encrypted data format conversion between different blockchains. It is a decentralized network composed of multiple independent institutions. They verify the "first cross-chain transaction" captured from the source chain through a consensus mechanism (such as a BFT-like algorithm), ensuring the transaction is valid, the data format is correct, and the source is legitimate. This replaces a single centralized intermediary, preventing single points of corruption or failure. It is responsible for the "translation" work. It converts and repackages the encrypted data, transaction format, and asset identifiers on the source chain according to the specifications of the target chain, achieving interoperability between heterogeneous blockchains.

[0305] Target-side collaborative blockchain: Deploy a smart contract system that includes a computing engine, a permission management module, and a token settlement module, wherein the computing engine supports collaborative computation of the encrypted field;

[0306] Data user node cluster: Deploy clients with private key management modules and decryption engines to submit computation requests, reconstruct private keys, and decrypt computation results;

[0307] Key Management Center: Generates and distributes encrypted key pairs through a secure multi-party computation (MPC) protocol, supporting distributed storage and reconstruction of private key components; based on the secure multi-party computation (MPC) protocol. During system initialization, multiple pre-defined authoritative parties participate jointly to generate a system master-public-private key pair without revealing their respective generated random number components. The complete master-private key never exists at any time or place.

[0308] In embodiments of the present invention, the clients of the data provider node and the data user node further integrate:

[0309] The off-chain pre-computation module performs some homomorphic operations locally, reducing on-chain computational overhead; this is key to improving performance. Before performing complex collaborative computations (such as large-scale matrix operations), this module can preprocess encrypted data locally.

[0310] The transaction monitoring module tracks the status of cross-chain assets and the progress of collaborative computing tasks in real time; it provides a visual dashboard that allows data providers to view in real time when their registered crypto assets are crossed across chains, by whom they are invoked, which calculations they participate in, and the resulting revenue, thus enhancing the system's observability and trustworthiness.

[0311] The smart contract interface adaptation layer supports standardized interaction with smart contracts on different blockchain platforms. By defining a unified API interface, it shields the differences in smart contracts from different source blockchains (such as Ethereum, FISCO BCOS, and Fabric), enabling data providers to interact with various blockchains in a unified manner, thus reducing the complexity of access and maintenance.

[0312] In embodiments of the present invention, the smart contract system of the target-side collaborative blockchain further includes:

[0313] The computation task scheduling module dynamically allocates on-chain computing resources based on data size and computational complexity; this is crucial for ensuring system performance and fairness. It acts like a "scheduler," dynamically allocating on-chain computing resources (Gas Limit, execution time slice) based on the urgency of the tasks, the required data size, and computational complexity when multiple complex computation tasks are received, preventing network congestion and ensuring high system throughput.

[0314] The result preservation module uses a combination of encrypted hashes and timestamps to preserve the calculation results in an tamper-proof manner. It not only stores the encrypted calculation results but also calculates the hash value (such as SHA-256) of the result and records it along with an authoritative timestamp. Any tampering with the stored results will cause a change in the hash value, thus providing court-level evidence for the integrity and existence of the data.

[0315] The audit log module records all encrypted asset calls and calculation processes, supporting compliance audits by regulatory nodes. It automatically records the complete operation pipeline: which address called which encrypted assets at what time, what calculations were performed, what results were generated, and what amount of settlement was triggered. This provides regulatory agencies with a transparent "God's-eye view," greatly simplifying the compliance audit process.

[0316] The embodiments disclosed in this invention are preferred embodiments, but are not limited thereto. Those skilled in the art can easily understand the spirit of this invention based on the above embodiments and make different extensions and variations, but as long as they do not depart from the spirit of this invention, they are all within the protection scope of this invention.

Claims

1. A multi-party marketing data privacy collaboration method based on cross-chain technology, characterized in that, The method is implemented based on a multi-party marketing data privacy collaboration system based on cross-chain technology, and the multi-party marketing data privacy collaboration method based on cross-chain technology includes: The data provider node locally extracts features from the raw marketing data to generate feature vectors, and then encrypts the feature vectors using a homomorphic encryption algorithm to obtain encrypted feature data. The data provider node registers encrypted feature data and data description information including data feature type, statistical distribution label and industry traceability identifier as encrypted assets to the source blockchain through the first cross-chain transaction; The cross-chain intermediary service listens to transaction events on the source blockchain. After verifying the legality of the first cross-chain transaction with zero-knowledge proof, it relays the encrypted feature data and data description information to the target collaborative blockchain according to the cross-chain protocol. The data user node submits a collaborative computing request to the target side collaborative blockchain. The request includes the preset collaborative computing logic to be invoked and the cryptographic asset identifiers participating in the computing. The target side's collaborative blockchain smart contract responds to the request, calls the preset homomorphic operation module to perform collaborative computation on multi-source encrypted feature data, and generates encrypted computation results; The smart contract stores the encrypted computation results on the target side's collaborative blockchain and triggers a token settlement mechanism based on preset rules; The data user node obtains the encrypted computation results from the target side collaborative blockchain, and decrypts the complete private key by combining the private key components securely distributed by multiple parties, thus obtaining readable collaborative analysis results. The multi-party marketing data privacy collaboration method based on cross-chain technology includes the following steps: S1: Key System Initialization and Distribution Operating entities: Key management center, data provider node, and data user node; Process description: The key management center generates globally homomorphic cryptographic public keys using the Secure Multi-Party Computation (SMPC) protocol. and N private key components The threshold condition K≥N (K is the minimum number of private key components required for decryption) is satisfied. Global public key The broadcast is sent to all data provider nodes, and the private key component is distributed to N data user nodes via a secure channel. Data provider nodes verify the validity of the public key and store it locally, while data user nodes confirm the integrity of the private key component. S2: Raw Data Feature Extraction and Encryption Operating entity: Data provider node; Process description: The data provider performs feature engineering on the raw marketing data (such as user behavior, purchase records, etc.) locally: Cleaning and noise reduction: Remove outliers and missing values; Feature transformation: Converting unstructured data into structured features (such as One-Hot encoding, standardization). Generate m-dimensional feature vectors: Use global public key Perform homomorphic encryption on the feature vector: Encryption function Satisfies additive homomorphism: Generate encrypted feature data: Generate data description information M, including: Feature metadata: feature type (continuous / discrete), dimension range; Quality indicators: data completeness score q∈[0,1], sample size n; Security identifiers: industry traceability code, usage permission attribute set; S3: Source Blockchain Asset Registration Operating entities: data provider nodes and source-side blockchain; Process description: Data provider builds the first cross-chain transaction Transaction Structure: =(sender address, , (Digital signature, timestamp); Digital signatures use the ECDSA algorithm: Private key provider submits On the source blockchain, nodes verify the legality of transactions: Verify signature validity: Verify data format compliance (feature dimensions, completeness of descriptive information); Once verification is successful, the source blockchain generates a unique identifier for the encrypted asset: ( ∥Block Height∥ (hash); Will Linked storage completes asset on-chain recording; S4: Cross-chain data relay and verification Operating entities: cross-chain intermediary services, target-side collaborative blockchain; Process description: Cross-chain intermediary services monitor source blockchain transaction events and capture newly registered crypto assets: Obtain the latest block from the source chain through the block header synchronization mechanism; Parse the block transaction list and extract transaction records containing crypto assets; Execute the cross-chain verification process: Zero-knowledge proof verification: generating proof of transaction validity This proves that the transaction has not been tampered with and complies with the rules; Consensus cluster verification: M verification nodes use the PBFT algorithm to verify the proof. To reach a consensus, the consensus result must be agreed upon by ≥2 / 3M nodes. Block hash verification: Verify the source chain block header hash and transaction Merkel proof to confirm the authenticity of the data source; Data format conversion and relay: The cross-chain protocol gateway converts the source chain asset format into a target chain-compatible format (preserving the ciphertext state) to generate a second cross-chain transaction. , including the converted Will Submit to the target-side collaborative blockchain to complete cross-chain asset registration; S5: Collaborative Computing Request Submission and Permission Verification Operating entities: data user nodes and target-side collaborative blockchain smart contracts; Process description: Data users can query available encrypted assets: Submit query conditions through the smart contract's attribute encryption query interface; Provide attribute proof θ to obtain a list of data description information that meets the criteria; Construct a collaborative computing request R: Request content: R=(requester address, (Deposit amount, signature) in Preset collaborative computing logic identifiers (such as "logistic regression training" and "user similarity calculation"); Smart contract permission verification: Verify whether the requester has the corresponding computing permissions (match the permission identifier in the data description information); Lock the requested amount of tokens as a deposit to ensure that computing fees are paid; Verify the status of the crypto assets involved in the computation (whether they are available or locked); S6: Ciphertext Domain Collaborative Computation Execution Operating entities: target-side collaborative blockchain smart contracts and computing engines; Process description: Smart contracts call pre-defined computational logic Load the corresponding encrypted feature data: Retrieve from on-chain storage: Initialize the computing engine (homogeneous operation module) and allocate on-chain computing resources; Performing encrypted domain collaborative computation (taking a typical scenario as an example): Scenario 1: Distributed Model Training Initialize model parameters Iterative execution: Calculate the encryption gradient: Update encryption parameters: ( (learning rate) Scenario 2: Feature Similarity Calculation Calculate the encrypted cosine similarity: Generate encrypted calculation results This includes the calculation output, intermediate process hash, and calculation timestamp; S7: Result Notarization and Automatic Token Settlement Operating entity: Target-side collaborative blockchain smart contract; Process description: Evidence of calculation results: Result hash generated: =SHA256( / / Block hash / / / Request ID); Will and Write it to the blockchain, linking the requester's address with the computation task ID; Record the audit log of the calculation process (participating assets, calculation steps, resource consumption); Token settlement mechanism execution: Calculate the contribution weight of each data provider: ( (The feature importance coefficient is preset by the smart contract). Distribute tokens: Data provider i obtains: =(1−α)·deposit· ; Cross-chain service fee: Fee = α·deposit (α∈[0.01,0.05]); The smart contract automatically executes token transfers to unlock unused deposits; S8: Decryption and Acquisition of Calculation Results Operating entity: Data user node; Process description: Data users obtain encrypted computation results from the target-side blockchain. : Query related transactions by transaction hash or task ID. Verification result hash Consistency with on-chain records ensures that the data has not been tampered with. Private key reconstruction and decryption: Collect private key components {sk1,...,skK} from at least K data users; Reconstruct the complete private key using Lagrange interpolation: Decrypt using the private key: Obtain readable collaborative analysis results; Result verification: The reasonableness of the decryption result was verified locally.

2. The multi-party marketing data privacy collaboration method based on cross-chain technology according to claim 1, characterized in that, The homomorphic encryption algorithm employs either partial or full homomorphic encryption mechanisms and is executed before encryption: A trusted initialization node generates a global public key and N private key components through a secret sharing protocol, where N is the number of data user nodes; The global public key is broadcast to all data provider nodes, and the N private key components are distributed to the corresponding N data user nodes through a secure channel. Any data user node needs to collect at least K private key components (K≤N) to reconstruct the complete private key. The value of K is preset by the initialization node.

3. The multi-party marketing data privacy collaboration method based on cross-chain technology according to claim 1, characterized in that, The preset collaborative computing logic includes: The distributed gradient descent training module based on encrypted feature vectors supports parameter updates for linear regression, logistic regression, or neural network models in encrypted form. The encrypted statistical analysis module for multi-source data can perform weighted summation, covariance matrix calculation, and quantile statistics in the encrypted domain; The feature similarity calculation module can calculate the cosine similarity or Euclidean distance between encrypted vectors without revealing the original features.

4. The multi-party marketing data privacy collaboration method based on cross-chain technology according to claim 1, characterized in that, The data description information also includes a data quality score and a usage permission identifier, and the method further includes: The smart contract of the target-side collaborative blockchain provides an attribute-encrypted query interface, and data user nodes obtain data description information that meets the conditions by submitting attribute proofs; The smart contract automatically verifies the calling permissions of the data user based on the usage permission identifier in the data description information, and only allows collaborative computing requests that meet the permission requirements to be executed.

5. The multi-party marketing data privacy collaboration method based on cross-chain technology according to claim 1, characterized in that, The token settlement mechanism includes: The smart contract locks a preset number of tokens in the data-using node when a collaborative computing request is submitted; The contribution weights of each crypto asset involved in the calculation are determined by a combination of data feature importance, data volume, and quality score. After the calculation is completed, tokens are automatically distributed to the cross-chain accounts of the corresponding data provider nodes according to the contribution weight, and a preset proportion of cross-chain service fees are deducted.

6. The multi-party marketing data privacy collaboration method based on cross-chain technology according to claim 1, characterized in that, The cross-chain intermediary service adopts a hybrid cross-chain architecture, including: A consensus cluster consisting of M verification nodes uses a practical Byzantine fault-tolerant algorithm to reach consensus on cross-chain messages. The cross-chain verification module verifies the legitimacy of the source of encrypted feature data by verifying the block header hash and transaction Merkel proof of the source blockchain. The data conversion module converts the encrypted asset format of the source blockchain into a format compatible with the target collaborative blockchain, while maintaining the encrypted state.

7. A multi-party marketing data privacy collaboration system based on cross-chain technology according to claim 1, characterized in that, include: Data provider node cluster: Deploys clients with feature engineering modules and homomorphic encryption engines to generate encrypted feature data and initiate cross-chain asset registration; Cross-chain relay network: includes a validator node pool and a cross-chain protocol gateway. The validator node pool verifies cross-chain transactions through a consensus mechanism, and the cross-chain protocol gateway realizes the conversion of encrypted data formats between different blockchains. Target-side collaborative blockchain: Deploy a smart contract system that includes a computing engine, a permission management module, and a token settlement module, wherein the computing engine supports collaborative computation of the encrypted field; Data user node cluster: Deploy clients with private key management modules and decryption engines to submit computation requests, reconstruct private keys, and decrypt computation results; Key Management Center: Generates and distributes encrypted key pairs through a secure multi-party computation protocol, and supports distributed storage and reconstruction of private key components.

8. A multi-party marketing data privacy collaboration system based on cross-chain technology according to claim 7, characterized in that, The clients of the data provider node and the data user node also integrate: The off-chain pre-computation module is used to perform some preprocessing of homomorphic operations locally, reducing on-chain computation overhead; The transaction monitoring module tracks the status of cross-chain assets and the progress of collaborative computing tasks in real time. The smart contract interface adaptation layer supports standardized interaction with smart contracts on different blockchain platforms.

9. A multi-party marketing data privacy collaboration system based on cross-chain technology according to claim 7, characterized in that, The smart contract system of the target-side collaborative blockchain also includes: The computation task scheduling module dynamically allocates on-chain computing resources based on data size and computational complexity. The result storage module uses a combination of encrypted hash and timestamp to store the calculation results in an anti-tampering manner. The audit log module records all calls to encrypted assets and the calculation process, supporting compliance audits by regulatory nodes.

Images