Privacy protection data linking method and system based on secure multi-party computing

Abstract

The present invention relates to a privacy-preserving data link method and system based on secure multi-party computing, and belongs to the technical field of privacy-preserving data link. The method of the present invention comprises: a first participant and a second participant encrypt data based on a hash function, and convert the data into hash data; either the first participant or the second participant converts the hash data into a secret share based on a secure multi-party computing protocol, retains the first share, and sends the second share to the other party; the first participant or the second participant performs data link screening under privacy protection based on the received share and the held share, and screens the serial numbers of the same entity records; based on the serial numbers of the same entity records screened out, the screening initiator obtains the screening results based on secure multi-party computing. The present invention improves the screening accuracy of the existing privacy-preserving data link method, does not require the participation of a trusted third party, and has security under a semi-honest attack model.

Description

Technical Field

[0001] The present invention relates to a privacy protection data linking method and system based on secure multi-party computing, and belongs to the technical field of privacy protection data linking. Background Art

[0002] Data is growing and accumulating rapidly, but data islands still exist. Promoting the circulation, development and utilization of data elements has become an important task in the current era, and it is urgent to carry out key technological innovations to support the safe sharing, circulation and use of data elements. Among them, privacy-preserving data linking is one of the core technologies for the current safe and effective integration of massive multi-source data sources. This technology can link data records representing the same entity in databases across organizations, institutions or systems, while protecting the privacy of entities stored in the database.

[0003] Privacy-preserving data linking technology can ensure that in the process of record linking, only the final matching results are shared between data sources, and other unmatched data record information is not leaked. Currently, this technology has been widely used in government services, healthcare, crime detection, and commercial applications. For example, in the new drug development scenario in the healthcare field, new drug developers need to integrate data from different hospitals, clinics, and pharmacies to investigate the adverse reactions of new drugs to support research and analysis. Based on privacy-preserving data linking technology, only the symptoms of patients after using the new drug need to be obtained, without obtaining other information about the patients; in crime detection scenarios, crime investigation agencies need to use relevant sensitive data from law enforcement agencies, Internet service providers, and financial institutions, and based on privacy-preserving data linking technology, only the identity information of the person corresponding to the record on the link (the identity information of the suspicious person) is obtained, without having to obtain all the data.

[0004] Currently, privacy-preserving data linking technology has been widely studied. However, in actual application, it is found that, on the one hand, the data sets of two or more parties that link data records may be inaccurate, data fields may be missing or erroneous, and a data set may contain multiple very similar data records, which greatly affects the accuracy of the link. On the other hand, in order to achieve secure linking or querying of data between two or more parties without sharing the original data, it is necessary to use encryption and other related technical means or introduce a trusted third party to achieve data comparison and linking between two or more parties, which results in extremely low efficiency and is difficult to promote to application scenarios with massive multi-source data. Summary of the invention

[0005] The present invention provides a privacy-preserving data linking method and system based on secure multi-party computing. The present invention solves the problem that most of the existing privacy-preserving data linking methods require the participation of a trusted third party, and the screening accuracy is low in scenarios where the original data is missing, erroneous, and duplicated. The present invention realizes privacy-preserving data linking without the participation of a third party in a semi-honest scenario based on secure multi-party computing, and can be applied to a variety of data screening scenarios.

[0006] The technical solution of the present invention is: in the first aspect, the present invention provides a privacy protection data linking method based on secure multi-party computing, which specifically includes:

[0007] S1. The first participant and the second participant obtain a set of data to be processed and pre-process the data to be processed;

[0008] S2. The first party and the second party encrypt the pre-processed data based on a hash function, converting the pre-processed data into hash data to protect the security of the original data;

[0009] S3. Either the first party or the second party converts the hash data into a secret share based on the secure multi-party computing protocol, retains the first share, and sends the second share to the other party; neither party can recover the original hash data of the other party based on the second share received from the other party;

[0010] S4. The first party or the second party performs a data linkage screening under privacy protection based on the received shares and the held shares, screening the serial numbers of the same entity records;

[0011] S5. Based on the serial numbers of the same entity records screened out, the screening initiator obtains the screening results based on secure multi-party computing.

[0012] Furthermore, in S1, preprocessing the data to be processed specifically includes:

[0013] S11. The preprocessing of the data to be processed refers to checking the data to be processed, standardizing the missing data, data errors and data formats, and adding a sequence number to the held data in order;

[0014] S12. The data to be processed is preprocessed, missing data is uniformly replaced by the number 0, and data errors are uniformly deleted.

[0015] Furthermore, in S2, the first party and the second party encrypt data based on a hash function to convert the pre-processed data into hash data, which specifically includes:

[0016] S21. The first party and the second party select the same hash function to encrypt the preprocessed data and select a hash function that satisfies the security strength for encryption;

[0017] S22. The first party and the second party select the same hash function to encrypt the pre-processed data, select the SHA256 algorithm, and convert the pre-processed data into a SHA256 hash, specifically including:

[0018] The preprocessed data is preprocessed again, including first padding to make the final length a multiple of 512 bits, and then dividing the data into blocks of 512 bits. ;

[0019] Calculate the hash value for the block data, process the data blocks one by one, and start with a fixed initial hash value. To begin, calculate:

[0020]

[0021] in, is the compression function of SHA256, "It's mod The addition of It is The hash value of a data block, N represents the total number of data blocks;

[0022] S23. The first participant and the second participant select the same hash function to encrypt the pre-processed data, wherein the encrypted pre-processed data includes a newly added serial number, that is, the serial number of each data piece also needs to generate a hash.

[0023] Furthermore, the S3 specifically includes:

[0024] S31. Any one of the first party and the second party converts the hash data into a secret share based on a secure multi-party computing protocol, which means that any one of the first party and the second party uses the same secure multi-party computing protocol to perform calculations and converts the held hash data into two shares;

[0025] S32. Either the first party or the second party uses the same secure multi-party computing protocol to perform calculations, including using an arithmetic secret sharing protocol and a Boolean secret sharing protocol to perform calculations:

[0026] The calculation using the arithmetic secret sharing protocol means that either the first participant or the second participant will Integer on Split into two shares and , the first share Hold it yourself and give the second share Sent to another party, where ; Arithmetic secret sharing has additive homomorphism, that is,

[0027]

[0028] in, and Represent two integers respectively. and for The two secret shares of and for Two secret shares of;

[0029] The Boolean secret sharing protocol is used for calculation, which means that either the first party or the second party splits the data through XOR operation, that is, , ,in and is the secret share obtained based on the Boolean secret sharing protocol. Hold it yourself and give the second share Send to the other party to complete share calculation and sending;

[0030] S33. Any of the first party and the second party uses the same secure multi-party computing protocol to perform calculations, including splitting the serial number of each piece of data into shares to protect the security of the serial number information.

[0031] Furthermore, the S4 specifically includes:

[0032] S41. The first participant or the second participant performs data link screening under privacy protection based on the received shares and the held shares, which means that the first participant selects the first data held by the first participant and performs similarity screening with all the data held by the second participant;

[0033] S42. The similarity screening means that the first participant selects the first data and performs similarity calculation between the first data share and the second share of all data received from the second participant, so that the first participant obtains the calculation result of the similarity between the first data and all data of the second participant;

[0034] S43. The first participant traverses and calculates all the data held by the first participant according to the steps in S42 to obtain the similarity screening results of all the data of the first participant and the second participant.

[0035] Furthermore, the S5 specifically includes:

[0036] S51. The serial number based on screening out the same entity record refers to obtaining the serial number of the record of the same entity of the first participant and the second participant based on the calculation in S4, and recovering the plain text of the serial number based on the secret recovery algorithm of the secure multi-party computing protocol used;

[0037] S52. The screening initiator obtains the screening result based on the serial number of the same entity record based on the secure multi-party computing, which means that the screening initiator obtains the data information of the required entity based on the serial number obtained in S51 and the secure multi-party computing protocol;

[0038] S53. The shared acquisition of the required entity data information based on the secure multi-party computing protocol means that the screening initiator obtains the data information of the entity to be queried based on the secret share of the same entity data that has been obtained and the secret recovery algorithm based on the secure multi-party computing protocol, and no other data information of the entity will be obtained except the information to be queried.

[0039] In a second aspect, the present invention provides a privacy-preserving data linking system based on secure multi-party computing, the system comprising a module for executing the method of the first aspect.

[0040] In a third aspect, the present invention also provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned privacy-preserving data linking method based on secure multi-party computing when executing the program.

[0041] In a fourth aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned privacy-preserving data linking method based on secure multi-party computing.

[0042] In a fifth aspect, the present invention also provides a computer program product, including a computer program, which, when executed by a processor, implements the above-mentioned privacy-preserving data linking method based on secure multi-party computing.

[0043] The beneficial effects of the present invention are as follows: the present invention solves the problem that most of the existing privacy-preserving data linking methods require the participation of a trusted third party, and the screening and matching accuracy is low in scenarios where the original data is missing, erroneous, and duplicated. The present invention realizes privacy-preserving data linking without the participation of a third party in a semi-honest scenario based on secure multi-party computing, and can be applied to a variety of data screening scenarios; the method proposed in the present invention is applicable to two-party and multi-party scenarios without loss of generality, and a two-party scenario is taken as an example in the introduction of the method. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0045] Figure 1 It is a flow chart of a privacy protection data linking method based on secure multi-party computing provided by the present invention;

[0046] Figure 2 It is a module connection diagram of the privacy protection data link system based on secure multi-party computing provided by the present invention;

[0047] Figure 3 It is a structural schematic diagram of the electronic device provided by the present invention.

[0048] Reference numerals:

[0049] 110: Hash function-based encryption module; 120: Secure multi-party computing module; 130: Data link screening module; 140: Screening result recovery module.

[0050] 310: processor; 320: communication interface; 330: memory; 340: communication bus. DETAILED DESCRIPTION

[0051] Example 1: Figure 1-Figure 3 As shown in FIG. 1 , the privacy-preserving data linking method based on secure multi-party computing specifically includes:

[0052] S1. The first participant and the second participant obtain a set of data to be processed and pre-process the data to be processed;

[0053] Preprocess the data to be processed, including:

[0054] S11. The pre-processing of the data to be processed refers to the first party and the second party checking the data to be processed, standardizing the missing data, data errors and data formats, and adding a series of serial numbers to the data in sequence; the serial number is a serial number used to identify the data held by each party;

[0055] S12. The data to be processed is preprocessed, and the missing data is uniformly replaced by the number 0, and the data errors are uniformly deleted. After completing the above standardization processing, the first participant and the second participant shuffle the data held by row to protect the real serial number information of the data from being leaked;

[0056] S2. The first party and the second party encrypt the pre-processed data based on a hash function and convert the pre-processed data into hash data to protect the security of the original data; specifically, the following steps are included:

[0057] S21. The first party and the second party select the same hash function to encrypt the preprocessed data and select a hash function that satisfies the security strength for encryption;

[0058] S22. The first party and the second party select the same hash function to encrypt the pre-processed data, select the SHA256 algorithm, and convert the pre-processed data into a SHA256 hash to protect data security, specifically including:

[0059] The preprocessed data is preprocessed again, including first padding to make the final length a multiple of 512 bits, and then dividing the data into blocks of 512 bits. ;

[0060] Calculate the hash value for the block data, process the data blocks one by one, and start with a fixed initial hash value. To begin, calculate:

[0061]

[0062] in, is the compression function of SHA256, "It's mod The addition of It is The hash value of a data block, N represents the total number of data blocks;

[0063] S23. The first participant and the second participant select the same hash function to encrypt the pre-processed data, wherein the encrypted pre-processed data includes a newly added serial number, that is, the serial number of each data piece also needs to generate a hash.

[0064] All data held by the first party and the second party are converted into SHA256 hashes by the above SHA256 algorithm.

[0065] S3. Either the first party or the second party converts the hash data into a secret share based on the secure multi-party computing protocol, retains the first share, and sends the second share to the other party; neither party can recover the original hash data of the other party based on the second share received from the other party;

[0066] The S3 specifically includes:

[0067] S31. Any one of the first party and the second party converts the hash data into a secret share based on a secure multi-party computing protocol, which means that any one of the first party and the second party uses the same secure multi-party computing protocol to perform calculations and converts the held hash data into two shares;

[0068] S32. To better support the subsequent screening link algorithm, any one of the first party and the second party uses the same secure multi-party computing protocol for calculation, including using an arithmetic secret sharing protocol and a Boolean secret sharing protocol for calculation:

[0069] The calculation using the arithmetic secret sharing protocol means that either the first participant or the second participant will Integer on Split into two shares and , the first share Hold it yourself and give the second share Sent to another party, where ; Arithmetic secret sharing has additive homomorphism, that is,

[0070]

[0071] in, and Represent two integers respectively. and for The two secret shares of and for Two secret shares of;

[0072] The Boolean secret sharing protocol is used for calculation, which means that either the first party or the second party splits the data through XOR operation, that is, , ,in and is the secret share obtained based on the Boolean secret sharing protocol. Hold it yourself and give the second share Send to the other party to complete share calculation and sending;

[0073] By using the arithmetic secret sharing protocol and the Boolean secret sharing protocol, the data sequence number and the held data of the first and second participants can be converted into secret shares. In addition, in order to improve the calculation efficiency, the hash data of SHA256 can be further processed. For each data block, 12 hexadecimal data can be selected from the 64-bit hexadecimal hash data for calculation. There is a probability of 1 / (16^12) that a collision will occur, so the probability of a collision with more than 500,000 samples can be ignored.

[0074] S33. Any of the first party and the second party uses the same secure multi-party computing protocol to perform calculations, including splitting the serial number of each piece of data into shares to protect the security of the serial number information.

[0075] S4. The first participant or the second participant performs data link screening under privacy protection based on the received shares and the held shares, and screens the serial numbers of the same entity records; S4 specifically includes:

[0076] S41. The first participant or the second participant performs data link screening under privacy protection based on the received shares and the held shares, which means that the first participant selects the first data held by the first participant and performs similarity screening with all the data held by the second participant;

[0077] S42. The similarity screening means that the first participant selects the first data and performs similarity calculation between the first data share and the second share of all data received from the second participant, so that the first participant obtains the calculation result of the similarity between the first data and all data of the second participant;

[0078] The first participant selects the first piece of data after reordering , assuming the first piece of data contains Block, that is Without loss of generality, assume The data block represents the first data of the first participant. feature items, and different feature items represent different specific meanings, such as the name, ID number, mobile phone number, address, etc. of a personal entity. Assume that the second participant has M data, that is, ,in Therefore, the first data of the first participant and the second data of the second participant The formula for calculating the similarity of data is:

[0079]

[0080] The second party The similarity calculation formula between the first data of the first participant and the first data of the first participant is:

[0081]

[0082] Based on the above calculation results and , the secret recovery algorithm based on secure multi-party computing can obtain the first data of the first participant and the second data of the second participant. The similarity matching results of the data are as follows:

[0083]

[0084] Through the above calculation formula, the first data of the first participant and all M data of the second participant are calculated similarly to obtain M similarity results:

[0085]

[0086] By repeating the above steps, the first participant can use the N pieces of data held by the convenience to complete the similarity screening calculation of the two parties' data and obtain the serial number of the final matching result.

[0087] Another preferred method for increasing the accuracy of data link screening is to assign different weights according to the error rate, missing rate and importance of different block data, so that the first data of the first participant and the second data of the second participant are The formula for calculating the similarity of data is:

[0088]

[0089] in, To indicate the The weight of each data block is set according to the importance of the data and the error rate.

[0090] The second party The similarity calculation formula between the first data of the first participant and the first data of the first participant is:

[0091]

[0092] in, To indicate the The weight of each data block is set according to the importance of the data and the error rate, and meets the .

[0093] S43. The first participant traverses and calculates all the data held by the first participant according to the steps in S42 to obtain the similarity screening results of all the data of the first participant and the second participant.

[0094] S5. Based on the sequence numbers of the same entity records screened out, the screening initiator obtains the screening results based on secure multi-party computing. S5 specifically includes:

[0095] S51. The serial number based on screening out the same entity record refers to obtaining the serial number of the record of the same entity of the first participant and the second participant based on the calculation in S4, and recovering the plain text of the serial number based on the secret recovery algorithm of the secure multi-party computing protocol used;

[0096] S52. Based on the serial numbers of the same entity records screened out, the screening initiator obtains the screening results based on secure multi-party computing, which means that the screening initiator obtains the data information of the required entity based on the serial number obtained in S51 and the secure multi-party computing protocol;

[0097] S53. The shared acquisition of the required entity data information based on the secure multi-party computing protocol means that the screening initiator obtains the data information of the entity to be queried based on the secret share of the same entity data that has been obtained and the secret recovery algorithm of the secure multi-party computing protocol, and no other data information of the entity except the information to be queried will be obtained.

[0098] like Figure 2 As shown, the present invention also provides a privacy protection data link system based on secure multi-party computing, the system comprising:

[0099] The hash function-based encryption module 110 is used by the first participant and the second participant to encrypt the pre-processed data based on the hash function, and convert the pre-processed data into hash data;

[0100] A secure multi-party computing module 120, configured for any one of the first party and the second party to convert the hash data into a secret share based on a secure multi-party computing protocol, retain the first share, and send the second share to the other party;

[0101] A data link screening module 130, for the first participant or the second participant to perform data link screening under privacy protection based on the received shares and the held shares, and screen the serial numbers of the same entity records;

[0102] The screening result recovery module 140 is used for obtaining the screening result based on the serial number of the same entity record screened out by the screening initiator based on secure multi-party computing.

[0103] In a feasible implementation, the hash function-based encryption module 110 is also used for the first participant and the second participant to obtain a data set to be processed and pre-process the data to be processed.

[0104] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned privacy protection data linking method based on secure multi-party computing when executing the program.

[0105] Specifically, Figure 3 An example of a physical structure diagram of an electronic device is provided, and the electronic device may include: a processor 310, a communication interface 320, a memory 330, and a communication bus 340, wherein the processor 310, the communication interface 320, and the memory 330 communicate with each other through the communication bus 340. The processor 310 may call the logic instructions in the memory 330 to execute the privacy protection data linking method based on secure multi-party computing provided in the present embodiment;

[0106] On the other hand, the present invention further provides a computer program product, the computer program product includes a computer program, the computer program can be stored on a non-transitory computer-readable storage medium, when the computer program is executed by a processor, the computer can execute the privacy protection data linking method based on secure multi-party computing provided by the above methods;

[0107] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the privacy-preserving data linking method based on secure multi-party computing provided by the above-mentioned methods.

[0108] In order to illustrate the effect of the present invention, the present invention conducted a comparative experiment to verify the improvement of the accuracy of the method of the present invention.

[0109] The idash2022 competition question 4 data set is collected and used as experimental data. The first participant holds data set A, including 500,000 data (processed by SHA256); the second participant holds data set B, including 500,000 data (processed by SHA256); there are 10 columns in the two data sets, that is, 10 features, such as last name, first name, gender, ID number, birthday, email, telephone, address, etc. Data sets A and B have 10% of the data in common, and except for gender and ID number, other features may have different error rates (2%~35%), and there are missing values ​​to varying degrees (the missing values ​​are also hashed).

[0110] The first participant and the second participant preprocess the data in the hashed data sets A and B respectively, identify missing values, and replace them with 0;

[0111] The first participant and the second participant select the same random rule respectively, and randomly select 12 hexadecimal digits from each 64-bit hexadecimal feature of each data, so as to reduce the length of screening and comparison and improve computing efficiency. A collision will occur with a probability of 1 / (16^12), so the probability of a collision with more than 500,000 samples can be ignored (longer data feature bits can be selected as the data volume increases).

[0112] The first party and the second party respectively convert the data held by the first party and the second party (the characteristics of each piece of data are 12-digit hexadecimal) into secret shares through the Boolean secret sharing protocol, and send the second share to the other party;

[0113] According to the different error rates of each feature (2%~35%), the present invention proposes to give different weights to different features. , indicating the The importance of each data block (or feature) (the lower the error rate, the higher the importance), without loss of generality, the first participant obtains the familiarity of each data by calculating the following expression:

[0114]

[0115] in, Indicates the first data of the first participant and the second data of the second participant The similarity of the data. If the calculation result is 0, it means similarity. If the calculation result is not equal to 0, it means dissimilarity. The closer the calculation result is to 0, the more similar it is. Indicates the first data of the first participant The first share of data features, Indicates that the second party Data No. The second share of the data feature;

[0116] The first participant and the second participant each hold 500,000 pieces of data, each piece of data has 10 feature columns, and the screening and comparison are performed under plain text conditions and cipher text conditions, respectively. The comparison under plain text conditions indicates that the first participant and the second participant are aggregated for screening and comparison, and the comparison under cipher text indicates that the original data information of the first participant and the second participant is not leaked, and the data is not aggregated, and is held by the two parties respectively. Among them, the present invention proposes a new comparison method based on error rate weighting, which improves the screening accuracy compared with the method without weighting, and is close to the calculation result under plain text (the screening accuracy does not reach 100% because in order to improve the calculation efficiency, the data features are cropped, and only 12 hexadecimal digits of each feature are selected).

[0117] Table 1 shows the screening comparison accuracy

[0118]

[0119] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0120] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0121] It should be noted that, in the present invention, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the term "comprises" or any other variant thereof is intended to cover non-exclusive inclusion, so that an article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such article or device. In the absence of further restrictions, the elements defined by the sentence "comprises..." do not exclude the existence of other identical elements in the article or device including the elements.

[0122] 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 make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A privacy-preserving data linking method based on secure multi-party computing, characterized by: include: S1. The first participant and the second participant obtain a set of data to be processed and pre-process the data to be processed; S2. The first party and the second party encrypt the pre-processed data based on a hash function, and convert the pre-processed data into hash data; S3. Either the first party or the second party converts the hash data into a secret share based on a secure multi-party computing protocol, retains the first share, and sends the second share to the other party; S4. The first party or the second party performs a data linkage screening under privacy protection based on the received shares and the held shares, screening the serial numbers of the same entity records; S5. Based on the serial numbers of the same entity records screened out, the screening initiator obtains the screening results based on secure multi-party computing; In S2, the first party and the second party encrypt data based on a hash function and convert the pre-processed data into hash data, which specifically includes: S21. The first party and the second party select the same hash function to encrypt the preprocessed data and select a hash function that satisfies the security strength for encryption; S22. The first party and the second party select the same hash function to encrypt the pre-processed data, select the SHA256 algorithm, and convert the pre-processed data into a SHA256 hash, specifically including: The preprocessed data is preprocessed again, including first padding to make the final length a multiple of 512 bits, and then dividing the data into blocks of 512 bits. ; Calculate the hash value for the block data, process the data blocks one by one, and start with a fixed initial hash value. To begin, calculate: ； in, is the compression function of SHA256, "It's mod The addition of It is The hash value of a data block, N represents the total number of data blocks; S23. The first party and the second party select the same hash function to encrypt the pre-processed data, wherein the encrypted pre-processed data includes a newly added serial number, and the serial number of each piece of data also generates a hash hash; The S3 specifically includes: S31. Any one of the first party and the second party converts the hash data into a secret share based on a secure multi-party computing protocol, which means that any one of the first party and the second party uses the same secure multi-party computing protocol to perform calculations and converts the held hash data into two shares; S32. Either the first party or the second party uses the same secure multi-party computing protocol to perform calculations, including using an arithmetic secret sharing protocol and a Boolean secret sharing protocol to perform calculations: The calculation using the arithmetic secret sharing protocol means that either the first participant or the second participant will Integer on Split into two shares and , the first share Hold it yourself and give the second share Sent to another party, where ; Arithmetic secret sharing has additive homomorphism; ； in, and Represent two integers respectively. and for The two secret shares of and for Two secret shares of; The Boolean secret sharing protocol is used for calculation, which means that either the first party or the second party splits the data through XOR operation. , ,in and is the secret share obtained based on the Boolean secret sharing protocol. Hold it yourself and give the second share Send to the other party to complete share calculation and sending; S33. Any one of the first party and the second party uses the same secure multi-party computing protocol to perform calculations, including splitting the serial number of each piece of data into shares to protect the security of the serial number information; The S4 specifically includes: S41. The first participant or the second participant performs data link screening under privacy protection based on the received shares and the held shares, which means that the first participant selects the first data held by the first participant and performs similarity screening with all the data held by the second participant; S42. The similarity screening means that the first participant selects the first data and performs similarity calculation between the first data share and the second share of all data received from the second participant, so that the first participant obtains the calculation result of the similarity between the first data and all data of the second participant; The first data of the first participant and the second data of the second participant Similarity of data The calculation formula is as follows: ; in, Indicates the first data of the first participant and the second data of the second participant The similarity of the data. If the calculation result is 0, it means similarity. If the calculation result is not equal to 0, it means dissimilarity. The closer the calculation result is to 0, the more similar it is. Indicates the first data of the first participant The first share of data features, Indicates that the second party Data No. The second share of data features, To indicate the The weight of each data block is set according to the importance and error rate of the data; The second party The similarity calculation formula between the first data of the first participant and the first data of the first participant is: ； in, To indicate the The weight of each data block is set according to the importance and error rate of the data; S43. The first participant traverses and calculates all the data held by the first participant according to the steps in S42 to obtain a similarity screening result of all the data of the first participant and the second participant.

2. The privacy-preserving data linking method based on secure multi-party computing according to claim 1 is characterized in that: In S1, the data to be processed is preprocessed, specifically including: S11. The preprocessing of the data to be processed refers to checking the data to be processed, standardizing the missing data, data errors and data formats, and adding a sequence number to the held data in order; S12. The data to be processed is preprocessed, missing data is uniformly replaced by the number 0, and data errors are uniformly deleted.

3. The privacy-preserving data linking method based on secure multi-party computing according to claim 1, characterized in that: The S5 specifically includes: S51. The serial number based on screening out the same entity record refers to obtaining the serial number of the record of the same entity of the first participant and the second participant based on the calculation in S4, and recovering the plain text of the serial number based on the secret recovery algorithm of the secure multi-party computing protocol used; S52. Based on the serial numbers of the same entity records screened out, the screening initiator obtains the screening results based on secure multi-party computing, which means that the screening initiator obtains the data information of the required entity based on the serial number obtained in S51 and the secure multi-party computing protocol; S53. The shared acquisition of the required entity data information based on the secure multi-party computing protocol means that the screening initiator obtains the data information of the entity to be queried based on the secret share of the same entity data that has been obtained and the secret recovery algorithm of the secure multi-party computing protocol, and no other data information of the entity except the information to be queried will be obtained.

4. A privacy-preserving data link system based on secure multi-party computing, characterized in that: It includes a module for executing the privacy-preserving data linking method based on secure multi-party computing as described in any one of claims 1 to 3.

5. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the privacy-preserving data linking method based on secure multi-party computing as described in any one of claims 1 to 3 is implemented.

6. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the privacy-preserving data linking method based on secure multi-party computing as described in any one of claims 1 to 3 is implemented.

7. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the privacy-preserving data linking method based on secure multi-party computing as described in any one of claims 1 to 3 is implemented.

Images