A sensitive frequent information hiding method, device, system and medium

Abstract

The application provides a sensitive frequent information hiding method, device, system and medium, and belongs to the technical field of privacy protection data mining, and comprises the following steps: combining an original transaction database, mining a frequent item set according to a given support threshold, and generating a sensitive frequent item set from the frequent item set; the purpose of the application is to obtain a target database through an optimization algorithm data cleaning according to the data, to reduce the support of the sensitive item set to the given threshold, so as to achieve a hiding effect, to balance data privacy and data utility under the condition of the same mining threshold, and to reduce data privacy leakage and retain normal utility of the data as much as possible in the cleaning process.

Description

Technical Field

[0001] This invention relates to a method, apparatus, system, and medium for hiding sensitive and frequently accessed information, belonging to the field of privacy-preserving data mining technology. Background Technology

[0002] With the development of data science, data mining techniques have been widely applied in various fields to extract valuable information and patterns from massive amounts of data. Against this backdrop, frequent itemset mining, as an important task in data mining, has attracted widespread attention from researchers. Frequent itemset mining aims to discover itemsets that appear frequently in a dataset. These frequent itemsets often contain key patterns and sensitive associations hidden in the data, and have dual significance in practical applications: for example, in the medical field, they may reveal a strong correlation between patient privacy and disease characteristics; in the financial field, they may expose the correlation patterns between customer identity information and transaction behavior; and in e-commerce, they may identify the special consumption preferences of high-value users. However, with increasingly stringent data privacy regulations, how to effectively hide sensitive information when conducting frequent itemset mining has become an important issue that urgently needs to be addressed.

[0003] When processing massive amounts of data, practitioners need to extract necessary information, and data mining algorithms are constantly being improved and refined to meet these demands. However, the strong association rules involved in frequent itemset mining often contain sensitive information, and these highly supported association patterns are easily exploited maliciously during processing. In an era where data value is increasingly prominent, reverse engineering attacks targeting sensitive frequent patterns are becoming more frequent, posing a serious risk of privacy breaches. Sensitive frequent itemset hiding, as a core technology to address this challenge, has become an important research direction in the field of privacy-preserving data mining in recent years.

[0004] The hiding of sensitive frequent itemsets faces unique technical challenges. Existing research, both domestically and internationally, has yielded numerous methods for hiding sensitive frequent itemsets. Some of these studies lean towards traditional rule-based processing, while others employ machine learning optimization algorithms to hide sensitive information. This research addresses the issues of machine learning algorithms in incurring significant database loss or having limited effectiveness in hiding sensitive information.

[0005] Existing methods often require significant deletion of original data transactions to hide sensitive and frequent itemsets when processing datasets, leading to obvious damage to the database structure. When the data distortion exceeds a critical value, a contradictory phenomenon occurs where "hiding failure" and "loss of true patterns" coexist. How to effectively hide sensitive and frequent patterns while maximizing the accuracy of data mining results remains a technical bottleneck that urgently needs to be overcome. Summary of the Invention

[0006] The purpose of this invention is to provide a method, apparatus, system, and medium for hiding sensitive and frequent information. By identifying and selecting appropriate target transactions (the set of transaction numbers to be deleted corresponding to the victim item) and sensitive items (the victim item), the sensitive and frequent itemset is hidden, thereby achieving privacy protection.

[0007] To achieve the above objectives / to solve the above technical problems, the present invention is implemented using the following technical solution.

[0008] On one hand, the present invention provides a method for hiding sensitive and frequently accessed information, comprising:

[0009] The frequent itemsets of a transaction are obtained based on the support of the transaction in the original transaction database, and the frequent itemsets are divided into sensitive frequent itemsets containing victim items and non-sensitive frequent itemsets according to the set requirements.

[0010] Based on whether the transactions in the original transaction database contain victim entries, transactions that do not contain victim entries are directly input into the target database;

[0011] The transaction cleansing process that includes victim items is hidden, and the transactions after hiding the victim items are input into the target database to form the final target database;

[0012] Specifically, the method for obtaining the victim item includes:

[0013] Calculate the candidate count of all "1" itemsets in the sensitive frequent itemset;

[0014] Arrange the "1" itemsets according to the candidate count, and the "1" itemsets with the candidate count from smallest to largest are the victim items. Delete all sensitive and frequent itemsets containing the current victim item until the sensitive and frequent itemsets are empty, determine all victim items, and number them in order.

[0015] The cleaning process specifically includes:

[0016] The victim items, after being ordered, are used as sub-particles. The original transaction database is then cleaned based on the size of each sub-particle and the protozoan optimization algorithm strategy.

[0017] Furthermore, the step of obtaining frequent itemsets from the original transaction database based on the support of the data information specifically includes:

[0018] The data information in the original transaction database is divided into frequent itemsets and infrequent itemsets based on the set support threshold.

[0019] Furthermore, the calculation of the candidate count of the "1" itemset in the sensitive frequent itemset specifically includes:

[0020] ;

[0021] in, and For custom parameters, For candidate count, The support of the "1" itemset in the sensitive frequent itemset. The support of the "1" itemset in the non-sensitive frequent itemset.

[0022] The potential damage to the database from modifying a sensitive item is quantified by the candidate count; the smaller the candidate count, the less damage, and the larger the candidate count, the more damage.

[0023] Furthermore, the method for calculating the size of the sub-particles is expressed as follows:

[0024] ;

[0025] in, Let i be the size of the sub-particle of the i-th victim item. This is the scaling factor. Represents the i-th victim item. Let p be the p-th sensitive frequent itemset in S, where S is the set of sensitive frequent itemsets.

[0026] This is a function that maximizes the support of frequent itemsets containing victim items.

[0027] This represents the sensitive frequent itemset where the i-th victim item exists.

[0028] To find the minimum support threshold for frequent itemsets,

[0029] |D| represents the size of the original transaction database.

[0030] By designing the size of sub-particles, we can ensure that sensitive items are removed as much as possible, which also prevents data utility loss due to excessive deletion. Generally, n defaults to 1, which ensures that the support of the sensitive itemset is reduced to below the minimum threshold without excessive deletion that would result in data utility loss. The larger n is, the better the confidentiality, but the greater the loss of data utility. The smaller n is, the worse the confidentiality, but the data utility will be preserved.

[0031] Furthermore, the step of cleaning the original transaction database by combining the size of each sub-particle with the protozoan optimization algorithm strategy to obtain the target database hiding sensitive and frequent information specifically includes:

[0032] Particles are composed of each sub-particle size combined with a pre-built sensitive transaction retrieval table:

[0033] Wherein: the sensitive transaction retrieval table is a set of transaction numbers that appear in the original transaction database corresponding to the victim item; the sub-particle is a victim item and a target transaction number; the target transaction number is a set of transaction numbers to be deleted in the original transaction database corresponding to the victim item; the number of transaction numbers in the set is the size of the sub-particle corresponding to the victim item; and the particle is a set of all the sub-particles.

[0034] Calculate and sort the fitness value of each particle. The particle with the smallest fitness value after sorting is the victim item of the preliminary global optimal solution and its corresponding target transaction number.

[0035] A set proportion of particles are selected to enter the dormancy or reproduction phase, while particles that do not enter the dormancy or reproduction phase will enter the foraging phase.

[0036] The expression for obtaining particles entering the dormant or reproductive phase is as follows:

[0037] ;

[0038] in, For particles to enter dormancy or reproduce, For population size, To customize the maximum hibernation or breeding ratio, A random number between 0 and 1 For population randomization function,

[0039] The fitness values ​​of each particle entering the dormancy or reproduction phase are sorted to calculate the dormancy parameter. Then, the particle's entry into the dormancy or reproduction phase is determined based on the dormancy parameter. The expression is:

[0040] ;

[0041] ;

[0042] in, These are the hibernation parameters. The fitness ranking of particles that have entered the dormant or reproductive phase within the population;

[0043] Particles that enter the dormant phase will be deleted and replaced by new particles that are randomly generated. Particles that enter the reproduction phase will modify the sub-particles of the initial non-global optimal solution in the internal part.

[0044] The foraging parameters are calculated based on the random numbers generated for particles entering the foraging phase and the number of iterations. These parameters determine whether a particle enters the autotrophic or heterotrophic phase. The expression is as follows:

[0045] ;

[0046] ;

[0047] in, For foraging parameters, This is the current iteration number. The maximum number of iterations;

[0048] During the foraging phase, each particle replaces a set proportion of the transactions of each of its subparticles with new transactions; during the autotrophic phase, particles will select new transactions from the initial set of non-global optimal solution candidates for replacement, and during the heterotrophic phase, particles will select new transactions from the initial global optimal solution for replacement.

[0049] The fitness values ​​of all particles after the replacement are calculated to obtain a new global optimal solution. The global optimal solution is then updated. The next iteration will perform transaction replacement based on the new global optimal solution. The above replacement process is repeated until the maximum number of iterations is reached and the optimal solution is obtained, thus completing the cleaning of the original transaction database.

[0050] By classifying particles into different stages such as foraging, dormancy, and reproduction, and iteratively optimizing within the maximum number of iterations, combined with particle fitness values, the cleaning effect is improved, so that a particle with the optimal fitness value is determined during the particle update process, which is the best solution.

[0051] Furthermore, the fitness value is calculated using the following expression:

[0052] ;

[0053] in: Let be the fitness value of the i-th particle. , , For custom parameters, and , Let be the total number of sensitive frequent itemsets that the i-th particle failed to hide during the hiding process. Let be the total number of insensitive frequent itemsets lost by the i-th particle during the hiding process. Let be the total number of infrequent itemsets incorrectly mined by the i-th particle.

[0054] In a second aspect, the present invention provides a device for hiding sensitive and frequently accessed information, comprising:

[0055] The acquisition module is used to obtain the frequent itemsets of a transaction based on the support of the transaction in the original transaction database, and to divide the frequent itemsets into sensitive frequent itemsets containing victim items and non-sensitive frequent itemsets according to the set requirements.

[0056] The processing module is used to determine whether the transactions in the original transaction database contain victim items. Transactions that do not contain victim items are directly input into the target database; transactions that contain victim items are cleaned; and transactions that hide victim items are input into the target database to form the final target database.

[0057] Specifically, the method for obtaining the victim item includes:

[0058] Calculate the candidate count of all "1" itemsets in the sensitive frequent itemset;

[0059] Arrange the "1" itemsets according to the candidate count, and the "1" itemsets with the candidate count from smallest to largest are the victim items. Delete all sensitive and frequent itemsets containing the current victim item until the sensitive and frequent itemsets are empty, determine all victim items, and number them in order.

[0060] The cleaning process specifically includes:

[0061] The victim items, after being ordered, are used as sub-particles. The original transaction database is then cleaned based on the size of each sub-particle and the protozoan optimization algorithm strategy.

[0062] Thirdly, the present invention provides a sensitive and frequently used information hiding system, characterized in that it includes:

[0063] Memory, used to store computer programs / instructions;

[0064] A processor for executing the computer program / instructions to implement the steps of the above-described sensitive and frequent information hiding method.

[0065] Fourthly, the present invention provides a computer-readable storage medium having a computer program / instructions stored thereon, characterized in that, when the computer program / instructions are executed by a processor, they implement the steps of the above-described sensitive and frequently accessed information hiding method.

[0066] Fifthly, the present invention provides a computer program product, including a computer program / instruction, characterized in that, when the computer program / instruction is executed by a processor, it implements the steps of the above-described sensitive and frequent information hiding method.

[0067] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: by cleaning the privacy database and inputting the transactions after hiding the victim items into the target database, the present invention can achieve a better effect in protecting the privacy information of electronic health records; by reducing the support of sensitive itemsets, the same minimum frequency mining threshold is applied without generating false costs.

[0068] This invention, by selecting victim items with smaller Icount values ​​and combining them with appropriate victim transactions obtained through iteration, results in less data loss (lower loss cost) when processing datasets. It also significantly reduces data loss in dense datasets while preserving a certain level of data utility.

[0069] This invention achieves good results in preserving itemsets and the overall database structure by choosing a cleaning strategy that deletes victim items within the victim transaction, instead of deleting the entire transaction; it can ensure that the overall database structure is above 99.99%. Attached Figure Description

[0070] Figure 1 This is a schematic diagram of the process of the present invention;

[0071] Figure 2 This is a schematic diagram of the cleaning process of the present invention. Detailed Implementation

[0072] It should be noted that:

[0073] Hiding Failure: This refers to situations where hiding privacy itemsets fails under the same threshold parameters. In other words, frequent itemsets that should have been hidden are still being discovered. For sensitive and frequent itemsets extracted from the cleaned database, generally, make This represents sensitive frequent itemsets that were not hidden during the hiding process. The total number is expressed by the following formula: ;

[0074] Loss Cost: This refers to the situation where non-sensitive privacy itemsets that should have been discovered after the cleansing process were not. NSRI' represents the non-sensitive frequent itemsets discovered after database cleansing. Generally, it allows... This represents the non-sensitive frequent itemsets lost during the hiding process. )total: ;

[0075] False costs: After cleaning, itemsets that shouldn't be frequent itemsets are discovered. This is partly due to false information generated by database modifications. Generally, let... This represents the total number of infrequent itemsets that were incorrectly mined. ;

[0076] Itemset support: Itemset I in transactional databases Support in It represents an itemset. Number of times it appears in the transaction database;

[0077] Itemset frequency: The frequency of itemset I in the transaction database is expressed as... It means, set Represents database The total number of transactions, then ;

[0078] Frequent itemsets: given parameters ( When an itemset The frequency in the database meets the condition. Then we can call it an itemset. For frequent itemsets ;

[0079] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations thereof. In the absence of conflict, the embodiments of the present invention and the technical features in the embodiments can be combined with each other.

[0080] The term "and / or" simply describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. Additionally, the character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0081] Example 1

[0082] like Figure 1 One embodiment shown provides a method for hiding sensitive and frequently accessed information, including:

[0083] Step 1: Obtain the frequent itemsets of the transaction based on the support of the transaction in the original transaction database, and divide the frequent itemsets into sensitive frequent itemsets containing victim items and non-sensitive frequent itemsets according to the set requirements;

[0084] Step 1.1: The original transaction database obtained in this embodiment is an electronic health record database.

[0085] , Represents a unique transaction number, in which Representing the Each transaction contains a further transaction. . Representing the first in this transaction Itemsets; if a certain itemset contains items... If the set is k, then this set is called the "k" itemset, as shown in Table 1.

[0086] Table 1 Electronic Health Record Database

[0087] ;

[0088] As shown in the table above, it consists of 10 electronic health records;

[0089] Step 1.1: Set the frequent itemset mining threshold (support threshold) to 0.4. Then, according to the frequent itemset mining rules, the frequent itemsets of this database are obtained, and the results are shown in Table 2:

[0090] Table 2 Frequent Itemsets

[0091] ;

[0092] Based on the principle of hidden requirements, {a2}, {c1}, {c1, g1} are set as sensitive frequent itemsets.

[0093] Step 2: Based on whether the transactions in the original transaction database contain victim entries, directly input the transactions that do not contain victim entries into the target database;

[0094] Step 3: Input the transactions containing victim entries into the privacy database, clean the privacy database, and input the transactions after hiding the victim entries into the target database to form the final target database;

[0095] Specifically, the method for obtaining the victim item includes:

[0096] Step 3.1: Calculate the candidate count of the "1" itemset in the sensitive frequent itemset, specifically including:

[0097] ;

[0098] in, and For custom parameters, For candidate count, The support of the "1" itemset in the sensitive frequent itemset. The support of the "1" itemset in the non-sensitive frequent itemset;

[0099] In this implementation ;

[0100] Step 3.2: Arrange the "1" itemsets according to the candidate count, with the "1" itemsets in ascending order of candidate count as victim items. Delete all sensitive frequent itemsets containing "1" itemsets until the sensitive frequent itemsets are empty. Determine all victim itemsets and their sub-particle sizes: and And number them sequentially according to their Icount values: c1, a2;

[0101] Step 3.3: Construct a sensitive item and victim item index table. The sensitive transaction retrieval table is a set of transaction IDs corresponding to the sensitive frequent itemsets in the original transaction database, such as {a2}→{ , },{c1}→ .

[0102] like Figure 2 As shown, the cleaning process specifically includes:

[0103] Step 3.4: Using the sequenced victim items as sub-particles, construct particles by combining the size of each sub-particle;

[0104] The method for calculating the size of the sub-particles is expressed as follows:

[0105] ;

[0106] in, Let i be the size of the sub-particle of the i-th victim item. This is the scaling factor. Representing the current victims, Let p be the p-th sensitive frequent itemset in S, where S is the set of sensitive frequent itemsets.

[0107] This is a function that maximizes the support of frequent itemsets containing victim items.

[0108] This represents the sensitive frequent itemset where the i-th victim item exists.

[0109] To find the minimum support threshold for frequent itemsets,

[0110] |D| represents the original database size;

[0111] Wherein: the sub-particle is a victim item and a target transaction number, the target transaction number is the set of transaction numbers to be deleted corresponding to the victim item in the original transaction database, the number of the transaction number set is the size of the sub-particle corresponding to the victim item, and the particle is the set of all the sub-particles;

[0112] Step 3.5: Clean the original transaction database according to the protozoan optimization algorithm strategy (APO2DT algorithm strategy) based on the size of each sub-particle.

[0113] Step 3.51: The cleaning phase begins. Based on Step 3, the initialized particles and their fitness values ​​are obtained. Assume the population size is 3, containing 3 particles P1, P2, and P3. The parameters are set as follows: , , The fitness value is calculated using the fitness function formula:

[0114] ;

[0115] , ,

[0116] .

[0117] Based on the fitness values ​​of each particle, the particle with the smallest fitness value is identified as the transaction number of the preliminary global optimal solution, as shown in Table 3:

[0118] Table 3 Initialized Population Representation

[0119] ;

[0120] Step 3.52:

[0121] A certain percentage of particles will be selected to enter the dormancy or reproduction phase, while particles that do not enter the dormancy or reproduction phase will enter the foraging phase.

[0122] The expression for obtaining particles entering the dormant or reproductive phase is as follows:

[0123]

[0124] in, For particles to enter dormancy or reproduce, For population size, To customize the maximum hibernation or breeding ratio, A random number between 0 and 1 For population randomization function,

[0125] The fitness values ​​of each particle entering the dormancy or reproduction phase are sorted to calculate the dormancy parameter. Then, the particle's entry into the dormancy or reproduction phase is determined based on the dormancy parameter. The expression is:

[0126] ;

[0127] ;

[0128] in, A random number between (0, 1) These are the hibernation parameters. The fitness value of a particle that has entered a dormant or reproductive phase is ranked within the population. Population size;

[0129] Particles that enter the dormant phase will be deleted and replaced by new particles that are randomly generated. Particles that enter the reproduction phase will modify the sub-particles of the initial non-global optimal solution in the internal part.

[0130] The foraging parameters are calculated based on the random numbers generated for particles entering the foraging phase and the number of iterations. These parameters determine whether a particle enters the autotrophic or heterotrophic phase. The expression is as follows:

[0131] ;

[0132] ;

[0133] in, A random number between (0, 1) For foraging parameters, This is the current iteration number. The maximum number of iterations;

[0134] To select the global optimal solution, since the fitness values ​​are the same, a particle P2 is randomly selected as the optimal solution. The three particles generate corresponding random parameters and proceed to the next stage. Here, according to the formula: Since the results are obtained randomly, we assume here that... With P2. Generate random numbers with P2, assuming Random parameters less than Parameters, therefore Entering the dormant phase. The P2 random parameter is greater than... The parameters are then processed, and the particle enters the reproduction phase. Other particles enter the autotrophic and heterotrophic selection phase. Similarly, random parameters are generated for P3, assuming the random parameters of P3 are less than... Parameters, P3 enters heterotrophic operation.

[0135] Based on the above, P1 performs dormancy, P2 performs reproduction, and P3 performs heterotrophic operations, therefore:

[0136] Table 4 Population representation after iteration

[0137] ;

[0138] Therefore, particles Since there is a minimum fitness value, P3 is the optimal solution after iteration.

[0139] Step 3.53: Repeat the above process until the maximum iteration value is reached to obtain the optimal result, such as the solution provided by particle P3: delete sensitive item c1 from transactions T2, T7, T8; delete sensitive item a2 from transactions T3, T10;

[0140] Based on the optimal solution, the original database was cleaned, resulting in Table 5:

[0141] Table 5 Target Database After Complete Cleaning

[0142] ;

[0143] Step 3.54: Target database To verify this, frequent itemset mining was performed on the target database in Table 5, with the threshold also set to 0.4, resulting in Table 6:

[0144] Table 6 Frequent itemsets obtained from the target database

[0145] ;

[0146] As shown in the table above, the target database Sensitive information has been hidden, while more non-sensitive information has been retained.

[0147] This invention achieves good results in preserving itemsets and the overall database structure by choosing a cleaning strategy that deletes victim items within the victim transaction, instead of deleting the entire transaction; it can ensure that the overall database structure is above 99.99%.

[0148] Example 2

[0149] Based on the sensitive and frequently accessed information hiding method described in Embodiment 1, this embodiment provides a sensitive and frequently accessed information hiding device, including:

[0150] The acquisition module is used to obtain the frequent itemsets of a transaction based on the support of the transaction in the original transaction database, and to divide the frequent itemsets into sensitive frequent itemsets containing victim items and non-sensitive frequent itemsets according to the set requirements.

[0151] The processing module is used to determine whether the transactions in the original transaction database contain victim items. Transactions that do not contain victim items are directly input into the target database; transactions that contain victim items are cleaned; and transactions that hide victim items are input into the target database to form the final target database.

[0152] Specifically, the method for obtaining the victim item includes:

[0153] Calculate the candidate count of all "1" itemsets in the sensitive frequent itemset;

[0154] Arrange the "1" itemsets according to the candidate count, and the "1" itemsets with the candidate count from smallest to largest are the victim items. Delete all sensitive and frequent itemsets containing the current victim item until the sensitive and frequent itemsets are empty, determine all victim items, and number them in order.

[0155] The cleaning process specifically includes:

[0156] The victim items, after being ordered, are used as sub-particles. The original transaction database is then cleaned based on the size of each sub-particle and the protozoan optimization algorithm strategy.

[0157] Example 3

[0158] Based on the sensitive and frequently used information hiding method described in Embodiment 1, this embodiment provides a sensitive and frequently used information hiding system, including:

[0159] Memory, used to store computer programs / instructions;

[0160] A processor for executing the computer program / instructions to implement the steps of the sensitive and frequent information hiding method described in Embodiment 1.

[0161] Example 4

[0162] Based on the sensitive and frequent information hiding method described in Embodiment 1, this embodiment provides a computer-readable storage medium storing a computer program / instruction thereon. When the computer program / instruction is executed by a processor, it implements the steps of the sensitive and frequent information hiding method described in Embodiment 1.

[0163] Example 5

[0164] Based on the sensitive and frequent information hiding method described in Embodiment 1, this embodiment provides a computer program product, including a computer program / instruction, characterized in that the computer program / instruction, when executed by a processor, implements the steps of the sensitive and frequent information hiding method described in Embodiment 1.

[0165] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0166] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0167] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0168] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0169] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims. All of these forms are within the protection scope of the present invention.

Claims

1. A sensitive frequent information hiding method, characterized in that, The application relates to a method for constructing a target database from an original transaction database. The method comprises the following steps: According to the support degree of a transaction in the original transaction database, the frequent item set of the transaction is obtained, and the frequent item set is divided into a sensitive frequent item set containing a victim item and a non-sensitive frequent item set according to a set requirement; According to whether the transaction in the original transaction database contains the victim item, the transaction not containing the victim item is directly input into the target database; The transaction containing the victim item is hidden through a cleaning process, and the transaction after the victim item is hidden is input into the target database to form a final target database; The method for obtaining the victim item comprises the following steps: The candidate count of all "1" item sets in the sensitive frequent item set is calculated; The "1" item sets are arranged according to the candidate count, and the "1" item sets with the candidate count from small to large are sequentially taken as the victim items; all the sensitive frequent item sets containing the current victim item are deleted until the sensitive frequent item set is empty, all the victim items are determined, and the victim items are sequentially coded; The cleaning process comprises the following steps: The coded victim items are taken as sub-particles respectively, and the original transaction database is cleaned according to the protozoan optimization algorithm strategy combined with the size of each sub-particle to hide the victim items, ； wherein, and is a self-defined parameter, is a candidate count, is the support of the "1" itemset in the sensitive frequent itemset, is the support of the "1" itemset in the non-sensitive frequent itemset, The method for calculating the candidate count of the "1" item set in the sensitive frequent item set comprises the following steps: ； wherein, is a sub-particle size of the ith victim item, is a scaling factor, represents the ith victim item, is the pth sensitive frequent itemset in S, S is a set of sensitive frequent itemsets, is a function of the maximum support of the frequent itemsets containing the victim item, a sensitive frequent itemset representing the existence of the i-th victim item, mining frequent itemsets, The calculation method of the size of the sub-particle is expressed as: |D| is the size of the original transaction database, The method for cleaning the original transaction database combined with the size of each sub-particle according to the protozoan optimization algorithm strategy to hide the victim items comprises the following steps: Each sub-particle size is combined with a pre-constructed sensitive transaction retrieval table to form a particle: The sensitive transaction retrieval table is a set of transaction numbers in which the victim item appears in the original transaction database, the sub-particle is a victim item and a target transaction number, the target transaction number is a set of transaction numbers corresponding to the victim item to be deleted in the original transaction database, the number of the set of transaction numbers is the size of the sub-particle, and the particle is a set of all the sub-particles; The fitness value of each particle is calculated and sorted, and the particle with the minimum fitness value after sorting is the victim item of the preliminary global optimal solution and the corresponding target transaction number; Particles in a set proportion are selected to enter a dormancy or breeding stage, and the particles not entering the dormancy or breeding stage enter a foraging stage; ； wherein, is the number of particles that enter dormancy or reproduce, is the population size, is a custom preset maximum dormancy or reproduction ratio, is a random number between 0 and 1, is a population randomization function; The expression of the particles entering the dormancy or breeding stage is as follows: ； ； wherein, is the dormancy parameter, is the rank of the fitness value of the particle entering the dormancy or reproductive phase in the population; The fitness value of each particle entering the dormancy or breeding stage is sorted to calculate a dormancy parameter, and the particle enters the dormancy or breeding stage according to the dormancy parameter, and the expression is as follows: The particle entering the dormancy stage is deleted and directly replaced by a newly generated particle, and the particle entering the breeding stage modifies the internal preliminary non-global optimal solution sub-particle; The foraging parameter is calculated according to the random number and the iteration number of the particle entering the foraging stage, and the particle enters an autotrophy or heterotrophy stage according to the foraging parameter, and the expression is as follows: ； ； wherein, is a foraging parameter, is a current iteration number, is a maximum iteration number; During the foraging phase, each particle replaces a set proportion of the transactions of each of its subparticles with new transactions; during the autotrophic phase, particles will select new transactions from the initial set of non-global optimal solution candidates for replacement, and during the heterotrophic phase, particles will select new transactions from the initial global optimal solution for replacement. The fitness values ​​of all particles after the replacement are calculated to obtain a new global optimal solution. The global optimal solution is then updated. The next iteration will perform transaction replacement based on the new global optimal solution. The above replacement process is repeated until the maximum number of iterations is reached, the optimal solution is obtained, the victim item is hidden, and the original transaction database is cleaned.

2. The sensitive frequent information hiding method of claim 1, wherein, The step of obtaining the frequent itemset of a transaction based on its support in the original transaction database specifically includes: The data information in the original transaction database is divided into frequent itemsets and infrequent itemsets based on the set support threshold.

3. The sensitive frequent information hiding method of claim 1, wherein, The fitness value is calculated using the following expression: ； wherein: is the fitness value of the ith particle, , , is a user-defined parameter, and , is the total number of sensitive frequent itemsets that the ith particle failed to hide in the hiding process, is the total number of non-sensitive frequent itemsets that the ith particle lost in the hiding process, is the total number of non-frequent itemsets that the ith particle mined by mistake.

4. A sensitive frequent information hiding device based on the method of claim 1, characterized by, include: The acquisition module is used to obtain the frequent itemsets of a transaction based on the support of the transaction in the original transaction database, and to divide the frequent itemsets into sensitive frequent itemsets containing victim items and non-sensitive frequent itemsets according to the set requirements. The processing module is used to determine whether the transactions in the original transaction database contain victim items. Transactions that do not contain victim items are directly input into the target database; transactions that contain victim items are cleaned; and transactions that hide victim items are input into the target database to form the final target database. Specifically, the method for obtaining the victim item includes: Calculate the candidate counts of all "1" itemsets in the sensitive frequent itemset; Arrange the "1" itemsets according to the candidate count, and the "1" itemsets with the candidate count from smallest to largest are the victim items. Delete all sensitive and frequent itemsets containing the current victim item until the sensitive and frequent itemsets are empty, determine all victim items, and number them in order. The cleaning process specifically includes: The victim items, after being ordered, are used as sub-particles. The original transaction database is then cleaned based on the size of each sub-particle and the protozoan optimization algorithm strategy.

5. A fragile steganographic system, characterized in that, include: Memory, used to store computer programs / instructions; A processor for executing the computer program / instructions to implement the steps of the sensitive and frequent information hiding method according to any one of claims 1-3.

6. A computer readable storage medium having stored thereon computer programs / instructions, characterized in that, When the computer program / instruction is executed by the processor, it implements the steps of the sensitive and frequent information hiding method according to any one of claims 1-3.

7. A computer program product comprising computer programs / instructions, characterized in that, When the computer program / instruction is executed by the processor, it implements the steps of the sensitive and frequent information hiding method according to any one of claims 1-3.

Images