CN108022654B - Association rule mining method and system based on privacy protection and electronic equipment - Google Patents
Association rule mining method and system based on privacy protection and electronic equipment Download PDFInfo
- Publication number
- CN108022654B CN108022654B CN201711391275.2A CN201711391275A CN108022654B CN 108022654 B CN108022654 B CN 108022654B CN 201711391275 A CN201711391275 A CN 201711391275A CN 108022654 B CN108022654 B CN 108022654B
- Authority
- CN
- China
- Prior art keywords
- mining
- data set
- association rule
- mined
- cloud server
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/60—Protecting data
- G06F21/62—Protecting access to data via a platform, e.g. using keys or access control rules
- G06F21/6218—Protecting access to data via a platform, e.g. using keys or access control rules to a system of files or objects, e.g. local or distributed file system or database
- G06F21/6245—Protecting personal data, e.g. for financial or medical purposes
- G06F21/6263—Protecting personal data, e.g. for financial or medical purposes during internet communication, e.g. revealing personal data from cookies
Abstract
The present application relates to the field of data mining technologies, and in particular, to an association rule mining method and system based on privacy protection, and an electronic device. The method comprises the following steps: the method comprises the steps that a sender conducts searchable encryption on a data set to be mined and uploads the encrypted data set to be mined to a cloud server; the sending party sends an association rule mining request to the cloud server and submits a mining keyword to the cloud server; and the cloud server side performs association rule mining on the encrypted data set to be mined by using an association rule mining algorithm based on differential privacy protection according to the mining key words. According to the method and the device, the data to be mined are encrypted by using the symmetric searchable encryption algorithm, the safety of data storage is guaranteed, and meanwhile, the association rule mining method of differential privacy protection is used for mining the association rule of the data to be mined, so that the privacy of the data in the whole flow in the data mining process, the intermediate caching process and the later result publishing process is guaranteed.
Description
Technical Field
The present application relates to the field of data mining technologies, and in particular, to an association rule mining method and system based on privacy protection, and an electronic device.
Background
With the development of society, people pay more attention to their health when enjoying the convenience brought by the development of science and technology and the progress of society. Along with the development of big data technology, the big medical data of the dust seal in the hospital is paid more attention again, and due to the particularity of the big medical data, people can dig out a lot of new and useful information from the big medical data of the dust seal through means such as data mining and the like for developing accurate auxiliary diagnosis and treatment.
However, because of the extremely high sensitivity of medical big data, how to guarantee the privacy of the medical big data in data mining is an important challenge. Chapter iii of "study of association rule mining algorithm for privacy protection based on MapReduce" by stamen, bear mentions privacy protection of data to be mined. Conhai swallow, section 2.1 of "review of differential privacy protection for application in data mining" mentions and compares various differential privacy based schema mining. Chapter iv of MapReduce-based symmetric searchable encryption scheme, by the party, mentions a MapReduce-based symmetric searchable encryption scheme.
Association rule mining (Association rule mining) is one of the most active research methods in data mining and can be used to discover the connections between things. For example, the rule { hypertension, heart disease } → { stroke } found in the medical data would indicate that if the patient had both hypertension and heart disease, the patient also had a stroke. Such rules may serve as a basis for making a physician's diagnosis and prognosis of a patient's disease for early prevention. However, the existing association rule mining technology mainly focuses on the performance problem in medical data mining, but does not solve the problem of how to guarantee the privacy of the medical data.
Disclosure of Invention
The application provides an association rule mining method, an association rule mining system and electronic equipment based on privacy protection, and aims to solve at least one of the technical problems in the prior art to a certain extent.
In order to solve the above problems, the present application provides the following technical solutions:
an association rule mining method based on privacy protection comprises the following steps:
step a: the method comprises the steps that a sender conducts searchable encryption on a data set to be mined and uploads the encrypted data set to be mined to a cloud server;
step b: the sending party sends an association rule mining request to the cloud server and submits a mining keyword to the cloud server;
step c: and the cloud server side performs association rule mining on the encrypted data set to be mined by using an association rule mining algorithm based on differential privacy protection according to the mining key words.
The technical scheme adopted by the embodiment of the application further comprises the following steps: before the step a, the method further comprises the following steps: a sending party submits a registration request to a cloud server, the cloud server creates a corresponding storage area for the sending party according to the registration request, and a server for providing data service for the sending party is started; the storage area is used for storing the user information of the data set to be mined uploaded by the sender.
The technical scheme adopted by the embodiment of the application further comprises the following steps: in the step a, the searchable encryption of the data set to be mined by the sender further includes: and the sender cleans the data to be mined, extracts the user ID and the user information keywords in the data to be mined and forms a data set to be mined.
The technical scheme adopted by the embodiment of the application further comprises the following steps: in the step a, the searchable encryption of the data set to be mined by the sender specifically comprises: encrypting all user information keywords in the data set to be mined by utilizing a searchable encryption algorithm; the method for encrypting the data set to be mined by the sender and uploading the encrypted data set to be mined to the cloud server specifically comprises the following steps:
step a 1: pre-encrypting the data set to be mined to obtain pre-encrypted data Cpre；
Step a 2: data C to be pre-encryptedpreDividing the pseudo random sequence into n-m bits and m bits, and generating a pseudo random sequence S with n-m bits by using a password;
step a 3: generating a pseudorandom value k using a password;
step a 4: using pseudo-random sequence S as parameter, using pseudo-random functions F and F to generate m bit value to form Salt value
Step a 5: data C to be pre-encryptedpreAnd Salt value TiPerforming XOR to obtain an original ciphertext data set;
step a 6: and uploading the original ciphertext data set to a server.
The technical scheme adopted by the embodiment of the application further comprises the following steps: in step a6, uploading the original ciphertext data set to a server further includes:
step a 7: the server preprocesses the original ciphertext data set to obtain a preprocessing result table; the preprocessing is to take the user ID in the original ciphertext data set as a key value and gather the user information keywords together;
step a 8: and after the original ciphertext data set and the preprocessing result table are respectively randomly disordered by the server, the original ciphertext data set and the preprocessing result table are stored in a storage area corresponding to the sender together.
The technical scheme adopted by the embodiment of the application further comprises the following steps: in the step b, the submitting and mining keywords to the cloud server specifically comprises: a sender designates a mining keyword, encrypts the mining keyword by using a symmetrical searchable encryption algorithm to generate a mining keyword trapdoor, and submits the generated mining keyword trapdoor to a server; and the mining keywords correspond to the user information keywords.
The technical scheme adopted by the embodiment of the application further comprises the following steps: in step c, the association rule mining, by the cloud server, the encrypted data set to be mined by using an association rule mining algorithm based on differential privacy protection according to the mining keyword further includes: matching the mining keyword trapdoor with the user information keywords in the preprocessing result table, judging whether the matching is successful or not, and if the matching is successful, mining the association rule of the original ciphertext data set according to the mining keyword trapdoor; and otherwise, returning the information of failure in digging the keyword trapdoor matching to the sender.
The technical scheme adopted by the embodiment of the application further comprises the following steps: in the step c, the performing association rule mining on the encrypted data set to be mined by using an association rule mining algorithm based on differential privacy protection specifically includes:
step c 1: filtering and sequencing the original ciphertext data set according to the mined keyword trapdoor to form a new ciphertext data set D*；
Step c 2: based on the new ciphertext data set D*Constructing a frequent pattern tree, searching for frequent patterns meeting conditions and the support degree of each frequent pattern through the frequent pattern tree, and selecting a frequent pattern set Cset with the support degree counting not less than a threshold value min _ count;
step c 3: selecting k frequent pattern sets which are most prone to privacy disclosure from the frequent pattern set Cset by adopting an exponential mechanism;
step c 4: adding noise to the support counts of the k frequent pattern sets;
step c 5: carrying out consistency constraint on the support counts of the k frequent pattern sets added with the noise;
step c 6: calculating an association rule index by using a noise count set;
step c 7: and returning the association rule mining result to the sender according to the association rule index calculation result.
Another technical scheme adopted by the embodiment of the application is as follows: an association rule mining system based on privacy protection comprises terminal equipment and a cloud server;
the terminal device includes:
a data encryption module: the cloud server side is used for encrypting the data set to be mined and uploading the encrypted data set to be mined to the cloud server side;
a data mining request module: the system comprises a cloud server and a server, wherein the cloud server is used for sending an association rule mining request to the cloud server;
a keyword submission module: the system is used for submitting mining keywords to a cloud server;
the cloud server comprises:
an association rule mining module: and the association rule mining method is used for mining the association rule of the encrypted data set to be mined by utilizing an association rule mining algorithm based on differential privacy protection according to the mining keyword after receiving an association rule mining request.
The technical scheme adopted by the embodiment of the application further comprises the following steps: the terminal device further includes:
a user information registration request module: the system comprises a cloud server and a server side, wherein the cloud server is used for submitting a registration request to the cloud server;
the cloud server further comprises:
a registration module: the server is used for establishing a corresponding storage area for the terminal equipment according to the registration request and starting up a server for providing data service for the terminal equipment; the storage area is used for storing the data set to be mined.
The technical scheme adopted by the embodiment of the application further comprises the following steps: the terminal device further includes:
a data processing module: the method is used for cleaning the data to be mined, extracting the user ID and the user information keywords in the data to be mined and forming a data set to be mined.
The technical scheme adopted by the embodiment of the application further comprises the following steps: the searchable encryption of the data set to be mined by the data encryption module is specifically as follows: encrypting all user information keywords in the data set to be mined by utilizing a searchable encryption algorithm; the data encryption module specifically comprises:
a pre-encryption unit: is used for pre-encrypting the data set to be mined to obtain pre-encrypted data Cpre；
A data division unit: for pre-encrypting data CpreDividing the pseudo random sequence into n-m bits and m bits, and generating a pseudo random sequence S with n-m bits by using a password;
a pseudo-random value generation unit: for generating a pseudo-random value k using a password;
a random salt value generation unit: for generating m-bit values by using pseudo-random functions F and F with pseudo-random sequence S as a parameter to form Salt value
An exclusive-or operation unit: for pre-encrypting data CpreAnd Salt value TiAnd performing XOR to obtain an original ciphertext data set, and uploading the original ciphertext data set to a server.
The technical scheme adopted by the embodiment of the application further comprises the following steps: the cloud server further comprises:
a data preprocessing module: the system is used for preprocessing an original ciphertext data set to obtain a preprocessing result table; the preprocessing is to take the ID in the original ciphertext data set as a key value and gather the user information keywords together;
a data storage module: and the data processing device is used for respectively randomly disordering the original ciphertext data set and the preprocessing result table and storing the original ciphertext data set and the preprocessing result table in a storage area corresponding to the terminal device.
The technical scheme adopted by the embodiment of the application further comprises the following steps: the method for submitting the mining keywords to the cloud server by the keyword submitting module specifically comprises the following steps: the method comprises the steps of appointing a mining keyword, encrypting the mining keyword by using a symmetrical searchable encryption algorithm to generate a mining keyword trapdoor, and submitting the generated mining keyword trapdoor to a cloud server; and the mining keywords correspond to the user information keywords.
The technical scheme adopted by the embodiment of the application further comprises the following steps: the cloud server further comprises:
a keyword matching module: the correlation rule mining module is used for matching the mining keyword trapdoor with the user information keywords in the preprocessing result table, judging whether the matching is successful or not, and mining the correlation rule of the original ciphertext data set through the correlation rule mining module if the matching is successful; otherwise, returning the information of failure in digging keyword trapdoor matching to the terminal equipment.
The technical scheme adopted by the embodiment of the application further comprises the following steps: the association rule mining module specifically comprises:
a sorting unit: for filtering and sorting the original ciphertext data set according to the mining keyword trapdoor to form a new ciphertext data set D*；
A frequent mode selection unit: for generating a new ciphertext data set D*Constructing a frequent pattern tree, searching for frequent patterns meeting conditions and the support degree of each frequent pattern through the frequent pattern tree, and selecting a frequent pattern set Cset with the support degree counting not less than a threshold value min _ count;
a frequent pattern screening unit: the method comprises the steps that k frequent pattern sets which are most prone to privacy disclosure are selected from the frequent pattern sets Cset by adopting an exponential mechanism;
a noise addition unit: adding noise to the support counts of the k frequent pattern sets;
a consistency constraint unit: carrying out consistency constraint on the support counts of the k frequent pattern sets for adding the noise;
an association rule index calculation unit: and the correlation rule index calculation module is used for calculating the correlation rule index by using the noise count set and returning a correlation rule mining result to the terminal equipment according to the correlation rule index calculation result.
The embodiment of the application adopts another technical scheme that: an electronic device, comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores instructions executable by the one processor to cause the at least one processor to perform the following operations of the privacy preserving based association rule mining method described above:
step a: the method comprises the steps that a sender conducts searchable encryption on a data set to be mined and uploads the encrypted data set to be mined to a cloud server;
step b: the sending party sends an association rule mining request to the cloud server and submits a mining keyword to the cloud server;
step c: and the cloud server side performs association rule mining on the encrypted data set to be mined by using an association rule mining algorithm based on differential privacy protection according to the mining key words.
Compared with the prior art, the embodiment of the application has the advantages that: according to the association rule mining method and system based on privacy protection and the electronic equipment, the data to be mined are encrypted by using the symmetric searchable encryption algorithm, the data storage safety is guaranteed, meanwhile, the association rule mining method based on the differential privacy protection is used for mining the association rule of the data to be mined, and the privacy of the whole data flow in the data mining process, the middle caching process and the later result publishing process is guaranteed.
Drawings
FIG. 1 is an overall flowchart of an association rule mining method based on privacy protection according to an embodiment of the present application;
FIG. 2 is a process flow diagram for data encryption according to an embodiment of the present application;
FIG. 3 is a process flow diagram of association rule mining according to an embodiment of the present application;
FIG. 4 is a schematic structural diagram of an association rule mining system based on privacy protection according to an embodiment of the present application;
fig. 5 is a schematic structural diagram of hardware equipment of an association rule mining method based on privacy protection according to an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
Please refer to fig. 1, which is a flowchart illustrating an association rule mining method based on privacy protection according to an embodiment of the present application. The association rule mining method based on privacy protection comprises the following steps:
step a: the method comprises the steps that a sender conducts searchable encryption on a data set to be mined and uploads the encrypted data set to be mined to a cloud server;
step b: the sending party sends an association rule mining request to the cloud server and submits a mining keyword to the cloud server;
step c: and the cloud server side performs association rule mining on the encrypted data set to be mined by using an association rule mining algorithm based on differential privacy protection according to the mining key words.
Based on the above, the association rule mining method of the embodiment of the application includes two stages of data encryption and association rule mining. In the data encryption stage, after encrypting data to be mined by a sender by using a symmetric searchable encryption algorithm, uploading the encrypted data to a cloud server; in the association rule mining stage, the cloud server side performs association rule mining on the encrypted data by using an association rule mining algorithm based on differential privacy protection without decrypting the encrypted data, so that the security of mining result publication is guaranteed. The method is applicable to association rule mining of various types of big data, such as enterprise data, machine and sensor data, social data, and the like.
Specifically, please refer to fig. 2, which is a flowchart illustrating a process of data encryption according to an embodiment of the present application. The data encryption method of the embodiment of the application comprises the following steps:
step 100: a sender cleans medical data in a database, and extracts a user ID and a user information keyword in the medical data to form a data set to be mined;
in step 100, data scrubbing refers to the last procedure to find and correct recognizable errors in data files, including checking data consistency, processing invalid and missing values, etc., with the purpose of deleting duplicate information, correcting existing errors, and providing data consistency. Taking medical data as an example, the formed data set to be mined is shown in table 1 below:
TABLE 1 data set to be mined
| ID | Items |
| ID1 | F1 |
| ID1 | F2 |
| ID1 | F4 |
| ID2 | F2 |
| ID2 | F4 |
| ID3 | F3 |
In table 1, ID denotes an ID number of a user (patient), and Items denotes a disease of the user (patient), that is, a user information keyword. Specifically, the ID and Items may be determined according to the type of data to be mined.
Step 110: the sending party submits a registration request to the cloud server;
in step 110, the registration request submitted by the sender includes at least registration information such as sender ID.
Step 120: after receiving the registration request, the cloud server creates a storage area for storing data for the sender at the cloud end, starts a server for providing data service for the sender, and returns registration success information to the sender after creation is completed;
step 130: after receiving the registration success information, the sender extracts a data set to be mined and conducts searchable encryption on the data set to be mined to obtain an encrypted original ciphertext data set;
in step 130, the embodiment of the present application encrypts the to-be-mined data set by using a Searchable Encryption algorithm (SE); searchable encryption is a cryptographic primitive which is developed in recent years and supports a user to search keywords in a ciphertext, a large amount of network and calculation expenses can be saved for the user, huge calculation resources of a cloud server are fully utilized to search the keywords in the ciphertext, and privacy of user data is guaranteed.
Specifically, the encryption mode for encrypting the data set to be mined by utilizing the searchable encryption algorithm comprises the following steps:
step 131: pre-encrypting the data set to be mined to obtain pre-encrypted data Cpre；
Step 132: data C to be pre-encryptedpreDividing the pseudo random sequence into n-m bits and m bits, and generating a pseudo random sequence S with n-m bits by using a password;
step 133: generating a pseudorandom value k using a password;
step 134: using pseudo-random sequence S as parameter and pseudo-random sequenceThe machine functions F and F generate m-bit values, which constitute the Salt value (random Salt value)
Step 135: data C to be pre-encryptedpreAnd Salt value TiAnd performing XOR to obtain an original ciphertext data set C.
Step 140: the sender uploads the original ciphertext data set to a server;
step 150: the server receives an original ciphertext data set uploaded by a sender, and preprocesses the original ciphertext data set to obtain a preprocessing result table;
in step 150, preprocessing is performed to group Items together with the ID in the original ciphertext data set as the key. The pretreatment results are shown in table 2 below:
TABLE 2 results of pretreatment
| ID | TimeStamp | Items |
| ID1 | Timestamp | F1,F2,F4 |
| ID2 | Timestamp | F2,F4 |
Step 160: and after randomly disordering the sequence of the original ciphertext data set and the preprocessing result table respectively, the server stores the original ciphertext data set and the preprocessing result table together in a storage area corresponding to the sender and returns successful uploading information to the sender.
Please refer to fig. 3, which is a flowchart illustrating an association rule mining process according to an embodiment of the present application. The association rule mining method comprises the following steps:
step 200: a sender sends an association rule mining request;
step 210: the server receives the association rule mining request and reads a preprocessing result table from a storage area corresponding to the sender;
step 220: the sender designates the mining keyword, encrypts the selected mining keyword to generate a mining keyword trapdoor (if a login processing system allows a specific user identification code, the identification code can bypass the common password check, and the user can visually understand that the user can log in the system to modify and the like through a special user name and a special password.
In step 220, the selected mining keyword is encrypted by using a symmetric searchable encryption algorithm in the embodiment of the present application, where the encryption manner is the same as that of the data set to be mined, and will not be described herein again. The mining keywords correspond to user information keywords in the data set to be mined, and the mining keyword trapdoor generation mode is as follows: a sender inputs a mining keyword W, encrypts the mining keyword W into a ciphertext X, and generates a trapdoor TD by using a random function f<X,fk(Xl)>。
Step 230: the server receives the mining keyword trapdoor submitted by the sender, matches the mining keyword trapdoor with the user information keyword in the preprocessing result table, judges whether the matching of the mining keyword trapdoor is successful or not, and executes step 240 if the matching is failed; if all the matching is successful, executing step 250;
in step 230, the mining keyword trapdoors are first matched before the association rule mining, and if the matching is successful, the original ciphertext is describedThe data set comprises the mining key word trapdoor, association rule mining can be carried out, decryption on an original ciphertext data set is not needed, and the safety of data storage is guaranteed; if the matching fails, the original ciphertext data set does not contain the mining keyword trapdoor, and subsequent association rule mining cannot be carried out. Specifically, the method for mining keyword matching is as follows: reading data C from Hbase (open source database), and performing XOR on X and C in TD to obtain<S,F>Using f in TDk(Xl) Judgment of
And if not, the matching of the mining keywords is successful, and if not, the matching of the mining keywords is failed.
Step 240: returning mining keyword trapdoor matching failure information to the sender, and re-executing the step 220;
step 250: reading an original ciphertext data set from a storage area corresponding to the sender, and calculating an association rule index of medical data in the original ciphertext data set by using an association rule mining algorithm according to a mining keyword trapdoor;
in step 250, the association rule mining algorithm based on differential privacy protection is used to calculate the association rule index of the medical data in the original ciphertext data set in the embodiment of the present application, and the differential privacy protection is a new data privacy protection method. Differential privacy guarantees the following: the personal data that an attacker can obtain is almost comparable to what they can obtain from a data set without this personal record. Although less defined for privacy than Dalenius, the guarantee is strong enough because it is consistent with real world motivation-an individual has no motivation to participate in a dataset because the analyst of the dataset will draw the same conclusions about the individual no matter he is not in the dataset. Since their sensitive personal information is almost completely irrelevant to the output of the system, users can be sure that the organization handling their data does not violate their privacy. Analysts have little "no personal information" meaning that they are limited to minor variations on any individual's opinion. (here and below, "change" means the change between using a data set and using the same data set minus any one person's record.) the extent of this change is controlled by a parameter epsilon that sets the boundaries of the change for any possible outcome. A low value of e, for example 0.1, means that the opinion on any person changes very little. A high value of e, for example 50, means that the change in opinion about the person is larger. It is understood that the present application is equally applicable to other types of association rule mining algorithms, such as Apriori algorithm, FP-tree frequency set algorithm, and the like.
Specifically, the calculation method for calculating the association rule index of the medical data in the original ciphertext data set by using the association rule mining algorithm based on the differential privacy protection comprises the following steps:
step 251: filtering and sequencing the original ciphertext data set according to the mined keyword trapdoor to form a new ciphertext data set D*；
Step 252: based on the new ciphertext data set D*Constructing a frequent pattern tree, searching for frequent patterns meeting conditions and the support degree of each frequent pattern through the frequent pattern tree, and selecting a frequent pattern set Cset with the support degree counting not less than a preset threshold value min _ count;
step 253: selecting the highest k frequent patterns from the frequent pattern set Cset by adopting an exponential mechanism, and recording the combination formed by the real support degree counts of the corresponding frequent patterns as Sset so that each frequent pattern p meets Pr (p) < exp > (epsilon)1X Rank (D, p)/2k), where Rank (p) is the scoring value for frequent pattern p.
In step 253, the value k may be set according to the actual application, and the k frequent patterns screened by the exponential mechanism are the frequent pattern set most prone to privacy disclosure.
Step 254: adding Laplace noise to the support counts of the selected k frequent patterns
Forming Pset;
in step 254, the embodiment of the present application adds noise to the frequent pattern set that is most likely to have privacy disclosure in association rule mining by using a laplacian mechanism, so as to reduce the possibility of privacy disclosure, and ensure the security of mining result publication.
Step 255: carrying out consistency constraint on the support counts of the frequent patterns containing noise in the Pset;
in step 255, consistency constraints are utilized to increase the availability of frequent pattern sets after noise addition.
Step 256: the association rule index is calculated using the noise count set RC.
Step 260: and returning the association rule mining result to the sender according to the association rule index calculation result.
Please refer to fig. 4, which is a schematic structural diagram of an association rule mining system based on privacy protection according to an embodiment of the present application. The association rule mining system based on privacy protection comprises terminal equipment and a cloud server, wherein a sender performs data cleaning, data encryption, data uploading, keyword submitting and other operations through the terminal equipment, sends a registration request and an association rule mining request to the cloud server, and receives an association rule mining result returned by the cloud server.
Specifically, the terminal equipment comprises a data processing module, a registration request module, a data encryption module, a data mining request module and a keyword submitting module;
a data processing module: the system comprises a database, a database and a database server, wherein the database is used for cleaning medical data in the database and extracting user ID and user information keywords in the medical data to form a data set to be mined; wherein, data cleaning refers to the last procedure for finding and correcting recognizable errors in the data file, including checking data consistency, processing invalid values and missing values, etc., aiming at deleting duplicate information, correcting existing errors, and providing data consistency. Taking medical data as an example, the formed data set to be mined is shown in table 1 below:
TABLE 1 data set to be mined
In table 1, ID denotes an ID number of a user (patient), and Items denotes a disease of the patient, that is, a user information keyword. Specifically, the ID and Items may be determined according to the type of data to be mined.
A registration request module: the system comprises a cloud server and a server side, wherein the cloud server is used for submitting a registration request to the cloud server; wherein, the submitted registration request at least comprises the registration information such as sender ID.
A data encryption module: the cloud server side is used for receiving registration success information returned by the cloud server side, extracting a data set to be mined, carrying out searchable encryption on the data set to be mined to obtain an encrypted original ciphertext data set, and uploading the original ciphertext data set to the cloud server side; the method comprises the steps of encrypting a data set to be mined by utilizing a searchable encryption algorithm; searchable encryption is a cryptographic primitive which is developed in recent years and supports a user to search keywords in a ciphertext, a large amount of network and calculation expenses can be saved for the user, huge calculation resources of a cloud server are fully utilized to search the keywords in the ciphertext, and privacy of user data is guaranteed.
Specifically, the data encryption module includes:
a pre-encryption unit: used for pre-encrypting the data set to be mined to obtain pre-encrypted data Cpre；
A data division unit: for pre-encrypting data CpreDividing the pseudo random sequence into n-m bits and m bits, and generating a pseudo random sequence S with n-m bits by using a password;
a pseudo-random value generation unit: for generating a pseudo-random value k using a password;
a random salt value generation unit: for generating m-bit values by using pseudo-random functions F and F with pseudo-random sequence S as a parameter to form Salt value
An exclusive-or operation unit: for pre-encrypting data CpreAnd Salt value TiAnd performing exclusive OR to obtain an original ciphertext data set C, and uploading the original ciphertext data set C to the cloud server.
A data mining request module: the system comprises a server, a rule mining module and a rule mining module, wherein the rule mining module is used for sending an association rule mining request to the server after receiving uploading success information returned by the server;
a keyword submission module: the mining keyword trapdoor generation method is used for appointing a mining keyword, generating a mining keyword trapdoor after the mining keyword is encrypted, and submitting the generated mining keyword trapdoor to a cloud server. The selected mining key words are encrypted by using a symmetric searchable encryption algorithm in the embodiment of the application, the encryption mode is the same as that of the data set to be mined, and details are not repeated here. The mining keywords correspond to user information keywords in the data set to be mined, and the mining keyword trapdoor generation mode is as follows: a sender inputs a mining keyword W, encrypts the mining keyword W into a ciphertext X, and generates a trapdoor TD by using a random function f<X,fk(Xl)>。
The cloud server comprises a registration module, a data preprocessing module, a data storage module, a data reading module, a keyword matching module and an association rule mining module,
A registration module: the system comprises a server, a terminal device and a server, wherein the server is used for receiving a registration request sent by the terminal device, creating a corresponding storage area for the terminal device according to the registration request, starting the server providing data service for the terminal device, and returning registration success information to the terminal device after creation is completed;
a data preprocessing module: the system comprises a data processing module, a data processing module and a data processing module, wherein the data processing module is used for receiving an original ciphertext data set uploaded by a terminal device and preprocessing the original ciphertext data set to obtain a preprocessing result table; wherein, the preprocessing is to take the ID in the original ciphertext data set as a key value to gather the Items together. The pretreatment results are shown in table 2 below:
TABLE 2 results of pretreatment
| ID | TimeStamp | Items |
| ID1 | Timestamp | F1,F2,F4 |
| ID2 | Timestamp | F2,F4 |
A data storage module: and the data processing device is used for respectively randomly disordering the original ciphertext data set and the preprocessing result table, storing the original ciphertext data set and the preprocessing result table in a storage area corresponding to the terminal device, and returning uploading success information to the terminal device.
A data reading module: the system comprises a storage area, a preprocessing result table and a correlation rule mining module, wherein the storage area is used for storing correlation rule mining requests sent by terminal equipment;
a keyword matching module: the device comprises a preprocessing result table, a key word trap door and a key word trap door matching table, wherein the preprocessing result table is used for receiving a mining key word trap door submitted by a terminal device, matching the mining key word trap door with user information key words in the preprocessing result table, judging whether the matching of the mining key word trap door is successful or not, if the matching is failed, returning mining key word trap door matching failure information to the terminal device, and waiting for the terminal device to submit the mining key word trap door again; if the mining keywords are matched completely successfully, mining the association rules through an association rule mining module; according to the method and the device, before association rule mining is carried out, firstly, mining keyword trapdoors are minedMatching, if matching is successful, the original ciphertext data set contains the disease keyword, association rule mining can be carried out, decryption of the original ciphertext data set is not needed, and safety of data storage is guaranteed; if the matching fails, the original ciphertext data set does not contain the disease keyword, and subsequent association rule mining cannot be carried out. Specifically, the method for mining keyword matching is as follows: reading data C from Hbase (open source database), and performing XOR on X and C in TD to obtain<S,F>Using f in TDk(Xl) Judgment of
And if not, the matching of the mining keywords is successful, and if not, the matching of the mining keywords is failed.
An association rule mining module: and the correlation rule mining module is used for reading the original ciphertext data set from the storage area corresponding to the terminal equipment after the matching of the mining keyword is successful, calculating the correlation rule index of the medical data in the original ciphertext data set by using the correlation rule mining algorithm according to the mining keyword trapdoor, and returning the correlation rule mining result to the terminal equipment according to the correlation rule index calculation result. The association rule index of the medical data in the original ciphertext data set is calculated by using an association rule mining algorithm based on differential privacy protection, and the differential privacy protection is a new data privacy protection method. Differential privacy guarantees the following: the personal data that an attacker can obtain is almost comparable to what they can obtain from a data set without this personal record. Although less defined for privacy than Dalenius, the guarantee is strong enough because it is consistent with real world motivation-an individual has no motivation to participate in a dataset because the analyst of the dataset will draw the same conclusions about the individual no matter he is not in the dataset. Since their sensitive personal information is almost completely irrelevant to the output of the system, users can be sure that the organization handling their data does not violate their privacy. Analysts have little "no personal information" meaning that they are limited to minor variations on any individual's opinion. (here and below, "change" means the change between using a data set and using the same data set minus any one person's record.) the extent of this change is controlled by a parameter epsilon that sets the boundaries of the change for any possible outcome. A low value of e, for example 0.1, means that the opinion on any person changes very little. A high value of e, for example 50, means that the change in opinion about the person is larger. It is understood that the present application is equally applicable to other types of association rule mining algorithms, such as Apriori algorithm, FP-tree frequency set algorithm, and the like.
Specifically, the association rule mining module comprises:
a sorting unit: for filtering and sorting the original ciphertext data set according to the mining keyword trapdoor to form a new ciphertext data set D*；
A frequent mode selection unit: for generating a new ciphertext data set D*Constructing a frequent pattern tree, searching for frequent patterns meeting conditions and the support degree of each frequent pattern through the frequent pattern tree, and selecting a frequent pattern set Cset with the support degree counting not less than a preset threshold value min _ count;
a frequent pattern screening unit: the method is used for selecting the highest k frequent patterns from the frequent pattern set Cset by adopting an exponential mechanism, and the combination formed by the real support degree counts of the corresponding frequent patterns is recorded as Sset, so that each frequent pattern p meets Pr (p) < epsilon >) exp (epsilon)1X Rank (D, p)/2k), where Rank (p) is the scoring value for frequent pattern p. The k value can be set according to practical application, and the k frequent patterns screened by using an exponential mechanism are the frequent pattern set which is most prone to privacy disclosure.
A noise addition unit: adding Laplace noise to the support counts of the selected k frequent patterns
Forming Pset; according to the method and the device, the noise is added to the frequent pattern set which is most prone to privacy disclosure in association rule mining by using the Laplace mechanism, the probability of privacy disclosure is reduced, and the fact that privacy disclosure is guaranteed to be dugSafety of discovering result publication.
A consistency constraint unit: carrying out consistency constraint on the support counts of frequent patterns containing noise in the Pset; wherein the consistency constraint is utilized to increase the availability of the frequent pattern set after adding noise.
An association rule index calculation unit: and the system is used for calculating the association rule index by using the noise count set RC and returning the association rule mining result to the terminal equipment according to the association rule index calculation result.
Referring to fig. 5, a schematic structural diagram of a hardware device of an association rule mining method based on privacy protection according to an embodiment of the present application is shown in fig. 5, where the device includes one or more processors and a memory. Taking a processor as an example, the apparatus may further include: an input device and an output device.
The processor, memory, input devices, and output devices may be connected by a bus or other means, as exemplified by the bus connection in fig. 5.
The memory, which is a non-transitory computer readable storage medium, may be used to store non-transitory software programs, non-transitory computer executable programs, and modules. The processor executes various functional applications and data processing of the electronic device, i.e., implements the processing method of the above-described method embodiment, by executing the non-transitory software program, instructions and modules stored in the memory.
The memory may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data and the like. Further, the memory may include high speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid state storage device. In some embodiments, the memory optionally includes memory located remotely from the processor, and these remote memories may be connected to the processing device over a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The input device may receive input numeric or character information and generate a signal input. The output device may include a display device such as a display screen.
The one or more modules are stored in the memory and, when executed by the one or more processors, perform the following for any of the above method embodiments:
step a: the method comprises the steps that a sender conducts searchable encryption on a data set to be mined and uploads the encrypted data set to be mined to a cloud server;
step b: the sending party sends an association rule mining request to the cloud server and submits a mining keyword to the cloud server;
step c: and the cloud server side performs association rule mining on the encrypted data set to be mined by using an association rule mining algorithm based on differential privacy protection according to the mining key words.
The product can execute the method provided by the embodiment of the application, and has the corresponding functional modules and beneficial effects of the execution method. For technical details that are not described in detail in this embodiment, reference may be made to the methods provided in the embodiments of the present application.
Embodiments of the present application provide a non-transitory (non-volatile) computer storage medium having stored thereon computer-executable instructions that may perform the following operations:
step a: the method comprises the steps that a sender conducts searchable encryption on a data set to be mined and uploads the encrypted data set to be mined to a cloud server;
step b: the sending party sends an association rule mining request to the cloud server and submits a mining keyword to the cloud server;
step c: and the cloud server side performs association rule mining on the encrypted data set to be mined by using an association rule mining algorithm based on differential privacy protection according to the mining key words.
Embodiments of the present application provide a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions that, when executed by a computer, cause the computer to perform the following:
step a: the method comprises the steps that a sender conducts searchable encryption on a data set to be mined and uploads the encrypted data set to be mined to a cloud server;
step b: the sending party sends an association rule mining request to the cloud server and submits a mining keyword to the cloud server;
step c: and the cloud server side performs association rule mining on the encrypted data set to be mined by using an association rule mining algorithm based on differential privacy protection according to the mining key words.
According to the association rule mining method and system based on privacy protection and the electronic equipment, the data to be mined are encrypted by using the symmetric searchable encryption algorithm, the data storage safety is guaranteed, meanwhile, the association rule mining method based on the differential privacy protection is used for mining the association rule of the data to be mined, and the privacy of the whole data flow in the data mining process, the middle caching process and the later result publishing process is guaranteed.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (11)
1. An association rule mining method based on privacy protection is characterized by comprising the following steps:
step a: the method comprises the steps that a sender conducts searchable encryption on a data set to be mined and uploads the encrypted data set to be mined to a cloud server;
step b: the sending party sends an association rule mining request to the cloud server and submits a mining keyword to the cloud server;
step c: the cloud server side performs association rule mining on the encrypted data set to be mined by using an association rule mining algorithm based on differential privacy protection according to the mining key words;
in the step a, the searchable encryption of the data set to be mined by the sender specifically comprises: encrypting all user information keywords in the data set to be mined by utilizing a searchable encryption algorithm; the method for searching and encrypting the data set to be mined by the sender and uploading the encrypted data set to be mined to the cloud server specifically comprises the following steps:
step a 1: pre-encrypting the data set to be mined to obtain pre-encrypted data Cpre；
Step a 2: data C to be pre-encryptedpreDividing the pseudo random sequence into n-m bits and m bits, and generating a pseudo random sequence S with n-m bits by using a password;
step a 3: generating a pseudorandom value k using a password;
step a 4: using pseudo-random sequence S as parameter, using pseudo-random functions F and F to generate m bit value to form Salt value
Step a 5: data C to be pre-encryptedpreAnd Salt value TiPerforming XOR to obtain an original ciphertext data set;
step a 6: uploading the original ciphertext data set to a server;
in the step b, the submitting and mining keywords to the cloud server specifically comprises: a sender designates a mining keyword, encrypts the mining keyword by using a symmetrical searchable encryption algorithm to generate a mining keyword trapdoor, and submits the generated mining keyword trapdoor to a server; the mining keywords correspond to user information keywords;
in the step c, the performing association rule mining on the encrypted data set to be mined by using an association rule mining algorithm based on differential privacy protection specifically includes:
step c 1: filtering and sequencing the original ciphertext data set according to the mined keyword trapdoor to form a new ciphertext data set D*；
Step c 2: based on the new ciphertext data set D*Constructing a frequent pattern tree, searching for frequent patterns meeting conditions and the support degree of each frequent pattern through the frequent pattern tree, and selecting a frequent pattern set Cset with the support degree counting not less than a threshold value min _ count;
step c 3: selecting k frequent pattern sets which are most prone to privacy disclosure from the frequent pattern set Cset by adopting an exponential mechanism;
step c 4: adding noise to the support counts of the k frequent pattern sets;
step c 5: carrying out consistency constraint on the support counts of the k frequent pattern sets added with the noise;
step c 6: calculating an association rule index by using a noise count set;
step c 7: and returning the association rule mining result to the sender according to the association rule index calculation result.
2. The association rule mining method based on privacy protection as claimed in claim 1, further comprising before the step a: a sending party submits a registration request to a cloud server, the cloud server creates a corresponding storage area for the sending party according to the registration request, and a server for providing data service for the sending party is started; the storage area is used for storing the data set to be mined uploaded by the sender.
3. The association rule mining method based on privacy protection as claimed in claim 2, wherein in the step a, the sending party performs searchable encryption on the data set to be mined further comprises: and the sender cleans the data to be mined, extracts the user ID and the user information keywords in the data to be mined and forms a data set to be mined.
4. The association rule mining method based on privacy protection as claimed in claim 3, wherein in the step a6, after uploading the original ciphertext data set to the server, the method further comprises:
step a 7: the server preprocesses the original ciphertext data set to obtain a preprocessing result table; the preprocessing is to take the user ID in the original ciphertext data set as a key value and gather the user information keywords together;
step a 8: and after the original ciphertext data set and the preprocessing result table are respectively randomly disordered by the server, the original ciphertext data set and the preprocessing result table are stored in a storage area corresponding to the sender together.
5. The association rule mining method based on privacy protection as claimed in claim 4, wherein in the step c, the association rule mining of the encrypted data set to be mined by the cloud server using an association rule mining algorithm based on differential privacy protection according to the mining keyword further comprises: matching the mining keyword trapdoor with the user information keywords in the preprocessing result table, judging whether the matching is successful or not, and if the matching is successful, mining the association rule of the original ciphertext data set according to the mining keyword trapdoor; otherwise, returning the information of failed mining keyword trapdoor matching to the sender.
6. An association rule mining system based on privacy protection is characterized by comprising terminal equipment and a cloud server;
the terminal device includes:
a data encryption module: the cloud server side is used for conducting searchable encryption on the data set to be mined and uploading the encrypted data set to be mined to the cloud server side;
a data mining request module: the system comprises a cloud server and a server, wherein the cloud server is used for sending an association rule mining request to the cloud server;
a keyword submission module: the system is used for submitting mining keywords to a cloud server;
the cloud server comprises:
an association rule mining module: the association rule mining method is used for mining the association rule of the encrypted data set to be mined by utilizing an association rule mining algorithm based on differential privacy protection according to the mining keyword after receiving an association rule mining request;
the searchable encryption of the data set to be mined by the data encryption module is specifically as follows: encrypting all user information keywords in the data set to be mined by utilizing a searchable encryption algorithm; the data encryption module specifically comprises:
a pre-encryption unit: is used for pre-encrypting the data set to be mined to obtain pre-encrypted data Cpre；
A data division unit: for pre-encrypting data CpreDividing the pseudo random sequence into n-m bits and m bits, and generating a pseudo random sequence S with n-m bits by using a password;
a pseudo-random value generation unit: for generating a pseudo-random value k using a password;
a random salt value generation unit: for generating m-bit values by using pseudo-random functions F and F with pseudo-random sequence S as a parameter to form Salt value
An exclusive-or operation unit: for pre-encrypting data CpreAnd Salt value TiPerforming XOR to obtain an original ciphertext data set, and uploading the original ciphertext data set to a server;
the method for submitting the mining keywords to the cloud server by the keyword submitting module specifically comprises the following steps: the method comprises the steps of appointing a mining keyword, encrypting the mining keyword by using a symmetrical searchable encryption algorithm to generate a mining keyword trapdoor, and submitting the generated mining keyword trapdoor to a cloud server; the mining keywords correspond to user information keywords;
the association rule mining module specifically comprises:
a sorting unit: for filtering and sorting the original ciphertext data set according to the mining keyword trapdoor to form a new ciphertext data set D*；
A frequent mode selection unit: for root ofAccording to the new ciphertext data set D*Constructing a frequent pattern tree, searching for frequent patterns meeting conditions and the support degree of each frequent pattern through the frequent pattern tree, and selecting a frequent pattern set Cset with the support degree counting not less than a threshold value min _ count;
a frequent pattern screening unit: the method comprises the steps that k frequent pattern sets which are most prone to privacy disclosure are selected from the frequent pattern sets Cset by adopting an exponential mechanism;
a noise addition unit: adding noise to the support counts of the k frequent pattern sets;
a consistency constraint unit: carrying out consistency constraint on the support counts of the k frequent pattern sets for adding the noise;
an association rule index calculation unit: and the correlation rule index calculation module is used for calculating the correlation rule index by using the noise count set and returning a correlation rule mining result to the terminal equipment according to the correlation rule index calculation result.
7. The privacy protection based association rule mining system of claim 6, wherein the terminal device further comprises:
a registration request module: the system comprises a cloud server and a server side, wherein the cloud server is used for submitting a registration request to the cloud server;
the cloud server further comprises:
a registration module: the server is used for establishing a corresponding storage area for the terminal equipment according to the registration request and starting up a server for providing data service for the terminal equipment; the storage area is used for storing the data set to be mined.
8. The privacy protection based association rule mining system of claim 7, wherein the terminal device further comprises:
a data processing module: the method is used for cleaning the data to be mined, extracting the user ID and the user information keywords in the data to be mined and forming a data set to be mined.
9. The privacy protection based association rule mining system of claim 8, wherein the cloud server further comprises:
a data preprocessing module: the system is used for preprocessing an original ciphertext data set to obtain a preprocessing result table; the preprocessing is to take the user ID in the original ciphertext data set as a key value and gather the user information keywords together;
a data storage module: and the data processing device is used for respectively randomly disordering the original ciphertext data set and the preprocessing result table and storing the original ciphertext data set and the preprocessing result table in a storage area corresponding to the terminal device.
10. The privacy protection based association rule mining system of claim 9, wherein the cloud server further comprises:
a keyword matching module: the correlation rule mining module is used for matching the mining keyword trapdoor with the user information keywords in the preprocessing result table, judging whether the matching is successful or not, and mining the correlation rule of the original ciphertext data set through the correlation rule mining module if the matching is successful; and otherwise, returning the information of failure in digging keyword trapdoor matching to the terminal equipment.
11. An electronic device, comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the privacy preserving based association rule mining method of any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711391275.2A CN108022654B (en) | 2017-12-20 | 2017-12-20 | Association rule mining method and system based on privacy protection and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711391275.2A CN108022654B (en) | 2017-12-20 | 2017-12-20 | Association rule mining method and system based on privacy protection and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108022654A CN108022654A (en) | 2018-05-11 |
| CN108022654B true CN108022654B (en) | 2021-11-30 |
Family
ID=62074523
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711391275.2A Active CN108022654B (en) | 2017-12-20 | 2017-12-20 | Association rule mining method and system based on privacy protection and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108022654B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109614466A (en) * | 2018-11-15 | 2019-04-12 | 阿里巴巴集团控股有限公司 | The keyword rule generating method and its device excavated based on Frequent Set |
| CN109784721B (en) * | 2019-01-15 | 2021-01-26 | 广东度才子集团有限公司 | Employment data analysis and data mining analysis platform system |
| CN110221784B (en) * | 2019-06-12 | 2022-09-06 | 青岛大学 | Data deduplication method, device and equipment based on project reaction theory |
| CN113141329B (en) * | 2020-01-16 | 2023-04-18 | 中移(上海)信息通信科技有限公司 | Big data mining method, device, equipment and storage medium |
| CN111680329B (en) * | 2020-08-14 | 2020-11-10 | 成都中轨轨道设备有限公司 | Data processing method for improving data security |
| CN112464292B (en) * | 2021-01-27 | 2021-08-20 | 支付宝(杭州)信息技术有限公司 | Method and device for training neural network based on privacy protection |
| CN112597524B (en) * | 2021-03-03 | 2021-05-18 | 支付宝(杭州)信息技术有限公司 | Privacy intersection method and device |
| CN113779630A (en) * | 2021-09-09 | 2021-12-10 | 新疆大学 | DICOM-based CT medical image reversible desensitization method |
| CN114564749B (en) * | 2022-03-04 | 2022-12-23 | 厦门熙重电子科技有限公司 | User information protection method and server for smart cloud service |
| CN114896477A (en) * | 2022-06-08 | 2022-08-12 | 徐州医科大学 | Data mining safety visualization system and method supporting multiple language algorithms |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104050267B (en) * | 2014-06-23 | 2017-10-03 | 中国科学院软件研究所 | The personalized recommendation method and system of privacy of user protection are met based on correlation rule |
| CN104765848B (en) * | 2015-04-17 | 2017-12-01 | 中国人民解放军空军航空大学 | What support result efficiently sorted in mixing cloud storage symmetrically can search for encryption method |
| CN106503575B (en) * | 2016-09-22 | 2019-03-05 | 广东工业大学 | A kind of Mining Association Rules in Distributed Environments method for protecting privacy |
| CN106778314A (en) * | 2017-03-01 | 2017-05-31 | 全球能源互联网研究院 | A kind of distributed difference method for secret protection based on k means |
| CN107239693B (en) * | 2017-05-17 | 2020-12-04 | 杰创智能科技股份有限公司 | Analysis method and system based on password coding rule |
| CN107203725A (en) * | 2017-05-23 | 2017-09-26 | 广东工业大学 | A kind of vertical distribution formula association rule mining method for protecting privacy |
-
2017
- 2017-12-20 CN CN201711391275.2A patent/CN108022654B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN108022654A (en) | 2018-05-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108022654B (en) | Association rule mining method and system based on privacy protection and electronic equipment | |
| AU2018361246B2 (en) | Data protection via aggregation-based obfuscation | |
| US9977918B2 (en) | Method and system for verifiable searchable symmetric encryption | |
| US10235335B1 (en) | Systems and methods for cryptographically-secure queries using filters generated by multiple parties | |
| WO2020253467A1 (en) | Blockchain system-based data processing method, system and device | |
| JP2021527274A (en) | Distributed access, storage, and transfer of blockchain | |
| US9852306B2 (en) | Conjunctive search in encrypted data | |
| TW202009776A (en) | Secure multi-party computation method and apparatus, and electronic device | |
| JP2022549581A (en) | Computing system, method, non-transitory computer-readable medium and computer program product for determining the sequential order of blocks in a DAG-structured blockchain | |
| US20200159847A1 (en) | Contribution of multiparty data aggregation using distributed ledger technology | |
| CN106776904A (en) | The fuzzy query encryption method of dynamic authentication is supported in a kind of insincere cloud computing environment | |
| CN107113183A (en) | The controlled shared system and method for big data | |
| US11860856B2 (en) | Managing distributed ledger storage space | |
| WO2022068355A1 (en) | Encryption method and apparatus based on feature of information, device, and storage medium | |
| CN103607420A (en) | Safe electronic medical system for cloud storage | |
| CN104270448A (en) | Secret sharing cloud storage method for electronic medical records capable of being outsourced and reconstructed | |
| Layton et al. | Automating open source intelligence: algorithms for OSINT | |
| US20230315877A1 (en) | Managing machine-learning models via non-fungible tokens on a digital ledger | |
| CN110737912A (en) | thesis duplicate checking method based on homomorphic encryption | |
| Rajkumar et al. | Fuzzy-Dedup: A secure deduplication model using cosine based Fuzzy interference system in cloud application | |
| Amorim et al. | Homomorphic Encryption: An Analysis of its Applications in Searchable Encryption | |
| Layton et al. | Automating open source intelligence | |
| CN112487065A (en) | Data retrieval method and device | |
| Liao et al. | BCDP: a blockchain-based credible data publishing system | |
| Neelamegam et al. | Health Data Deduplication Using Window Chunking-Signature Encryption in Cloud. |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |