User behavior data processing method and system based on privacy protection

Abstract

This application provides a privacy-preserving user behavior data processing method and system, relating to the field of information security technology. The method comprises: converting acquired user behavior data into an ordered utility sequence to form a utility sequence database; extracting high-utility sequence patterns from the utility sequence database and their position and utility information within the utility sequence; for each high-utility sequence pattern, obtaining the sum of the utilities of each item in the high-utility sequence pattern, selecting the item with the largest utility sum as the victim item, and modifying the utility of the victim item in the utility sequence to update the utility value of the high-utility sequence pattern until the utility value of the high-utility sequence pattern is less than a minimum utility threshold, at which point modification ceases; and sharing the modified utility sequence database. In this way, the utility of the original database can be maintained as much as possible while ensuring that sensitive information is hidden.

Description

Technical Field

[0001] The present application belongs to the field of information security technology, and in particular relates to a method and system for processing user behavior data based on privacy protection. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] With the advent of big data and computer technology, information about people's purchasing behavior in supermarkets and on e-commerce platforms can be collected and used for knowledge discovery. However, this data may contain sensitive information, leading to privacy leaks. When this sensitive knowledge is maliciously obtained by a third party, it can lead to profit losses for businesses and organizations. Privacy leaks have become an unavoidable problem. Therefore, protecting sensitive information from discovery and thereby better serving businesses is a pressing issue.

[0004] Mining high-utility sequence patterns is a key research topic in pattern recognition, with applications in various scenarios, including business decision-making and stock market analysis. Currently, many effective high-utility sequence pattern mining methods have been proposed to discover valuable utility sequence patterns, but this also carries the risk of information leakage. For example, user purchase behavior data may contain high-utility combination patterns that can support leadership decisions. However, if these combination patterns are discovered and exploited by competitors, the company may suffer losses. In this case, the data loses its confidentiality.

[0005] High-utility sequence pattern hiding is a branch of privacy-preserving data mining (PPDM). It is implemented by modifying the original database and hiding all high-utility sequence patterns found in the database under a given minimum utility threshold, so that others cannot find these patterns in the cleaned database using the same or higher minimum utility threshold. When using hiding methods on utility sequence databases, certain side effects will occur, namely utility loss. The purpose of high-utility sequence pattern hiding is to reduce the analyzability of shared data, ensure data privacy to a certain extent, and at the same time ensure the integrity of the database, so that the modified database can still be effectively mined using the same pattern mining framework. The inventors found that the existing high-utility sequence pattern hiding methods still face some problems: (1) The use of utility matrix and utility chain structure to store the relevant information of the original database and high-utility sequence patterns results in a lot of time spent on identifying the items to be modified (victim items) and their locations in the original database; (2) In order to identify and modify the victim items in the utility sequence database, the hiding process requires a large number of search operations; (3) The hiding process usually leads to a great loss of utility in the original database. Summary of the Invention

[0006] To overcome the deficiencies of the prior art, the present application provides a user behavior data processing method and system based on privacy protection, which is used to achieve privacy protection of user behavior data.

[0007] The technical solutions adopted in this application are as follows:

[0008] In a first aspect, an embodiment of the present invention provides a method for processing user behavior data based on privacy protection, comprising:

[0009] Convert the acquired user behavior data into an ordered utility sequence to form a utility sequence database;

[0010] Extracting a high-utility sequence pattern from the utility sequence database and its position and utility information of the utility sequence in which it is located, wherein the utility value of the high-utility sequence pattern is greater than or equal to a minimum utility threshold;

[0011] For each high-utility sequence pattern, obtain the sum of the utilities of each item in the high-utility sequence pattern, select the item with the largest utility sum as the victim item, and modify the utility of the victim item in the utility sequence to update the utility value of the high-utility sequence pattern until the utility value of the high-utility sequence pattern is less than the minimum utility threshold, then stop modifying;

[0012] A modified utility sequence database is shared to achieve privacy protection of user behavior data.

[0013] In one possible implementation, after extracting the high-utility sequence pattern from the utility sequence database and the position and utility information of the utility sequence in which it is located, the method further includes: assigning a sequence weight to all utility sequences in the utility sequence database according to the extracted high-utility sequence pattern; sorting the modification order of the target utility sequence according to the sequence weight, and preferentially selecting the target utility sequence with the largest sequence weight for modification.

[0014] In a possible implementation, the number of high-utility sequence patterns in each utility sequence is determined as the sequence weight of the utility sequence.

[0015] In one possible implementation, the external utility of the victim item is obtained, and a decline rate of the victim item is determined based on the external utility, the difference between the utility value of the high-utility sequence pattern and the minimum utility threshold, and the sum of the utility of the victim item; and the utility of the victim item in the utility sequence is modified based on the decline rate.

[0016] In one possible implementation, converting the acquired user behavior data into an ordered utility sequence to form a utility sequence database includes: acquiring the user's shopping information within a preset time period, the shopping information including shopping time, items, and consumption amount; treating all items purchased at the same shopping time and the corresponding consumption amount of each item as an item set of the utility sequence, and arranging the item set in order of shopping time to form an ordered utility sequence; and the utility sequences of multiple users constitute the utility sequence database.

[0017] In a possible implementation, a preset high-utility sequence pattern mining algorithm is used to extract high-utility sequence patterns from a utility sequence database and the position and utility information of the utility sequence in which they are located.

[0018] In a second aspect, an embodiment of the present application provides a user behavior data processing system based on privacy protection, comprising:

[0019] An acquisition module, used to convert the acquired user behavior data into an ordered utility sequence to form a utility sequence database;

[0020] an extraction module, configured to extract a high-utility sequence pattern from the utility sequence database and its position and utility information, wherein the utility value of the high-utility sequence pattern is greater than or equal to a minimum utility threshold;

[0021] a modification module configured to obtain, for each high-utility sequence pattern, the sum of the utilities of each item in the high-utility sequence pattern, select the item with the largest utility sum as the victim item, and modify the utility of the victim item in the utility sequence to update the utility value of the high-utility sequence pattern until the utility value of the high-utility sequence pattern is less than a minimum utility threshold, at which point the modification is stopped;

[0022] The sharing module is used to share the modified utility sequence database to achieve privacy protection of user behavior data.

[0023] In one possible embodiment, after the extraction module is used to extract the high-utility sequence pattern from the utility sequence database and the location and utility information of the utility sequence in which it is located, it also includes: assigning a sequence weight to all utility sequences in the utility sequence database according to the extracted high-utility sequence pattern; sorting the modification order of the target utility sequence according to the sequence weight, and preferentially selecting the target utility sequence with the largest sequence weight for modification.

[0024] In a third aspect, an embodiment of the present invention provides a computer device comprising: a processor, a memory, and a bus, wherein the memory stores machine-readable instructions executable by the processor. When the computer device is running, the processor and the memory communicate through the bus, and when the machine-readable instructions are executed by the processor, the steps of the user behavior data processing method based on privacy protection are performed as described in the first aspect and any possible implementation of the first aspect.

[0025] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the user behavior data processing method based on privacy protection as described in the first aspect and any possible implementation method of the first aspect are executed.

[0026] Beneficial effects of this application:

[0027] (1) This application extracts high-utility sequence patterns from a utility sequence database and the position and utility information of the utility sequence in which they are located, thereby obtaining an array-based utility array structure for use in the hiding process. This structure can quickly and accurately locate the victim items without rescanning the original database. In the hiding process, the utility of the high-utility sequence patterns is reduced by directly modifying the utility of the victim items instead of changing the number of victim items, thereby ensuring the consistency of the original database and the cleaned database. In this way, the utility of the original database is maintained as much as possible while ensuring that sensitive information is hidden.

[0028] (2) The concept of sequence weight is proposed to define the weight of the sequence in the original database, and according to the weight of the sequence, the modification order of the victim sequence is changed to reduce the number of searches in the hiding process and improve the hiding efficiency of user behavior data. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The accompanying drawings, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0030] Figure 1 This is a flowchart of a method for processing user behavior data based on privacy protection provided by an embodiment of the present application;

[0031] Figure 2 is a flowchart of a method for processing user behavior data based on privacy protection provided by another embodiment of the present application;

[0032] Figure 3 This is a schematic diagram of the structure of a user behavior data processing system based on privacy protection provided by an embodiment of the present application;

[0033] Figure 4 This is a schematic diagram of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0034] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0035] It should be noted that the following detailed descriptions are illustrative and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present invention belongs.

[0036] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof.

[0037] See also Figure 1 , Figure 1 This is a flowchart of a method for processing user behavior data based on privacy protection provided by an embodiment of the present application. Figure 1 As shown in , the user behavior data processing method based on privacy protection includes the following steps:

[0038] S101: Convert the acquired user behavior data into an ordered utility sequence to form a utility sequence database.

[0039] In a specific implementation, the embodiment of the present application uses the data of user-purchased items as the data source for mining and hiding, and takes the shopping information of 6 users within 3 months as an example, as shown in Tables 1 and 2. Table 1 is the unit price (also called external utility) of items sold in the store, where the letters represent the item ID; Table 2 is a sorted utility sequence database D of user-purchased items, where SID represents the user's identity, and the item (item) ID purchased by the user and the consumption amount u(ID) are represented by (ID, u(ID)).

[0040] Table 1 Unit prices of items sold in stores

[0041] Item a b c d e f unit price 4 3 5 2 1 3

[0042] Table 2 The organized utility sequence database D of items purchased by users

[0043] User SID User purchase sequence <![CDATA[S1]]> <[(a,8)(c,15)],[(a,12)(c,10)],[(a,16)(b,15)(d,8)],[(b,3)(e,3)]> <![CDATA[S2]]> <[(a,4)(e,3)],[(a,20)(b,9)(d,4)],[(b,6)(c,5)(d,8)(e,3)]> <![CDATA[S3]]> <[(f,12)],[(c,10)(d,6)],[(a,12)(e,3)],[(b,12)(d,10)]> <![CDATA[S4]]> <[(b,6)(c,15)],[(a,20)(f,3)],[(b,12)(d,8)(e,5)]> <![CDATA[S5]]> <[(a,16)(c,15)],[(a,8)(b,15)(c,10)(d,8)(e,2)]> <![CDATA[S6]]> <[(e,6)],[(a,20)(b,9)],[(a,12)(d,8)]>

[0044] In an embodiment of the present application, as an optional embodiment, the acquired user behavior data is converted into an ordered utility sequence to form a utility sequence database, including: acquiring the user's shopping information within a preset time period, the shopping information including: shopping time, items and consumption amount; taking all items purchased at the same shopping time and the consumption amount corresponding to each item as an item set of the utility sequence, and arranging the item set in the order of shopping time to form an ordered utility sequence; the utility sequences of multiple users constitute the utility sequence database.

[0045] In practice, as shown in the data in Table 2, all purchase information for a user within a preset time period is converted into an ordered utility sequence, denoted by <>. An itemset represents all items purchased simultaneously by the user at a specific point in time, denoted by []. For convenience, the items in an itemset are ordered. A user may purchase the same item at different times, but the price of the item may be the same or different, meaning that an item appears once or multiple times in the sequence. For example, user S1 in Table 2 purchased item c both the first and second time, but the price paid for each purchase was 15 and 10, respectively. The itemset [(a,8)(c,15)] represents the simultaneous purchase of items a and c, with a total value of 23(8+15).

[0046] S102: extracting a high utility sequence pattern from the utility sequence database and its position and utility information, wherein the utility value of the high utility sequence pattern is greater than or equal to a minimum utility threshold.

[0047] In the specific implementation, Definition 1: The utility of a utility sequence s is defined as the sum of the utilities of each item i in s, denoted as:

[0048] u(s) = ∑{u(i)|i is contained in s};

[0049] For example: u(S1)=8+15+12+10+16+15+8+3+3=90.

[0050] Definition 2: Given a sequence t = <w1,w2,...,w m > and utility sequence s= <v1,v2,...,v n >, if m = n, and is included in each item set w k (1≤k≤m) and the terms contained in v k If the items in are the same, then t matches s, which is defined as t~s. For example: the sequence<a(bd)> There are two matches in S1, namely <[(a,8)][(b,15)(d,8)]> and <[(a,12)][(b,15)(d,8)]>.

[0051] Definition 3: The utility of sequence t in utility sequence s is the sum of the utilities of the items in s' that match t. When there are multiple matches, the one with the largest sum of utilities is selected as the utility of t in s, which is defined as

[0052]

[0053] For example: Sequence<a(bd)> The utilities of the two matches in S1 are 31(8+15+8) and 35(12+15+8), so the sequence<a(bd)> The utility u(<a(bd)> , S1)=max{31,35}=35.

[0054] Definition 4: The utility of sequence t in the utility sequence database D is the sum of the utility of t in each utility sequence, defined as

[0055] For example: Sequence<a(bd)> The utility in the utility sequence database D is u(t,D)=u(t,S1)+u(t,S2)+u(t,S3)+u(t,S4)+u(t,S5)+u(t,S6)=35+34+34+40+39=182.

[0056] Most companies frequently collect user purchase behavior data over long periods of time and use high-utility sequential pattern mining methods to discover high-utility sequential patterns (HUSPs) for decision-making. To protect HUSPs from malicious acquisition by competing companies, they need to be hidden. Currently, most hiding methods are applied to quantitative transaction databases and cannot meet the requirements of utility sequence databases. The few HUSP hiding algorithms that exist result in significant utility losses.

[0057] In order to hide high-utility sequence patterns, two important problems need to be solved: (1) how to find the victim items in each HUSP; (2) how to determine the utility value of the victim items that need to be reduced in each sequence containing HUSP.

[0058] Based on this, the embodiment of the present application extracts a high utility sequence pattern from the utility sequence database and its position and utility information of the utility sequence in which it is located, where the utility value of the high utility sequence pattern is greater than or equal to the minimum utility threshold.

[0059] As an optional embodiment, a preset high-utility sequence pattern mining algorithm is used to extract high-utility sequence patterns from the utility sequence database, as well as the location and utility information of the utility sequence in which they are located. Here, the high-utility sequence pattern mining algorithm includes but is not limited to the HUSP-ULL (HUSP mining with UL-list) algorithm.

[0060] In the specific implementation, the user purchase behavior data is subjected to the high-utility pattern mining algorithm HUSP-ULL to obtain all high-utility sequence patterns HUSP and the corresponding UAH structures.

[0061] Definition 5: High-Utility Sequential Pattern

[0062] Given a minimum utility threshold, minutil, a subsequence is considered a HUSP if its total utility in the utility sequence database is greater than or equal to minutil. During HUSP mining using HUSP-ULL, this embodiment of the present application utilizes the UAH structure to store the location and utility information of each HUSP in the sequence, which is used for high-utility sequence pattern hiding.

[0063] Definition 6: PUL structure

[0064] PUL is used to store the utility value of each match of HUSP in the sequence and the position of the item in each match.

[0065] Definition 7: UAH structure

[0066] UAH records the SID of the sequence where HUSP is located and the corresponding PUL structure.

[0067] For example, given minutil=125,<a(bd)> The utility in Table 2 is 182, so<a(bd)> is a HUSP, and its corresponding UAH structure is shown in Table 3. SID represents<a(bd)> The ID of the sequence, a_pos, b_pos, d_pos represent the positions of items a, b, and d in each match in the sequence, mat_u represents the sum of the utilities of each match, and mu_ind represents the index of the maximum mat_u value.<a(bd)> There are two matches in S1. The utility value of the first match is 31, and the positions of each item in the match in S1 are 1, 6, and 7. Similarly, the utility value of the first match is 35, and the positions of each item in the match in S1 are 3, 6, and 7. The utility of the second match is the largest (35), so mu_ind is 2, as shown in Table 3:

[0068] Table 3<a(bd)> UAH structure

[0069]

[0070] In an embodiment of the present application, as an optional embodiment, after step S102, i.e., extracting the high-utility sequence pattern from the utility sequence database and the position and utility information of the utility sequence in which it is located, the method further includes: assigning a sequence weight to all utility sequences in the utility sequence database according to the extracted high-utility sequence pattern; sorting the modification order of the utility sequences according to the sequence weights of the utility sequences, and preferentially selecting the utility sequence with the largest sequence weight for modification.

[0071] Here, the embodiment of the present application sorts all high-utility sequence patterns in descending order according to sequence weights and calculates the sequence weights of the sequences. Optionally, the number of high-utility sequence patterns in each utility sequence is determined as the sequence weight of the utility sequence.

[0072] In the specific implementation, Definition 8: Sequence weight

[0073] The weight of a sequence s in the utility sequence database is expressed as the number n of high-utility sequence patterns contained in s, that is, SW(s)=n.

[0074] S103: For each high-utility sequence pattern, obtain the sum of the utilities of each item in the high-utility sequence pattern, select the item with the largest utility sum as the victim item, and modify the utility of the victim item in the utility sequence to update the utility value of the high-utility sequence pattern. The modification is stopped until the utility value of the high-utility sequence pattern is less than the minimum utility threshold.

[0075] In the specific implementation, for each HUSP, the IEM method is used to select a victim item and change the utility of the victim item to modify the utility of the HUSP until the utility of the HUSP is less than the minimum utility threshold. After that, the utility values ​​of all HUSPs are recalculated, and the HUSPs with a utility greater than or equal to the minimum utility threshold are hidden.

[0076] In an embodiment of the present application, as an optional embodiment, the external utility of the victim item is obtained, and the decline rate of the victim item is determined based on the external utility, the difference between the utility value of the high-utility sequence pattern and the minimum utility threshold, and the sum of the utility of the victim item; and the utility of the victim item in the utility sequence is modified based on the decline rate.

[0077] In the specific implementation, the utility sum Sum of each item in HUSP is calculated, and the item with the largest utility sum is selected as the victim item. The decline rate α of the victim item is calculated, where p(i vic ) represents the external utility of the victim's item (unit price of the item), u(i vic ,pos,s) means i vic The utility in the utility sequence s, pos is i vic The position in s. The specific formula is as follows:

[0078]

[0079]

[0080] S104: Sharing the modified utility sequence database to achieve privacy protection of user behavior data.

[0081] In specific implementation, the pseudo code of the HHUSP-SW algorithm proposed in this application is as follows:

[0082] Input: D: utility sequence database of items purchased by users; minutil: minimum utility threshold; utable: unit price of items; U: set of high utility sequence patterns and their corresponding UAH

[0083] Output: Cleaned database

[0084]

[0085]

[0086] 22: Return the cleaned database; the returned database can be used for sharing, and its flow chart is as follows Figure 2 As shown in .

[0087] The contributions of the entire algorithm process are as follows:

[0088] First, all high-utility sequence patterns are effectively mined through the UL-List structure to accelerate the calculation of the utility of candidate sequences and their descendants. On this basis, two pruning strategies are introduced to pre-prune unpromising candidate sequences, avoiding combinatorial explosion in the search space.

[0089] Secondly, in the process of hiding HUSP, an improved array-based structure, called UAH, is proposed to store the location and utility information of each HUSP, which can quickly and accurately locate the victim item and effectively avoid the rescanning of the original database.

[0090] Third, the concept of sequence weight is proposed, which rearranges the order of sequence modification in the original database and prioritizes the sequence with the largest weight as the victim sequence of modification.

[0091] Fourth, the item elimination method (IEM) is used to identify the victim items and reduce their utility, which provides a feasible solution to maintain the consistency between the original database and the cleaned database.

[0092] Based on the same application concept, the embodiment of the present application also provides a user behavior data processing system based on privacy protection, such as Figure 3As shown in FIG, the user behavior data processing system 300 based on privacy protection includes:

[0093] An acquisition module 300 is used to convert the acquired user behavior data into an ordered utility sequence to form a utility sequence database;

[0094] An extraction module 310 is configured to extract a high-utility sequence pattern from the utility sequence database and its position and utility information in the utility sequence in which it is located, wherein the utility value of the high-utility sequence pattern is greater than or equal to a minimum utility threshold;

[0095] a modification module 320 configured to obtain, for each high-utility sequence pattern, the sum of the utilities of each item in the high-utility sequence pattern, select the item with the largest utility sum as the victim item, and modify the utility of the victim item in the utility sequence to update the utility value of the high-utility sequence pattern until the utility value of the high-utility sequence pattern is less than a minimum utility threshold, at which point the modification stops;

[0096] The sharing module 330 is used to share the modified utility sequence database to achieve privacy protection of user behavior data.

[0097] In the embodiment of the present application, as an optional embodiment, after the extraction module 310 is used to extract the high-utility sequence pattern from the utility sequence database and the location and utility information of the utility sequence in which it is located, it further includes: assigning a sequence weight to all utility sequences in the utility sequence database according to the extracted high-utility sequence pattern; and rearranging the modification order of each utility sequence in the sequence database according to a preset rule based on the sequence weight of the utility sequence, and preferentially selecting the utility sequence with the largest sequence weight for modification.

[0098] See also Figure 4 , Figure 4 This is a schematic diagram of a computer device provided in an embodiment of the present application. Figure 4 As shown in , an embodiment of the present application further provides a computer device, wherein the computer device 400 includes a processor 410 , a memory 420 and a bus 430 .

[0099] The memory 420 stores machine-readable instructions executable by the processor 410. When the computer device 400 is running, the processor 410 communicates with the memory 420 via the bus 430. When the machine-readable instructions are executed by the processor 410, the above-mentioned Figure 1 and Figure 2 The steps of the privacy-protection-based user behavior data processing method in the method embodiment shown are specifically implemented in accordance with the method embodiment and will not be described in detail here.

[0100] Based on the same application concept, an embodiment of the present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the user behavior data processing method based on privacy protection described in the above method embodiment are executed.

[0101] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing related hardware through a computer program. The program can be stored in a computer-readable storage medium, and when executed, the program can include the processes in the above-described method embodiments. The storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0102] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A method for processing user behavior data based on privacy protection, characterized in that: include: Convert the acquired user behavior data into an ordered utility sequence to form a utility sequence database; Extracting high-utility sequence patterns from the utility sequence database and their position and utility information, where the utility value of the high-utility sequence pattern is greater than or equal to a minimum utility threshold, and using a UAH structure to store the position and utility information of each high-utility sequence pattern in the sequence for use in the high-utility sequence pattern hiding process; the UAH structure records the SID of the sequence in which the high-utility sequence pattern is located and the corresponding PUL structure, where SID represents the ID of the sequence containing the corresponding high-utility sequence pattern, a_pos, b_pos, and d_pos represent the positions of items a, b, and d in the sequence for each match, respectively, mat_u represents the sum of the utilities of each match, and mu_ind represents the index of the maximum mat_u value; For each high-utility sequence pattern, obtain the sum of the utilities of each item in the high-utility sequence pattern, select the item with the largest utility sum as the victim item, and modify the utility of the victim item in the utility sequence to update the utility value of the high-utility sequence pattern until the utility value of the high-utility sequence pattern is less than the minimum utility threshold, then stop modifying; Sharing a modified utility sequence database to achieve privacy protection of user behavior data; After extracting the high-utility sequence pattern from the utility sequence database and the position and utility information of the utility sequence in which it is located, the method further includes: assigning a sequence weight to all utility sequences in the utility sequence database according to the extracted high-utility sequence pattern; The modification order of the target utility sequence is sorted according to the sequence weight, and the target utility sequence with the largest sequence weight is preferentially selected for modification.

2. The method for processing user behavior data based on privacy protection according to claim 1, characterized in that: The number of high-utility sequence patterns in each utility sequence is determined as the sequence weight of the utility sequence.

3. The method for processing user behavior data based on privacy protection according to claim 1, characterized in that: The external utility of the victim item is obtained, and a decrease rate of the victim item is determined based on the external utility, a difference between the utility value of the high-utility sequence pattern and a minimum utility threshold, and the sum of the utility of the victim item; and the utility of the victim item in the utility sequence is modified based on the decrease rate.

4. The method for processing user behavior data based on privacy protection according to claim 1, characterized in that: The method converts the acquired user behavior data into an ordered utility sequence to form a utility sequence database, including: acquiring the user's shopping information within a preset time period, wherein the shopping information includes: shopping time, items, and consumption amount; treating all items purchased at the same shopping time and the corresponding consumption amount of each item as an item set of the utility sequence, and arranging the item set in the order of shopping time to form an ordered utility sequence; and the utility sequences of multiple users constitute the utility sequence database.

5. The method for processing user behavior data based on privacy protection according to claim 1, characterized in that: The preset high-utility sequence pattern mining algorithm is used to extract the high-utility sequence patterns in the utility sequence database and their location and utility information in the utility sequence.

6. A user behavior data processing system based on privacy protection, characterized in that: include: An acquisition module, used to convert the acquired user behavior data into an ordered utility sequence to form a utility sequence database; An extraction module is configured to extract high-utility sequence patterns from the utility sequence database and the position and utility information of the utility sequence in which they are located. The utility value of the high-utility sequence pattern is greater than or equal to a minimum utility threshold. The position and utility information of the sequence in which each high-utility sequence pattern is located is stored using a UAH structure for use in the high-utility sequence pattern hiding process. The UAH structure records the SID of the sequence in which the high-utility sequence pattern is located and the corresponding PUL structure. The SID represents the ID of the sequence containing the corresponding high-utility sequence pattern. a_pos, b_pos, and d_pos represent the positions of items a, b, and d in the sequence for each match, respectively. mat_u represents the sum of the utilities of each match. mu_ind represents the index of the maximum mat_u value. a modification module configured to obtain, for each high-utility sequence pattern, the sum of the utilities of each item in the high-utility sequence pattern, select the item with the largest utility sum as the victim item, and modify the utility of the victim item in the utility sequence to update the utility value of the high-utility sequence pattern until the utility value of the high-utility sequence pattern is less than a minimum utility threshold, at which point the modification is stopped; A sharing module, used to share the modified utility sequence database to achieve privacy protection of user behavior data; After extracting the high-utility sequence pattern from the utility sequence database and the position and utility information of the utility sequence in which it is located, the method further includes: assigning a sequence weight to all utility sequences in the utility sequence database according to the extracted high-utility sequence pattern; The modification order of the target utility sequence is sorted according to the sequence weight, and the target utility sequence with the largest sequence weight is preferentially selected for modification.

7. A computer device, characterized in that: include: A processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor. When the computer device is running, the processor and the memory communicate via the bus. When the machine-readable instructions are executed by the processor, the steps of the user behavior data processing method based on privacy protection as described in any one of claims 1 to 6 are performed.

8. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, which, when executed by a processor, executes the steps of the method for processing user behavior data based on privacy protection according to any one of claims 1 to 6.

Images