CN108022654A - A kind of association rule mining method based on secret protection, system and electronic equipment - Google Patents
A kind of association rule mining method based on secret protection, system and electronic equipment Download PDFInfo
- Publication number
- CN108022654A CN108022654A CN201711391275.2A CN201711391275A CN108022654A CN 108022654 A CN108022654 A CN 108022654A CN 201711391275 A CN201711391275 A CN 201711391275A CN 108022654 A CN108022654 A CN 108022654A
- Authority
- CN
- China
- Prior art keywords
- keyword
- mining
- data set
- data
- excavated
- 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.)
- Granted
Links
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
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Bioethics (AREA)
- General Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Databases & Information Systems (AREA)
- Computer Security & Cryptography (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Medical Informatics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
This application involves data mining technology field, the more particularly to a kind of association rule mining method based on secret protection, system and electronic equipment.The described method includes：Sender treats mining data collection and carries out can search for encrypting, and encrypted data set to be excavated is uploaded to cloud server terminal；Sending direction cloud server terminal sends association rule mining request, and is submitted to cloud server terminal and excavate keyword；Cloud server terminal is associated rule digging using the association rules mining algorithm based on difference secret protection according to the excavation keyword to the encrypted data set to be excavated.The application can search for Encryption Algorithm and treat mining data being encrypted using symmetrical; the security of data storage is ensured; treating mining data using the association rule mining method of difference secret protection is associated rule digging at the same time; it ensure that the process in data mining, the full-range privacy of data in intermediate buffer process and anaphase issuing process.
Description
Technical field
This application involves data mining technology field, more particularly to a kind of association rule mining side based on secret protection
Method, system and electronic equipment.
Background technology
With the development of society, people also focus more on certainly when enjoying the facility that development in science and technology and social progress are brought
Oneself health.With the development of big data technology, the medical big data being covered with dust in hospital is paid attention to again, due to doctor
The particularity of big data is treated, people can excavate many by means such as data minings in the medical big data that these are covered with dust
New, useful information, for carrying out accurate assisting in diagnosis and treatment.
But since the sensitiveness of medical big data is high, the privacy of medical big data how is ensured in data mining is
One significant challenge.Bear richness stamen《The research of the association rules mining algorithm of secret protection based on MapReduce》Chapter 3
It refer to carry out secret protection to data to be excavated.Kang Haiyan exists《Difference secret protection is in data mining using summary》's
2.1 sections refer to a variety of mode excavations based on difference privacy, and they are compared.Party's dawn《Based on MapReduce's
Symmetrically it can search for encipherment scheme》Chapter 4, refer to symmetrically can search for encipherment scheme based on MapReduce.
Association rule mining (Association rule mining) be in data mining most active research method it
One, it can be used for finding the contact between thing.For example, the rule { hypertension, heart disease } found from medical data → in
Wind } it can show that the patient also suffers from apoplexy if patient while suffering from hypertension and heart disease.This rule-like can be used as and make
Diagnosis and prediction patient disease are so as to the basis of prevention ahead of time.But existing Association Rule Mining is mainly focused on curing
The problem for the treatment of the performance issue in data mining, not addressing how but to ensure medical data privacy.
The content of the invention
This application provides a kind of association rule mining method based on secret protection, system and electronic equipment, it is intended to extremely
It is few to solve one of above-mentioned technical problem of the prior art to a certain extent.
To solve the above-mentioned problems, this application provides following technical solution：
A kind of association rule mining method based on secret protection, including：
Step a：Sender treats mining data collection and carries out can search for encrypting, and encrypted data set to be excavated is uploaded
To cloud server terminal；
Step b：Sending direction cloud server terminal sends association rule mining request, and is submitted to cloud server terminal and excavate key
Word；
Step c：Cloud server terminal is calculated according to the excavation keyword using the association rule mining based on difference secret protection
Method is associated rule digging to the encrypted data set to be excavated.
The technical solution that the embodiment of the present application is taken further includes：Further included before the step a：Sending direction cloud server terminal
Submission registration request, cloud server terminal creates corresponding storage region according to the registration request for the sender, and starts to be somebody's turn to do
Sender provides the server of data service；The data set to be excavated that the storage region is used to store described sender upload is used
Family information.
The technical solution that the embodiment of the present application is taken further includes：In the step a, described sender treats mining data
Collection carries out can search for encryption further including：Sender treats mining data and is cleaned, extract User ID in data to be excavated and
User information keyword, forms data set to be excavated.
The technical solution that the embodiment of the present application is taken further includes：In the step a, described sender treats mining data
Collection carries out can search for encryption：All user informations in the data set to be excavated are closed using can search for Encryption Algorithm
Keyword is encrypted；Described sender is treated mining data collection and is encrypted, and encrypted data set to be excavated is uploaded to cloud
Server-side specifically includes：
Step a1：The data set to be excavated is pre-encrypted, obtains pre-add ciphertext data Cpre；
Step a2：By pre-add ciphertext data CpreN-m bits and m bits are divided into, and utilizes the puppet of password generation n-m bits
Random sequence S；
Step a3：Use password generation pseudorandom values k；
Step a4：Using pseudo-random sequence S as parameter, using the value of pseudo-random function F and f generation m bits, form
Salt values
Step a5：By pre-add ciphertext data CpreWith Salt values TiExclusive or, obtains original cipher text data set；
Step a6：Original cipher text data set is uploaded onto the server.
The technical solution that the embodiment of the present application is taken further includes：It is described by original cipher text data set in the step a6
Further included after uploading onto the server：
Step a7：Server pre-processes original cipher text data set, obtains pre-processed results table；The pretreatment is
Using the User ID in original cipher text data set as key assignments, user information keyword is flocked together；
Step a8：After original cipher text data set and pre-processed results table are upset order by server at random respectively, deposit together
Storage is in the corresponding storage region of the sender.
The technical solution that the embodiment of the present application is taken further includes：In the step b, described submitted to cloud server terminal is excavated
Keyword is specially：Sender specifies excavation keyword, using the symmetrical Encryption Algorithm that can search for by the excavation keyword encryption
Keyword trapdoor is excavated in generation afterwards, and the excavation keyword trapdoor of generation is submitted to server；Wherein, the excavation keyword
It is corresponding with user information keyword.
The technical solution that the embodiment of the present application is taken further includes：In the step c, the cloud server terminal is according to the digging
Keyword is dug, the encrypted data set to be excavated is closed using the association rules mining algorithm based on difference secret protection
Connection rule digging further includes：By the excavation keyword trapdoor and the user information keyword progress in pre-processed results table
Match somebody with somebody, and judge matching whether all success, if matching all successes, according to the excavation keyword trapdoor to described original close
Literary data set is associated rule digging；Otherwise, returned to sender and excavate keyword trapdoor it fails to match information.
The technical solution that the embodiment of the present application is taken further includes：It is described to be protected using based on difference privacy in the step c
The association rules mining algorithm of shield is associated rule digging to the encrypted data set to be excavated and specifically includes：
Step c1：According to keyword trapdoor is excavated, original cipher text data set is filtered and sorted, forms new ciphertext
Data set D*；
Step c2：According to new ciphertext data set D*Frequent pattern tree (fp tree) is constructed, is searched for by frequent pattern tree (fp tree) eligible
Frequent mode and each frequent mode support, and choose support counting be not less than threshold value min_count frequent mould
Formula set Cset；
Step c3：K are picked out from frequent mode set Cset be easiest to privacy leakage occur using index mechanism
Fuzzy frequent itemsets；
Step c4：Noise is added to the support counting of k fuzzy frequent itemsets；
Step c5：The support counting of the k fuzzy frequent itemsets to adding noise carries out consistency constraint；
Step c6：Added up to using noise counter set and calculate correlation rule index；
Step c7：Association rule mining result is returned to sender according to correlation rule index result of calculation.
Another technical solution that the embodiment of the present application is taken is：A kind of association rule mining system based on secret protection,
Including terminal device and cloud server terminal；
The terminal device includes：
Data encryption module：It is encrypted for treating mining data collection, and encrypted data set to be excavated is uploaded to
Cloud server terminal；
Data mining request module：For sending association rule mining request to cloud server terminal；
Keyword submits module：Keyword is excavated for being submitted to cloud server terminal；
The cloud server terminal includes：
Association rule mining module：For after association rule mining request is received, according to the excavation keyword, profit
Rule digging is associated to the encrypted data set to be excavated with the association rules mining algorithm based on difference secret protection.
The technical solution that the embodiment of the present application is taken further includes：The terminal device further includes：
User information registration request module：For submitting registration request to cloud server terminal；
The cloud server terminal further includes：
Registration module：For creating corresponding storage region according to the registration request for the terminal device, and start and be
The terminal device provides the server of data service；The storage region is used to store the data set to be excavated.
The technical solution that the embodiment of the present application is taken further includes：The terminal device further includes：
Data processing module：Cleaned for treating mining data, extract the User ID in data to be excavated and user
Information key, forms data set to be excavated.
The technical solution that the embodiment of the present application is taken further includes：The data encryption module treats the progress of mining data collection can
Searching for encryption is specially：All user information keywords in the data set to be excavated are carried out using can search for Encryption Algorithm
Encryption；The data encryption module specifically includes：
Pre-encrypt unit：For the data set to be excavated to be pre-encrypted, pre-add ciphertext data C is obtainedpre；
Data partitioning unit：For by pre-add ciphertext data CpreN-m bits and m bits are divided into, and utilizes password generation n-
The pseudo-random sequence S of m bits；
Pseudorandom values generation unit：For using password generation pseudorandom values k；
Random salt figure generation unit：For using pseudo-random sequence S as parameter, m to be generated using pseudo-random function F and f
The value of bit, forms Salt values
Xor operation unit：For by pre-add ciphertext data CpreWith Salt values TiExclusive or, obtains original cipher text data set, and
Original cipher text data set is uploaded onto the server.
The technical solution that the embodiment of the present application is taken further includes：The cloud server terminal further includes：
Data preprocessing module：For being pre-processed to original cipher text data set, pre-processed results table is obtained；It is described pre-
Processing flocks together user information keyword i.e. using the ID in original cipher text data set as key assignments；
Data memory module：After original cipher text data set and pre-processed results table are upset order at random respectively, one
Rise and be stored in the corresponding storage region of the terminal device.
The technical solution that the embodiment of the present application is taken further includes：The keyword is submitted module to be submitted to cloud server terminal and is excavated
Keyword is specially：Specify and excavate keyword, will be generated using the symmetrical Encryption Algorithm that can search for after the excavation keyword encryption
Keyword trapdoor is excavated, and the excavation keyword trapdoor of generation is submitted into cloud server terminal；Wherein, the excavation keyword is with using
Family information key is corresponding.
The technical solution that the embodiment of the present application is taken further includes：The cloud server terminal further includes：
Keywords matching module：For the excavation keyword trapdoor and the user information in pre-processed results table is crucial
Whether all word is matched, and judge matching success, if matching all successes, pass through the association rule mining module pair
The original cipher text data set is associated rule digging；Otherwise returned to terminal device and excavate keyword trapdoor it fails to match letter
Breath.
The technical solution that the embodiment of the present application is taken further includes：The association rule mining module specifically includes：
Sequencing unit：For according to keyword trapdoor is excavated, original cipher text data set being filtered and being sorted, formed new
Ciphertext data set D*；
Frequent mode chooses unit：For according to new ciphertext data set D*Frequent pattern tree (fp tree) is constructed, passes through frequent mode
Tree searches for the support of qualified frequent mode and each frequent mode, and chooses support counting and be not less than threshold value min_
The frequent mode set Cset of count；
Frequent mode screening unit：It is easiest to for picking out k from frequent mode set Cset using index mechanism
There are the fuzzy frequent itemsets of privacy leakage；
Noise adding device：For adding noise to the support counting of k fuzzy frequent itemsets；
Consistency constraint unit：Support counting for the k fuzzy frequent itemsets to adding noise carries out uniformity about
Beam；
Correlation rule exponent calculation unit：Correlation rule index is calculated for adding up to using noise counter set, and according to association
Regularity index result of calculation returns to association rule mining result to terminal device.
The another technical solution that the embodiment of the present application is taken is：A kind of electronic equipment, including：
At least one processor；And
The memory being connected with least one processor communication；Wherein,
The memory storage has the instruction that can be performed by one processor, and described instruction is by least one place
Manage device perform so that at least one processor be able to carry out the above-mentioned association rule mining method based on secret protection with
Lower operation：
Step a：Sender treats mining data collection and carries out can search for encrypting, and encrypted data set to be excavated is uploaded
To cloud server terminal；
Step b：Sending direction cloud server terminal sends association rule mining request, and is submitted to cloud server terminal and excavate key
Word；
Step c：Cloud server terminal is calculated according to the excavation keyword using the association rule mining based on difference secret protection
Method is associated rule digging to the encrypted data set to be excavated.
Relative to the prior art, the beneficial effect that the embodiment of the present application produces is：The embodiment of the present application based on privacy
Association rule mining method, system and the electronic equipment of protection can search for Encryption Algorithm and treat mining data being added using symmetrical
Close processing, has ensured the security of data storage, while utilizes the association rule mining method of difference secret protection to be excavated
Data are associated rule digging, it is ensured that in the process of data mining, intermediate buffer process and anaphase issuing process
The middle full-range privacy of data.
Brief description of the drawings
Fig. 1 is the overall flow figure of the association rule mining method based on secret protection of the embodiment of the present application；
Fig. 2 is the process chart of the data encryption of the embodiment of the present application；
Fig. 3 is the process chart of the association rule mining of the embodiment of the present application；
Fig. 4 is the structure diagram of the association rule mining system based on secret protection of the embodiment of the present application；
Fig. 5 is the hardware device structural representation of the association rule mining method based on secret protection of the embodiment of the present application
Figure.
Embodiment
It is with reference to the accompanying drawings and embodiments, right in order to which the object, technical solution and advantage of the application are more clearly understood
The application is further elaborated.It should be appreciated that specific embodiment described herein is only to explain the application, not
For limiting the application.
Referring to Fig. 1, it is the overall flow figure of the association rule mining method based on secret protection of the embodiment of the present application.
The association rule mining method based on secret protection of the embodiment of the present application comprises the following steps：
Step a：Sender treats mining data collection and carries out can search for encrypting, and encrypted data set to be excavated is uploaded
To cloud server terminal；
Step b：Sending direction cloud server terminal sends association rule mining request, and is submitted to cloud server terminal and excavate key
Word；
Step c：Cloud server terminal is calculated according to the excavation keyword using the association rule mining based on difference secret protection
Method is associated rule digging to the encrypted data set to be excavated.
Based on above-mentioned, the association rule mining method of the embodiment of the present application includes data encryption and association rule mining two
Stage.In data encryption stage, sender is treated after mining data is encrypted using the symmetrical Encryption Algorithm that can search for, and will be encrypted
Data are uploaded to cloud server terminal；In the association rule mining stage, cloud server terminal utilizes the correlation rule based on difference secret protection
Mining algorithm is associated rule digging to encryption data, without encryption data is decrypted, has ensured Result public affairs
The security of cloth.The application is suitable for the association rule mining of polytype big data, such as business data, machine and sensing
Device data, social data etc., for the ease of fairly setting out the technical solution of the application, only with medical big data in following embodiments
Association rule mining exemplified by be specifically described.
Specifically, referring to Fig. 2, being the process chart of the data encryption of the embodiment of the present application.The embodiment of the present application
Data ciphering method comprises the following steps：
Step 100：Sender cleans the medical data in database, and extract the User ID in medical data and
User information keyword, forms data set to be excavated；
In step 100, data cleansing refers to find and corrects identifiable wrong last one of journey in data file
Sequence, including check data consistency, processing invalid value and missing values etc., it is therefore intended that it is wrong existing for deletion duplicate message, correction
By mistake, and data consistency is provided.By taking medical data as an example, the data set to be excavated of formation is as shown in table 1 below：
1 data set to be excavated of table
| ID | Items |
| ID1 | F1 |
| ID1 | F2 |
| ID1 | F4 |
| ID2 | F2 |
| ID2 | F4 |
| ID3 | F3 |
In table 1, ID represents the ID number of user (patient), and Items represents the disease of user (patient), is user information
Keyword.Specifically, ID and Items can be determined according to the type of data to be excavated.
Step 110：Sending direction cloud server terminal submits registration request；
In step 110, the log-on messages such as sender ID are included at least in the registration request that sender submits.
Step 120：After cloud server terminal receives registration request, created for the sender be used to store depositing for data beyond the clouds
Storage area domain, starts the server that data service is provided for the sender, and returns and succeed in registration to sender after completion is created
Information；
Step 130：After sender receives the information that succeeds in registration, data set to be excavated is extracted, and treats mining data collection
Carry out can search for encrypting, obtain encrypted original cipher text data set；
In step 130, the embodiment of the present application, which utilizes, can search for Encryption Algorithm ((Searchable Encryption, letter
Claim SE)) treat mining data collection and be encrypted；Can search for encryption is that one kind developed in recent years supports user in ciphertext
The cryptographic primitive of keyword lookup is carried out, it can be that user saves substantial amounts of network and computing cost, and make full use of cloud
Hold the huge computing resource of server to carry out the keyword in ciphertext to search, ensure that the privacy of user data.
Specifically, following step is included using can search for Encryption Algorithm and treat the cipher mode that mining data collection is encrypted
Suddenly：
Step 131：Data set to be excavated is pre-encrypted, obtains pre-add ciphertext data Cpre；
Step 132：By pre-add ciphertext data CpreTwo parts of n-m bits and m bits are divided into, and utilize password generation n-m
The pseudo-random sequence S of bit；
Step 133：Use password generation pseudorandom values k；
Step 134：Using pseudo-random sequence S as parameter, using the value of pseudo-random function F and f generation m bits, form
Salt values (random salt figure)
Step 135：By pre-add ciphertext data CpreWith Salt values TiExclusive or, obtains original cipher text data set C.
Step 140：Sender uploads onto the server original cipher text data set；
Step 150：Server receives the original cipher text data set that sender uploads, and original cipher text data set is carried out pre-
Processing, obtains pre-processed results table；
In step 150, pretreatment flocks together Items i.e. using the ID in original cipher text data set as key assignments.In advance
Handling result table is as shown in table 2 below：
2 pre-processed results of table
| ID | TimeStamp | Items |
| ID1 | Timestamp | F1,F2,F4 |
| ID2 | Timestamp | F2,F4 |
Step 160：After original cipher text data set and pre-processed results table are upset order by server at random respectively, deposit together
Storage returns to sender in the corresponding storage region of the sender and uploads successful information.
Referring to Fig. 3, it is the process chart of the association rule mining of the embodiment of the present application.The association of the embodiment of the present application
Rule digging method comprises the following steps：
Step 200：Sender sends association rule mining request；
Step 210：Server receives association rule mining request, and is read in advance out of the sender corresponding storage region
Handling result table；
Step 220：Sender specifies excavation keyword, and generation after selected excavation keyword encryption is excavated keyword and is fallen into
Door (if one logs in processing system and allows a specific CUSTOMER ID, common mouth can be bypassed by the identification code
Order checks intuitively understand to be exactly that the operation such as modify can be logged in by a special username and password.This peace
Complete dangerous referred to as trapdoor, also known as unauthorized access, hide key word information using trapdoor here and ensure key word information at the same time
Can be identified), and the excavation keyword trapdoor of generation is submitted into server.
In a step 220, the embodiment of the present application is encrypted selected excavation keyword using symmetrically can search for Encryption Algorithm,
Cipher mode is identical with the cipher mode of data set to be excavated, will not be described in great detail herein.Excavate keyword and data set to be excavated
In user information keyword it is corresponding, excavate keyword trapdoor generating mode be：Sender, which inputs, excavates keyword W, will dig
Pick keyword W is encrypted to ciphertext X, utilizes random function f generation trapdoors TD=<X,fk(Xl)>。
Step 230：Server receives the excavation keyword trapdoor that sender submits, and will excavate keyword trapdoor and pretreatment
As a result whether all the user information keyword in table is matched, and judge to excavate keyword trapdoor matching success, if
With failure, step 240 is performed；If matching all successes, perform step 250；
In step 230, matched first to excavating keyword trapdoor before rule digging is associated, if
With success, then illustrate to include the excavation keyword trapdoor in original cipher text data set, rule digging can be associated, without
Original cipher text data set is decrypted, has ensured the security of data storage；If it fails to match, illustrate original cipher text number
Do not include the excavation keyword trapdoor according to concentrating, follow-up association rule mining can not be carried out.Specifically, Keywords matching is excavated
Mode is：Data C is read from Hbase (PostgreSQL database), the X in TD and C are done into exclusive or, obtained<S,F>, using in TD
fk(Xl), judgeWhether set up, represent to excavate Keywords matching success if setting up, if invalid
Then represent to excavate Keywords matching failure.
Step 240：Returned to sender and excavate keyword trapdoor it fails to match information, and re-execute step 220；
Step 250：Original cipher text data set is read from the corresponding storage region of the sender, and according to excavation keyword
Trapdoor, the correlation rule index of medical data in original cipher text data set is calculated using association rules mining algorithm；
In step 250, the embodiment of the present application is calculated former using the association rules mining algorithm based on difference secret protection
The correlation rule index of medical data in beginning ciphertext data set, difference secret protection are a kind of data-privacy protection sides of new proposition
Method.Difference privacy reassurances herein below：The personal data that attacker can the obtain almost number with their never this personal record
It can be obtained according to concentration very nearly the same.Although weaker than definition of the Dalenius to privacy, be ensured of it is powerful enough, because
Meeting the motivation of real world for it --- personal no motivation is not involved in data set, no matter because whether oneself or not data set
In, the analyst of the data set will draw identical conclusion on the individual.Due to its sensitive personal information and system
Output it is almost uncorrelated, therefore user be believed that handle its data tissue will not invade their privacy.Analysis
Person, which almost " can not obtain personal information ", means that they are limited in the minor variations on any personal view.(
Here and below, " change " refers to subtract the change between anyone record using data set and using identical data set
Change.) scope of this change controls by parameter ε, to any possible as a result, the parameter setting border of change.ε's is low
Value, such as 0.1, it is meant that the change on any personal view is considerably less.The high level of ε, such as 50, it is meant that on individual
View change bigger.It is appreciated that the application is equally applicable to other kinds of association rules mining algorithm, such as
Apriori algorithm, FP- tree frequency set algorithms etc..
Specifically, medical treatment in original cipher text data set is calculated using the association rules mining algorithm based on difference secret protection
The calculation of the correlation rule index of data comprises the following steps：
Step 251：According to keyword trapdoor is excavated, original cipher text data set is filtered and sorted, formed new close
Literary data set D*；
Step 252：According to new ciphertext data set D*Frequent pattern tree (fp tree) is constructed, is searched for by frequent pattern tree (fp tree) eligible
Frequent mode and each frequent mode support, and choose support counting be not less than predetermined threshold value min_count frequency
Numerous set of modes Cset；
Step 253：Highest k frequent mode is picked out from frequent mode set Cset using index mechanism, accordingly
The combination that the true support counting of frequent mode is formed is denoted as Sset so that each frequent mode p meets Pr (p) ∝ exp (ε1
× Rank (D, p)/2k), wherein Rank (p) is the marking value of frequent mode p.
In step 253, k values can be set according to practical application, the k frequent mode that utilization index mechanism filters out
As it is easiest to the fuzzy frequent itemsets for privacy leakage occur.
Step 254：The support counting of the k frequent mode to being selected adds Laplce's noiseFormed
Pset；
In step 254, the embodiment of the present application is hidden to being easiest to occur in association rule mining using Laplce's mechanism
The fuzzy frequent itemsets addition noise of private leakage, reduces it and reveals the possibility of privacy, ensured the security that Result is announced.
Step 255：Consistency constraint is carried out to the support counting of the frequent mode containing noise in Pset；
In step 255, the availability of the fuzzy frequent itemsets after addition noise is improved using consistency constraint.
Step 256：Correlation rule index is calculated using noise count set RC.
Step 260：Association rule mining result is returned to sender according to correlation rule index result of calculation.
Referring to Fig. 4, it is the structure diagram of the association rule mining system based on secret protection of the embodiment of the present application.
The association rule mining system based on secret protection of the embodiment of the present application includes terminal device and cloud server terminal, and sender passes through
The operations such as terminal device carries out data cleansing, data encryption, data upload, keyword is submitted, and send and register to cloud server terminal
Request, correlation rule excavate request, and the correlation rule for receiving cloud server terminal return excavates result.
Specifically, terminal device is asked including data processing module, registration request module, data encryption module, data mining
Modulus block and keyword submit module；
Data processing module：For being cleaned to the medical data in database, and extract the user in medical data
ID and user information keyword, form data set to be excavated；Wherein, data cleansing refers to find and corrects in data file to know
Other wrong last one of program, including check data consistency, processing invalid value and missing values etc., it is therefore intended that delete weight
Mistake existing for complex information, correction, and data consistency is provided.By taking medical data as an example, the data set to be excavated of formation is as follows
Shown in table 1：
1 data set to be excavated of table
In table 1, ID represents the ID number of user (patient), and Items represents the disease of patient, is that user information is crucial
Word.Specifically, ID and Items can be determined according to the type of data to be excavated.
Registration request module：For submitting registration request to cloud server terminal；Wherein, included at least in the registration request of submission
The log-on messages such as sender ID.
Data encryption module：For after the information that succeeds in registration of cloud server terminal return is received, extracting data to be excavated
Collection, treats mining data collection and carries out can search for encrypting, and obtains encrypted original cipher text data set, and by original cipher text data set
Upload to cloud server terminal；Wherein, the embodiment of the present application is encrypted using can search for Encryption Algorithm and treat mining data collection；It can search
Suo Jiami is a kind of cryptographic primitive for supporting user to carry out keyword lookup in ciphertext developed in recent years, it can
Substantial amounts of network and computing cost are saved for user, and makes full use of the huge computing resource of cloud server to carry out in ciphertext
Keyword is searched, and ensure that the privacy of user data.
Specifically, data encryption module includes：
Pre-encrypt unit：For data set to be excavated to be pre-encrypted, pre-add ciphertext data C is obtainedpre；
Data partitioning unit：For by pre-add ciphertext data CpreTwo parts of n-m bits and m bits are divided into, and are utilized close
The pseudo-random sequence S of code generation n-m bits；
Pseudorandom values generation unit：For using password generation pseudorandom values k；
Random salt figure generation unit：For using pseudo-random sequence S as parameter, m to be generated using pseudo-random function F and f
The value of bit, forms Salt values
Xor operation unit：For by pre-add ciphertext data CpreWith Salt values TiExclusive or, obtains original cipher text data set C, and
Original cipher text data set C is uploaded into cloud server terminal.
Data mining request module：For after the upload successful information of server return is received, being sent to server
Association rule mining is asked；
Keyword submits module：Keyword is excavated for specified, generation excavation keyword is fallen into after excavating keyword encryption
Door, and the excavation keyword trapdoor of generation is submitted into cloud server terminal.Wherein, the embodiment of the present application utilizes and symmetrically can search for encrypting
Algorithm encrypts selected excavation keyword, and cipher mode is identical with the cipher mode of data set to be excavated, will no longer go to live in the household of one's in-laws on getting married herein
State.It is corresponding with the user information keyword in data set to be excavated to excavate keyword, excavating keyword trapdoor generating mode is：
Sender, which inputs, excavates keyword W, will excavate keyword W and is encrypted to ciphertext X, utilizes random function f generation trapdoors TD=<X,fk
(Xl)>。
Cloud server terminal includes registration module, data preprocessing module, data memory module, data read module, keyword
Matching module and association rule mining module,
Registration module：The registration request sent for receiving terminal apparatus, creates according to registration request for the terminal device
Corresponding storage region, and start the server that data service is provided for the terminal device, to terminal device after completion is created
Return is succeeded in registration information；
Data preprocessing module：The original cipher text data set uploaded for receiving terminal apparatus, and to original cipher text data
Collection is pre-processed, and obtains pre-processed results table；Wherein, pretreatment, will i.e. using the ID in original cipher text data set as key assignments
Items flocks together.Pre-processed results table is as shown in table 2 below：
2 pre-processed results of table
| ID | TimeStamp | Items |
| ID1 | Timestamp | F1,F2,F4 |
| ID2 | Timestamp | F2,F4 |
Data memory module：After original cipher text data set and pre-processed results table are upset order at random respectively, one
Rise and be stored in the corresponding storage region of the terminal device, and returned to terminal device and upload successful information.
Data read module：The association rule mining request sent for receiving terminal apparatus, and from the terminal device pair
Pre-processed results table is read in the storage region answered；
Keywords matching module：The excavation keyword trapdoor submitted for receiving terminal apparatus, will excavate keyword trapdoor
Matched with the user information keyword in pre-processed results table, and judge to excavate the matching of keyword trapdoor whether all into
Work(, if it fails to match, returns to terminal device and excavates keyword trapdoor it fails to match information, and wait terminal device to carry again
Hand over and excavate keyword trapdoor；If excavating Keywords matching all successes, rule is associated by association rule mining module
Excavate；Wherein, the embodiment of the present application is matched to excavating keyword trapdoor first before rule digging is associated, if
Successful match, then illustrate to include the disease keyword in original cipher text data set, rule digging can be associated, without right
Original cipher text data set is decrypted, and has ensured the security of data storage；If it fails to match, illustrate original cipher text data
Concentration does not include the disease keyword, can not carry out follow-up association rule mining.Specifically, Keywords matching mode is excavated
For：Data C is read from Hbase (PostgreSQL database), the X in TD and C are done into exclusive or, obtained<S,F>, utilize the f in TDk
(Xl), judgeWhether set up, represent to excavate Keywords matching success if setting up, if invalid
Represent to excavate Keywords matching failure.
Association rule mining module：For excavate Keywords matching success after, from the corresponding memory block of the terminal device
Original cipher text data set is read in domain, according to keyword trapdoor is excavated, original cipher text number is calculated using association rules mining algorithm
According to the correlation rule index for concentrating medical data, and according to correlation rule index result of calculation correlation rule is returned to terminal device
Result.Wherein, the embodiment of the present application calculates original cipher text using the association rules mining algorithm based on difference secret protection
The correlation rule index of medical data in data set, difference secret protection are a kind of data-privacy guard methods of new proposition.Difference
Divide privacy reassurances herein below：The personal data that attacker can obtain almost with they the never data set of this personal record
It can obtain very nearly the same.Although weaker than definition of the Dalenius to privacy, be ensured of it is powerful enough because it accord with
Close the motivation of real world --- personal no motivation is not involved in data set, no matter because oneself or not in data set, the number
Identical conclusion on the individual will be all drawn according to the analyst of collection.Since the output of its sensitive personal information and system is several
It is completely uncorrelated, therefore user is believed that the tissue for handling its data will not invade their privacy.Analyst's almost " nothing
Method, which obtains personal information, " means that they are limited in the minor variations on any personal view.(herein with
Face, " change " refer to subtract the change between anyone record using data set and using identical data set.) this
The scope of change is controlled by parameter ε, to any possible as a result, the border of parameter setting change.The low value of ε, such as
0.1, it is meant that the change on any personal view is considerably less.The high level of ε, such as 50, it is meant that on personal view
Change bigger.It is appreciated that the application is equally applicable to other kinds of association rules mining algorithm, such as Apriori is calculated
Method, FP- tree frequency set algorithms etc..
Specifically, association rule mining module includes：
Sequencing unit：For according to keyword trapdoor is excavated, original cipher text data set being filtered and being sorted, formed new
Ciphertext data set D*；
Frequent mode chooses unit：For according to new ciphertext data set D*Frequent pattern tree (fp tree) is constructed, passes through frequent mode
Tree searches for the support of qualified frequent mode and each frequent mode, and chooses support counting and be not less than predetermined threshold value
The frequent mode set Cset of min_count；
Frequent mode screening unit：For picking out highest k from frequent mode set Cset using index mechanism
Frequent mode, the combination that the true support counting of corresponding frequent mode is formed are denoted as Sset so that each frequent mode p meets
Pr(p)∝exp(ε1× Rank (D, p)/2k), wherein Rank (p) is the marking value of frequent mode p.Wherein, k values can be according to reality
Border application is set, and the k frequent mode that utilization index mechanism filters out is the frequent mould for being easiest to privacy leakage occur
Formula collection.
Noise adding device：Support counting for the k frequent mode to being selected adds Laplce's noiseForm Pset；Wherein, the embodiment of the present application using Laplce's mechanism to being easiest to occur in association rule mining
The fuzzy frequent itemsets addition noise of privacy leakage, reduces it and reveals the possibility of privacy, ensured the safety that Result is announced
Property.
Consistency constraint unit：For carrying out uniformity about to the support counting of the frequent mode containing noise in Pset
Beam；Wherein, the availability of the fuzzy frequent itemsets after addition noise is improved using consistency constraint.
Correlation rule exponent calculation unit：For calculating correlation rule index using noise count set RC, and according to pass
Join regularity index result of calculation and return to association rule mining result to terminal device.
Referring to Fig. 5, it is the hardware device knot of the association rule mining method based on secret protection of the embodiment of the present application
Structure schematic diagram, as shown in figure 5, the equipment includes one or more processors and memory.By taking a processor as an example, this sets
It is standby to include：Input unit and output device.
Processor, memory, input unit and output device can be connected by bus or other modes, in Fig. 5 with
Exemplified by being connected by bus.
Memory as a kind of non-transient computer readable storage medium storing program for executing, available for store non-transient software program, it is non-temporarily
State computer executable program and module.Processor is by running non-transient software program stored in memory, instruction
And module, so as to perform various function application and the data processing of electronic equipment, that is, realize the place of above method embodiment
Reason method.
Memory can include storing program area and storage data field, wherein, storing program area can storage program area, extremely
A few required application program of function；Storage data field can store data etc..In addition, memory can be included at a high speed at random
Memory is accessed, can also include non-transient memory, a for example, at least disk memory, flush memory device or other are non-
Transient state solid-state memory.In certain embodiments, memory is optional including relative to the remotely located memory of processor, this
A little remote memories can pass through network connection to processing unit.The example of above-mentioned network includes but not limited to internet, enterprise
In-house network, LAN, mobile radio communication and combinations thereof.
Input unit can receive the numeral or character information of input, and produce signal input.Output device may include to show
The display devices such as display screen.
One or more of modules are stored in the memory, are performed when by one or more of processors
When, perform the following operation of any of the above-described embodiment of the method：
Step a：Sender treats mining data collection and carries out can search for encrypting, and encrypted data set to be excavated is uploaded
To cloud server terminal；
Step b：Sending direction cloud server terminal sends association rule mining request, and is submitted to cloud server terminal and excavate key
Word；
Step c：Cloud server terminal is calculated according to the excavation keyword using the association rule mining based on difference secret protection
Method is associated rule digging to the encrypted data set to be excavated.
The said goods can perform the method that the embodiment of the present application is provided, and possesses the corresponding function module of execution method and has
Beneficial effect.Not ins and outs of detailed description in the present embodiment, reference can be made to method provided by the embodiments of the present application.
The embodiment of the present application provides a kind of non-transient (non-volatile) computer-readable storage medium, and the computer storage is situated between
Matter is stored with computer executable instructions, which can perform following operation：
Step a：Sender treats mining data collection and carries out can search for encrypting, and encrypted data set to be excavated is uploaded
To cloud server terminal；
Step b：Sending direction cloud server terminal sends association rule mining request, and is submitted to cloud server terminal and excavate key
Word；
Step c：Cloud server terminal is calculated according to the excavation keyword using the association rule mining based on difference secret protection
Method is associated rule digging to the encrypted data set to be excavated.
The embodiment of the present application provides a kind of computer program product, and the computer program product includes being stored in non-temporary
Computer program on state computer-readable recording medium, the computer program include programmed instruction, when described program instructs
When being computer-executed, the computer is set to perform following operation：
Step a：Sender treats mining data collection and carries out can search for encrypting, and encrypted data set to be excavated is uploaded
To cloud server terminal；
Step b：Sending direction cloud server terminal sends association rule mining request, and is submitted to cloud server terminal and excavate key
Word；
Step c：Cloud server terminal is calculated according to the excavation keyword using the association rule mining based on difference secret protection
Method is associated rule digging to the encrypted data set to be excavated.
The association rule mining method based on secret protection, system and electronic equipment of the embodiment of the present application, which utilize, symmetrically may be used
Search Encryption Algorithm is treated mining data and is encrypted, and has ensured the security of data storage, while utilize difference privacy
The association rule mining method of protection treats mining data and is associated rule digging, it is ensured that in the process of data mining, in
Between the full-range privacy of data in process of caching and anaphase issuing process.
The foregoing description of the disclosed embodiments, enables professional and technical personnel in the field to realize or using the application.
A variety of modifications to these embodiments will be apparent for those skilled in the art, as defined herein
General Principle can be realized in other embodiments in the case where not departing from spirit herein or scope.Therefore, the application
The embodiments shown herein is not intended to be limited to, and is to fit to and the principles and novel features disclosed herein phase one
The most wide scope caused.
Claims (17)
- A kind of 1. association rule mining method based on secret protection, it is characterised in that including：Step a：Sender treats mining data collection and carries out can search for encrypting, and encrypted data set to be excavated is uploaded to cloud Server-side；Step b：Sending direction cloud server terminal sends association rule mining request, and is submitted to cloud server terminal and excavate keyword；Step c：Cloud server terminal utilizes the association rules mining algorithm pair based on difference secret protection according to the excavation keyword The encrypted data set to be excavated is associated rule digging.
- 2. the association rule mining method according to claim 1 based on secret protection, it is characterised in that in the step Further included before a：Sending direction cloud server terminal submits registration request, and cloud server terminal is created according to the registration request for the sender Corresponding storage region, and start the server that data service is provided for the sender；The storage region is used to store described The data set to be excavated that sender uploads.
- 3. the association rule mining method according to claim 2 based on secret protection, it is characterised in that in the step In a, described sender treats mining data collection and carries out can search for encryption further including：Sender treats mining data and is cleaned, The User ID and user information keyword in data to be excavated are extracted, forms data set to be excavated.
- 4. the association rule mining method according to claim 3 based on secret protection, it is characterised in that in the step In a, described sender, which treats mining data collection and carries out can search for encryption, is specially：Wait to dig to described using can search for Encryption Algorithm All user information keywords in pick data set are encrypted；Described sender is treated mining data collection and is encrypted, and will Encrypted data set to be excavated is uploaded to cloud server terminal and specifically includes：Step a1：The data set to be excavated is pre-encrypted, obtains pre-add ciphertext data Cpre；Step a2：By pre-add ciphertext data CpreN-m bits and m bits are divided into, and utilizes the pseudorandom of password generation n-m bits Sequence S；Step a3：Use password generation pseudorandom values k；Step a4：Using pseudo-random sequence S as parameter, using the value of pseudo-random function F and f generation m bits, Salt values are formedStep a5：By pre-add ciphertext data CpreWith Salt values TiExclusive or, obtains original cipher text data set；Step a6：Original cipher text data set is uploaded onto the server.
- 5. the association rule mining method according to claim 4 based on secret protection, it is characterised in that in the step In a6, it is described original cipher text data set is uploaded onto the server after further include：Step a7：Server pre-processes original cipher text data set, obtains pre-processed results table；The pretreatment is i.e. with original User ID in beginning ciphertext data set is key assignments, and the user information keyword is flocked together；Step a8：After original cipher text data set and pre-processed results table are upset order by server at random respectively, it is collectively stored in In the corresponding storage region of the sender.
- 6. the association rule mining method according to claim 5 based on secret protection, it is characterised in that in the step It is described to be specially to cloud server terminal submission excavation keyword in b：Sender specifies excavation keyword, can search for adding using symmetrical Generation after the excavation keyword encryption is excavated keyword trapdoor by close algorithm, and the excavation keyword trapdoor of generation is submitted to Server；Wherein, the excavation keyword is corresponding with user information keyword.
- 7. the association rule mining method according to claim 6 based on secret protection, it is characterised in that in the step In c, the cloud server terminal is according to the excavation keyword, using the association rules mining algorithm based on difference secret protection to institute State encrypted data set to be excavated and be associated rule digging and further include：By the excavation keyword trapdoor and pre-processed results table In user information keyword matched, and judge matching whether all success, if matching all success, according to the digging Pick keyword trapdoor is associated rule digging to the original cipher text data set；Otherwise excavation keyword is returned to sender to fall into Door it fails to match information.
- 8. the association rule mining method according to claim 7 based on secret protection, it is characterised in that in the step It is described that the encrypted data set to be excavated is closed using the association rules mining algorithm based on difference secret protection in c Connection rule digging specifically includes：Step c1：According to keyword trapdoor is excavated, original cipher text data set is filtered and sorted, forms new ciphertext data Collect D*；Step c2：According to new ciphertext data set D*Frequent pattern tree (fp tree) is constructed, is searched for by frequent pattern tree (fp tree) qualified frequent The support of pattern and each frequent mode, and choose the frequent mode set that support counting is not less than threshold value min_count Cset；Step c3：K are picked out from frequent mode set Cset be easiest to the frequent of privacy leakage occur using index mechanism Set of patterns；Step c4：Noise is added to the support counting of k fuzzy frequent itemsets；Step c5：The support counting of the k fuzzy frequent itemsets to adding noise carries out consistency constraint；Step c6：Added up to using noise counter set and calculate correlation rule index；Step c7：Association rule mining result is returned to sender according to correlation rule index result of calculation.
- 9. a kind of association rule mining system based on secret protection, it is characterised in that including terminal device and cloud server terminal；The terminal device includes：Data encryption module：Carry out can search for encrypting for treating mining data collection, and by encrypted data set to be excavated Reach cloud server terminal；Data mining request module：For sending association rule mining request to cloud server terminal；Keyword submits module：Keyword is excavated for being submitted to cloud server terminal；The cloud server terminal includes：Association rule mining module：For after association rule mining request is received, according to the excavation keyword, utilizing base Rule digging is associated to the encrypted data set to be excavated in the association rules mining algorithm of difference secret protection.
- 10. the association rule mining system according to claim 9 based on secret protection, it is characterised in that the terminal Equipment further includes：Registration request module：For submitting registration request to cloud server terminal；Cloud server terminal further includes described in user information：Registration module：For creating corresponding storage region according to the registration request for the terminal device, and it is the end to start End equipment provides the server of data service；The storage region is used to store the data set to be excavated.
- 11. the association rule mining system according to claim 10 based on secret protection, it is characterised in that the terminal Equipment further includes：Data processing module：Cleaned for treating mining data, extract User ID and user information in data to be excavated Keyword, forms data set to be excavated.
- 12. the association rule mining system according to claim 11 based on secret protection, it is characterised in that the data Encrypting module, which treats mining data collection and carries out can search for encryption, is specially：Using can search for Encryption Algorithm to the data to be excavated All user information keywords concentrated are encrypted；The data encryption module specifically includes：Pre-encrypt unit：For the data set to be excavated to be pre-encrypted, pre-add ciphertext data C is obtainedpre；Data partitioning unit：For by pre-add ciphertext data CpreN-m bits and m bits are divided into, and utilizes password generation n-m ratios Special pseudo-random sequence S；Pseudorandom values generation unit：For using password generation pseudorandom values k；Random salt figure generation unit：For using pseudo-random sequence S as parameter, m bits to be generated using pseudo-random function F and f Value, form Salt valuesXor operation unit：For by pre-add ciphertext data CpreWith Salt values TiExclusive or, obtains original cipher text data set, and will be original Ciphertext data set is uploaded onto the server.
- 13. the association rule mining system according to claim 12 based on secret protection, it is characterised in that the cloud clothes Business end further includes：Data preprocessing module：For being pre-processed to original cipher text data set, pre-processed results table is obtained；The pretreatment I.e. using the ID in original cipher text data set as key assignments, the user information keyword is flocked together；Data memory module：After original cipher text data set and pre-processed results table are upset order at random respectively, deposit together Storage is in the corresponding storage region of the terminal device.
- 14. the association rule mining system according to claim 13 based on secret protection, it is characterised in that the key Word submits module to excavate keyword to cloud server terminal submission：Specify and excavate keyword, encryption calculation is can search for using symmetrical Generation after the excavation keyword encryption is excavated keyword trapdoor by method, and the excavation keyword trapdoor of generation is submitted to cloud clothes Business end；Wherein, the excavation keyword is corresponding with user information keyword.
- 15. the association rule mining system according to claim 14 based on secret protection, it is characterised in that the cloud clothes Business end further includes：Keywords matching module：For by the user information keyword in excavation keyword trapdoor and the pre-processed results table into Row matching, and judge matching whether all success, if matching all successes, by the association rule mining module to described Original cipher text data set is associated rule digging；Otherwise, returned to terminal device and excavate keyword trapdoor it fails to match information.
- 16. the association rule mining system according to claim 15 based on secret protection, it is characterised in that the association Rule digging module specifically includes：Sequencing unit：For according to keyword trapdoor is excavated, original cipher text data set being filtered and being sorted, formed new close Literary data set D*；Frequent mode chooses unit：For according to new ciphertext data set D*Frequent pattern tree (fp tree) is constructed, is searched for by frequent pattern tree (fp tree) The support of qualified frequent mode and each frequent mode, and choose support counting and be not less than threshold value min_count Frequent mode set Cset；Frequent mode screening unit：It is easiest to occur for picking out k from frequent mode set Cset using index mechanism The fuzzy frequent itemsets of privacy leakage；Noise adding device：For adding noise to the support counting of k fuzzy frequent itemsets；Consistency constraint unit：Support counting for the k fuzzy frequent itemsets to adding noise carries out consistency constraint；Correlation rule exponent calculation unit：Correlation rule index is calculated for adding up to using noise counter set, and according to correlation rule Index result of calculation returns to association rule mining result to terminal device.
- 17. a kind of electronic equipment, including：At least one processor；AndThe memory being connected with least one processor communication；Wherein,The memory storage has the instruction that can be performed by one processor, and described instruction is by least one processor Perform, so that at least one processor is able to carry out above-mentioned correlation rule of 1 to 8 any one of them based on secret protection The following operation of method for digging：Step a：Sender treats mining data collection and carries out can search for encrypting, and encrypted data set to be excavated is uploaded to cloud Server-side；Step b：Sending direction cloud server terminal sends association rule mining request, and is submitted to cloud server terminal and excavate keyword；Step c：Cloud server terminal utilizes the association rules mining algorithm pair based on difference secret protection according to the excavation keyword The encrypted data set to be excavated is associated rule digging.
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 true CN108022654A (en) | 2018-05-11 |
| CN108022654B 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) |
Cited By (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 |
| CN109784721A (en) * | 2019-01-15 | 2019-05-21 | 东莞市友才网络科技有限公司 | A kind of plateform system of employment data analysis and data mining analysis |
| CN110221784A (en) * | 2019-06-12 | 2019-09-10 | 青岛大学 | A kind of data duplicate removal method based on item response theory, device and equipment |
| CN111680329A (en) * | 2020-08-14 | 2020-09-18 | 成都中轨轨道设备有限公司 | Data processing method for improving data security |
| CN112597524A (en) * | 2021-03-03 | 2021-04-02 | 支付宝(杭州)信息技术有限公司 | Privacy intersection method and device |
| CN113141329A (en) * | 2020-01-16 | 2021-07-20 | 中移(上海)信息通信科技有限公司 | Big data mining method, device, equipment and storage medium |
| CN113536383A (en) * | 2021-01-27 | 2021-10-22 | 支付宝(杭州)信息技术有限公司 | Method and device for training neural network based on privacy protection |
| CN113779630A (en) * | 2021-09-09 | 2021-12-10 | 新疆大学 | DICOM-based CT medical image reversible desensitization method |
| CN114564749A (en) * | 2022-03-04 | 2022-05-31 | 聊城保磊计算机科技有限公司 | 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 |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104050267A (en) * | 2014-06-23 | 2014-09-17 | 中国科学院软件研究所 | Individuality recommendation method and system protecting user privacy on basis of association rules |
| CN104765848A (en) * | 2015-04-17 | 2015-07-08 | 中国人民解放军空军航空大学 | Symmetrical searchable encryption method for supporting result high-efficiency sequencing in hybrid cloud storage |
| CN106503575A (en) * | 2016-09-22 | 2017-03-15 | 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 |
| 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 |
| CN107239693A (en) * | 2017-05-17 | 2017-10-10 | 杰创智能科技股份有限公司 | A kind of analysis method and system based on password coding rule |
-
2017
- 2017-12-20 CN CN201711391275.2A patent/CN108022654B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104050267A (en) * | 2014-06-23 | 2014-09-17 | 中国科学院软件研究所 | Individuality recommendation method and system protecting user privacy on basis of association rules |
| CN104765848A (en) * | 2015-04-17 | 2015-07-08 | 中国人民解放军空军航空大学 | Symmetrical searchable encryption method for supporting result high-efficiency sequencing in hybrid cloud storage |
| CN106503575A (en) * | 2016-09-22 | 2017-03-15 | 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 |
| CN107239693A (en) * | 2017-05-17 | 2017-10-10 | 杰创智能科技股份有限公司 | A kind of 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 |
Non-Patent Citations (6)
| Title |
|---|
| CHRISTIAN NEUHAUS等: "Secure Keyword Search over Data Archives in the Cloud Performance and Security Aspects of Searchable Encryption", 《THE 5TH INTERNATIONAL CONFERENCE ON CLOUD COMPUTING AND SERVICES SCIENCE》 * |
| DAWN XIAODONG SONG等: "Practical Techniques for Searches on Encrypted Data", 《PROCEEDING 2000 IEEE SYMPOSIUM ON SECURITY AND PRIVACY》 * |
| HAIBAT JADHAV等: "Association Rule Mining Methods for Applying Encryption Techniques in Transaction Dataset", 《2016 INTERNATIONAL CONFERENCE ON COMPUTER COMMUNICATION AND INFORMATICS》 * |
| 周李威: "基于全同态加密的隐私保护关联规则研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
| 安全: "可搜索加密1", 《HTTPS://BLOG.CSDN.NET/YALUNDONG/ARTICLE/DETAILS/72792973》 * |
| 廖勇: "满足差分隐私的关联规则挖掘算法研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Cited By (15)
| 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 |
| CN109784721A (en) * | 2019-01-15 | 2019-05-21 | 东莞市友才网络科技有限公司 | A kind of plateform system of employment data analysis and data mining analysis |
| 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 |
| CN110221784A (en) * | 2019-06-12 | 2019-09-10 | 青岛大学 | A kind of data duplicate removal method based on item response theory, device and equipment |
| CN113141329A (en) * | 2020-01-16 | 2021-07-20 | 中移(上海)信息通信科技有限公司 | Big data mining method, device, equipment and storage medium |
| 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 |
| CN111680329A (en) * | 2020-08-14 | 2020-09-18 | 成都中轨轨道设备有限公司 | Data processing method for improving data security |
| 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 |
| CN113536383A (en) * | 2021-01-27 | 2021-10-22 | 支付宝(杭州)信息技术有限公司 | Method and device for training neural network based on privacy protection |
| CN113536383B (en) * | 2021-01-27 | 2023-10-27 | 支付宝(杭州)信息技术有限公司 | Method and device for training graph neural network based on privacy protection |
| CN112597524A (en) * | 2021-03-03 | 2021-04-02 | 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 |
| CN114564749A (en) * | 2022-03-04 | 2022-05-31 | 聊城保磊计算机科技有限公司 | 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 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108022654B (en) | 2021-11-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108022654A (en) | A kind of association rule mining method based on secret protection, system and electronic equipment | |
| Dauterman et al. | Waldo: A private time-series database from function secret sharing | |
| CN111698088B (en) | Key alternation method, key alternation device, electronic equipment and medium | |
| CN105049196B (en) | The encryption method that multiple keywords of designated position can search in cloud storage | |
| CN108092766B (en) | Ciphertext search authority verification method and system | |
| CN112000632B (en) | Ciphertext sharing method, medium, sharing client and system | |
| Iliev et al. | Private information storage with logarithmic-space secure hardware | |
| CN103607420A (en) | Safe electronic medical system for cloud storage | |
| CN110413652A (en) | A kind of big data privacy search method based on edge calculations | |
| JP2004147218A (en) | Data division management method and program | |
| JP5388727B2 (en) | Information processing system, information processing apparatus, server apparatus, information processing method, and program | |
| CN115473703B (en) | Authentication ciphertext equivalent test method, device, system and medium based on identity | |
| CA3189628A1 (en) | Systems and methods for tracking propagation of sensitive data | |
| CN115765965A (en) | Medical data safety sharing method based on federal learning and double union block chains | |
| CN114598472A (en) | Conditional-hidden searchable agent re-encryption method based on block chain and storage medium | |
| CN116318617A (en) | Medical rescue material charity donation method based on RFID and blockchain | |
| CN113779597B (en) | Method, device, equipment and medium for storing and similar searching of encrypted document | |
| CN110535630B (en) | Key generation method, device and storage medium | |
| de Loaysa Babiano et al. | Evaluation of live forensic techniques, towards Salsa20-Based cryptographic ransomware mitigation | |
| CN109740378A (en) | A kind of safety of anti-keyword privacy leakage is to index construction and its search method | |
| WO2019104674A1 (en) | Ciphertext search authority verification method and system therefor | |
| Amorim et al. | Homomorphic Encryption: An Analysis of its Applications in Searchable Encryption | |
| CN113034139A (en) | Block chain multi-currency wallet based on living body biological feature authentication and implementation method thereof | |
| Zhang et al. | PPsky: Privacy-preserving skyline queries with secret sharing in ehealthcare | |
| Smriti et al. | Secure File Storage in Cloud Computing Using a Modified Cryptography Algorithm |
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 |