Method, system and apparatus for determining objects based on privacy protection

Abstract

The present specification provides a method, system and device for determining an object based on privacy protection, applied to a first system; the first system comprises a first-party device and a second-party device, and the method comprises: the first-party device obtaining attribute information corresponding to each of a plurality of reference objects; the plurality of reference objects comprise a seed object and a non-seed object; the attribute information corresponding to any reference object is used to indicate whether the reference object belongs to the seed object; the first-party device homomorphically encrypts each attribute information to obtain each of the plurality of reference objects corresponding to each of the attribute information in a ciphertext state, and sends the attribute information in the ciphertext state to the second-party device; the second-party device calculates the ciphertext correlation degree between each of a plurality of candidate objects and the seed object based on the attribute information in the ciphertext state; the first-party device obtains an object number obtained based on the ciphertext correlation degree from the second-party device, and obtains target information from the second-party device based on the object number; the target information is used to determine a target object.

Description

Technical Field

[0001] This specification relates to one or more embodiments in the field of secure computing, and particularly to a method, system, and apparatus for determining objects based on privacy protection. Background Technology

[0002] With the continuous development of information technology, the relationships between people, people and organizations, and organizations are becoming increasingly close and complex. Other objects can be categorized and filtered based on the relationships between different objects and known seed objects. However, in some scenarios, the data on objects and their relationships, as well as the seed list, may be held by different institutions. Therefore, filtering objects closely related to seed objects from this data carries the risk of privacy breaches. Currently, there is a need for a privacy-preserving solution to identify target objects from object and relationship data. Summary of the Invention

[0003] This specification provides one or more embodiments of a method, system, and apparatus for determining objects based on privacy protection.

[0004] According to a first aspect, a method for determining objects based on privacy protection is provided, applied to a first system; the first system includes a first-party device and a second-party device; the method includes:

[0005] The first-party device acquires attribute information corresponding to each of multiple reference objects; the multiple reference objects include seed objects and non-seed objects; the attribute information corresponding to any reference object is used to indicate whether the reference object belongs to a seed object;

[0006] The first-party device performs homomorphic encryption on each attribute information to obtain each encrypted attribute information corresponding to the plurality of reference objects, and sends the encrypted attribute information to the second-party device.

[0007] Based on the dense state attribute information, the second-party device calculates the dense state correlation between each candidate object among the multiple candidate objects and the seed object among the multiple reference objects;

[0008] The first device obtains the object number based on the dense state correlation from the second device, and obtains target information from the second device based on the object number; the target information is used to determine the target object.

[0009] According to a second aspect, a method for determining objects based on privacy protection is provided, applied to a first-party device included in a first system; the first system further includes a second-party device; the method includes:

[0010] Obtain the attribute information corresponding to each of multiple reference objects; the multiple reference objects include seed objects and non-seed objects; the attribute information corresponding to any reference object is used to indicate whether the reference object belongs to a seed object;

[0011] The attribute information is homomorphically encrypted to obtain the encrypted attribute information corresponding to each of the multiple reference objects, and the encrypted attribute information is sent to the second-party device; so that the second-party device can calculate the encrypted correlation between each candidate object among the multiple candidate objects and the seed object among the multiple reference objects based on the encrypted attribute information;

[0012] The object number obtained from the second-party device based on the dense state correlation is obtained, and target information is obtained from the second-party device based on the object number; the target information is used to determine the target object.

[0013] According to a third aspect, a method for determining objects based on privacy protection is provided, applied to a second-party device included in a first system; the first system further includes a first-party device; the method includes:

[0014] The first party device obtains encrypted attribute information corresponding to each of the multiple reference objects; wherein, the encrypted attribute information is obtained by the first party device through homomorphic encryption of the attribute information corresponding to each of the multiple reference objects; the multiple reference objects include seed objects and non-seed objects; the attribute information corresponding to any reference object is used to indicate whether the reference object belongs to the seed object;

[0015] Based on the dense state attribute information, the dense state correlation between each candidate object among the multiple candidate objects and the seed object among the multiple reference objects is calculated, so that the first device can obtain the object number based on the dense state correlation.

[0016] Provide the first-party device with target information for determining the target object based on the object number.

[0017] According to the fourth aspect, a system for determining objects based on privacy protection is provided; the system includes a first-party device and a second-party device;

[0018] The first-party device is configured to acquire attribute information corresponding to each of the multiple reference objects, perform homomorphic encryption on the attribute information to obtain encrypted attribute information corresponding to each of the multiple reference objects, and send the encrypted attribute information to the second-party device; wherein, the multiple reference objects include seed objects and non-seed objects; the attribute information corresponding to any reference object is used to indicate whether the reference object belongs to a seed object;

[0019] The second-party device is used to calculate the dense state correlation between each candidate object among the plurality of candidate objects and the seed object among the plurality of reference objects based on the dense state attribute information;

[0020] The first-party device is further configured to obtain an object number based on the dense state correlation from the second-party device, and obtain target information from the second-party device based on the object number; the target information is used to determine the target object.

[0021] According to a fifth aspect, a device for determining objects based on privacy protection is provided, applied to a first-party device included in a first system; the first system further includes a second-party device; the device includes:

[0022] The acquisition module is used to acquire the attribute information corresponding to each of multiple reference objects; the multiple reference objects include seed objects and non-seed objects; the attribute information corresponding to any reference object is used to indicate whether the reference object belongs to a seed object;

[0023] An encryption module is used to perform homomorphic encryption on each attribute information to obtain each encrypted attribute information corresponding to each of the plurality of reference objects, and send the encrypted attribute information to the second-party device; so that the second-party device can calculate the encrypted correlation between each candidate object among the plurality of candidate objects and the seed object among the plurality of reference objects based on the encrypted attribute information;

[0024] The determination module is used to obtain an object number from the second-party device based on the dense state correlation, and to obtain target information from the second-party device based on the object number; the target information is used to determine the target object.

[0025] According to a sixth aspect, a device for determining an object based on privacy protection is provided, applied to a second-party device included in a first system; the first system further includes a first-party device; the device includes:

[0026] The acquisition module is used to acquire each encrypted attribute information corresponding to a plurality of reference objects from the first party device; wherein, each encrypted attribute information is obtained by the first party device through homomorphic encryption of each attribute information corresponding to the plurality of reference objects; the plurality of reference objects include seed objects and non-seed objects; the attribute information corresponding to any reference object is used to indicate whether the reference object belongs to a seed object;

[0027] The calculation module is used to calculate the dense state correlation between each candidate object among the multiple candidate objects and the seed object among the multiple reference objects based on the dense state attribute information, so that the first-party device can obtain the object number based on the dense state correlation.

[0028] A module is provided to the first-party device to provide target information for determining the target object based on the object number.

[0029] According to a seventh aspect, a computer-readable storage medium is provided, the storage medium storing a computer program that, when executed by a processor, implements the method described in any one of the first, second, or third aspects above.

[0030] According to an eighth aspect, a computing device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, implements the method described in any one of the first, second, or third aspects above.

[0031] The technical solutions provided in the embodiments of this specification may include the following beneficial effects:

[0032] The method and apparatus for determining objects based on privacy protection provided in the embodiments of this specification involve a first-party device acquiring attribute information corresponding to multiple reference objects, including seed objects and non-seed objects. The attribute information corresponding to any reference object is used to indicate whether the reference object belongs to the seed object. The first-party device homomorphically encrypts each attribute information and sends the resulting encrypted attribute information to a second-party device. Based on the encrypted attribute information, the second-party device calculates the encrypted correlation between each candidate object and the seed object. The first-party device obtains an object number from the second-party device based on the encrypted correlation, and based on the object number, obtains target information from the second-party device for determining the target object. Since the reference objects include both seed objects and non-seed objects, and the first-party device homomorphically encrypts the attribute information corresponding to the reference objects, there is no need to expose the seed object information to the second-party device, thereby protecting user privacy, reducing the risk of privacy leakage, and improving data security.

[0033] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0034] To more clearly illustrate the technical solutions of the embodiments in this specification, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0035] Figure 1This specification is a schematic diagram illustrating a scenario for determining objects based on privacy protection, according to an exemplary embodiment.

[0036] Figure 2 This is a flowchart illustrating a method for determining objects based on privacy protection, according to an exemplary embodiment.

[0037] Figure 3 This specification illustrates a system block diagram for determining objects based on privacy protection, according to an exemplary embodiment.

[0038] Figure 4 This specification is a block diagram of an apparatus for determining objects based on privacy protection, according to an exemplary embodiment.

[0039] Figure 5 This is a block diagram of another device for determining objects based on privacy protection, as illustrated in this specification according to an exemplary embodiment. Detailed Implementation

[0040] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this specification.

[0041] In the following description, when referring to the accompanying drawings, the same numbers in different drawings denote the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0042] The terminology used in this disclosure is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. The singular forms “a,” “the,” and “the” as used herein are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.

[0043] It should be understood that although the terms first, second, third, etc., may be used in this disclosure to describe various information, such information should not be limited to these terms. These terms are used only to distinguish information of the same type from one another. For example, without departing from the scope of this disclosure, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."

[0044] With the continuous development of information technology, the relationships between individuals, organizations, and other entities are becoming increasingly close and complex. Graph-based techniques can be used to study these relationships and, based on these relationships and known seed entities, to classify and filter other entities. For example, given multiple known malicious users (i.e., seed users), a graph can be used to filter out potentially malicious users from a large pool of users based on their relationships with these known malicious users.

[0045] In some scenarios, graph data and seed lists are held by different organizations. For example, Organization A holds a list of seed users, while Organization B holds the graph data. In some related technologies, Organization A can provide Organization B with the list of seed users. Organization B, based on the list, can then filter out target users closely associated with the seed users from the graph data and provide the target users' information to Organization A. However, this requires both organizations to expose their own user data, which carries the risk of privacy breaches if the user data involves privacy concerns.

[0046] This embodiment provides a privacy-preserving approach for determining objects. A first-party device acquires attribute information corresponding to multiple reference objects, including seed objects and non-seed objects. The attribute information of any reference object indicates whether it belongs to the seed object. The first-party device homomorphically encrypts each attribute information and sends the resulting encrypted attribute information to a second-party device. Based on this encrypted attribute information, the second-party device calculates the encrypted correlation between each candidate object and the seed object. The first-party device obtains an object number from the second-party device based on the encrypted correlation and, based on this object number, obtains target information from the second-party device to determine the target object. Since the reference objects include both seed and non-seed objects, and the first-party device homomorphically encrypts the attribute information corresponding to the reference objects, there is no need to expose the seed object information to the second-party device. This protects user privacy, reduces the risk of privacy leakage, and improves data security.

[0047] Figure 1This is a schematic diagram illustrating a scenario for determining objects based on privacy protection, according to an exemplary embodiment.

[0048] like Figure 1 As shown, Organization A holds a seed list containing multiple seed objects. Organization A can construct a reference list based on the seed list using device 101. This reference list can include information records corresponding to multiple reference objects. These reference objects can include both seed and non-seed objects. Seed objects are all real objects, while non-seed objects can include both real and obfuscated objects. The information record corresponding to any reference object can include its object identifier and attribute information. This attribute information indicates whether the reference object belongs to a seed node or a non-seed node. For example, a non-zero value can be used to indicate that the reference object is a seed node, and 0 can be used to indicate that it is a non-seed node. Then, Organization A, using device 101, uses a homomorphic encryption algorithm to homomorphically encrypt the attribute information of each reference object in the reference list, obtaining encrypted attribute information. The object identifier of each reference object in the reference list, along with its encrypted attribute information, is then sent to device 102 held by Organization B.

[0049] Organization B possesses graph data, which includes multiple nodes and edges between them. Each node corresponds to an object, and the node information of each node includes the object identifier of the object corresponding to that node. Organization B can acquire the graph data through device 102 and select at least a portion of the nodes in the graph data as candidate nodes, and the objects corresponding to the candidate nodes as candidate objects. For example, all nodes in the graph data can be selected as candidate nodes, or the graph data can be expanded k-degree by each reference object to obtain a subgraph, and all nodes in the subgraph can be selected as candidate nodes. Then, using a preset node importance algorithm (such as a personalized PageRank algorithm), the correlation between each candidate object and each reference object is calculated based on the object identifier of each reference object and the object identifier corresponding to the node in the graph data. Furthermore, based on the dense state attribute information corresponding to each reference object, the dense state correlation between each candidate object and the seed object in the reference objects is calculated.

[0050] Device 101 can obtain the node number of the node corresponding to the target object in the graph data from device 102 through secure computation between the two parties, and use it as the object number of the target object. Then, using a privacy information retrieval method, based on the object number of the target object, it retrieves the encrypted data of the target object from the second-party device and decrypts the encrypted data. Based on the data obtained after decryption, the target object is identified.

[0051] The solutions provided in this specification will be described in detail below with reference to specific embodiments.

[0052] like Figure 2 As shown, Figure 2 This is a flowchart illustrating a privacy-preserving method for determining objects according to an exemplary embodiment, applied to a first system. The first system may include a first-party device and a second-party device. The first-party device may be a device held by a first party providing seed objects, and the second-party device may be a device held by a second party providing candidate objects. Both the first and second parties may be institutions or organizations, etc. Both the first and second devices can be implemented as any device, platform, server, or device cluster with computing and processing capabilities. The method includes the following steps:

[0053] In step 201, the first-party device acquires the attribute information corresponding to each of the multiple reference objects.

[0054] In this embodiment, the objects involved may include, but are not limited to, natural persons, accounts, institutions, and organizations. Reference objects can be used as a reference for selecting target objects. Multiple reference objects may include seed objects and non-seed objects, and the attribute information corresponding to any reference object is used to indicate whether the reference object belongs to a seed object. Seed objects can be objects known to have preset characteristics or belong to preset categories, while non-seed objects can be ordinary objects that do not have preset characteristics or belong to preset categories. For example, seed objects may be individuals known to have engaged in fraudulent activities, or users known to have high value to preset business, while non-seed objects may be ordinary individuals or users.

[0055] In some implementations, the reference object may include a first object belonging to a designated group and a second object not belonging to the designated group. The first object includes seed objects and non-seed objects, while the second object consists entirely of non-seed objects. The designated group may be a group related to the first party providing the seed object; for example, the designated group may be a group comprised of members, customers, managed objects within its jurisdiction, or service objects of the first party. The first object may be a real object belonging to the designated group. The second object may be an object used for obfuscation; the second object does not belong to the designated group and can be a real object or a virtual object. For example, if the designated group is the customer of organization A, the second object may be a real customer of organization B or a virtual user. Because this embodiment incorporates a second object for obfuscation into the reference object, the security of privacy data is increased.

[0056] In step 202, the first device performs homomorphic encryption on each attribute information to obtain the encrypted attribute information corresponding to each of the multiple reference objects, and in step 203, sends the encrypted attribute information to the second device.

[0057] In this embodiment, the first-party device first generates a homomorphic encryption key, and then uses this homomorphic encryption key to homomorphically encrypt the attribute information corresponding to each reference object, thereby obtaining the encrypted attribute information corresponding to each reference object. For example, y can be used i Representing the attribute information of the i-th reference object, using a homomorphic encryption key to pair y i After encryption, the encrypted attribute information corresponding to the i-th reference object is obtained, which can be represented by E(y). i ) represents the dense state attribute information corresponding to the i-th reference object.

[0058] In this embodiment, after obtaining the encrypted attribute information corresponding to each reference object, the first-party device can send the encrypted attribute information to the second-party device. Specifically, the first-party device can send the object identifiers corresponding to each reference object and the encrypted attribute information to the second-party device in association. For example, the first-party device can generate a list of reference objects based on the object identifiers and encrypted attribute information corresponding to each reference object. The list can include information records corresponding to each reference object, and the first-party device can send the list of reference objects to the second-party device. The information record corresponding to any reference object can include the object identifier and the encrypted attribute of that reference object. For example, it can be used... <ID i ,E(y i )> represents the information record corresponding to the i-th reference object, where ID i Let be the object identifier of the i-th reference object. It should be noted that the object identifier of the reference object can be a unique identifier of the reference object. For example, the object identifier can include, but is not limited to, identity identifiers, telephone numbers, organization numbers, etc.

[0059] In step 204, the second-party device calculates the dense state correlation between each candidate object and the seed object among the multiple reference objects based on the aforementioned dense state attribute information.

[0060] In this embodiment, the second-party device can first determine multiple candidate objects. Specifically, the second-party device can acquire a target graph, which may be, for example, a knowledge graph. The target graph may include multiple nodes representing objects and connecting edges between related nodes. The second-party device can select objects corresponding to at least some of the nodes in the target graph as candidate objects.

[0061] In one implementation, all objects corresponding to nodes in the target graph can be obtained as candidate objects. In another implementation, the nodes corresponding to reference objects in the target graph can be extended by k degrees to obtain a subgraph of the target graph, and all objects corresponding to nodes in this subgraph can be used as candidate objects. It is understood that multiple candidate objects can also be determined by any other reasonable method, and this embodiment is not limited in this respect.

[0062] In this embodiment, firstly, the relevance between each candidate object and each reference object can be calculated. For example, based on the object identifier corresponding to each reference object, a preset node importance algorithm can be used to calculate the relevance between each candidate object and each reference object through the target graph. The preset node importance algorithm may include, but is not limited to, a personalized PageRank algorithm, a TrustRank algorithm, etc.

[0063] Then, based on the correlation between each candidate object and each reference object, and the dense-state attribute information corresponding to each reference object, the dense-state correlation between each candidate object and the seed objects among the multiple reference objects is calculated. For example, for each candidate object, the weighted sum of the correlations between the candidate object and each of the multiple reference objects can be calculated as the dense-state correlation between the candidate object and the seed objects among the multiple reference objects. The weight of the correlation between the candidate object and any reference object can be the dense-state attribute information corresponding to that reference object.

[0064] Specifically, see the following formula:

[0065]

[0066] Among them, D i y represents the dense-state correlation between the i-th candidate object and the seed object among multiple reference objects. j d represents the attribute information corresponding to the j-th reference object. ij This represents the relevance between the i-th candidate object and the j-th reference object. E(y) represents the plaintext relevance between the i-th candidate object and the seed objects among the J reference objects in the non-closed state. j ) represents the dense state attribute information corresponding to the j-th reference object. The formula can be... This is a formula for calculating the dense state correlation between the i-th candidate object and the seed objects among the J reference objects.

[0067] In this embodiment, since the reference objects include seed objects and non-seed objects, the non-seed objects serve to obfuscate information, thus protecting the privacy of seed objects to a certain extent. When calculating the closed-state correlation between a candidate object and the seed objects among multiple reference objects, the correlation between the candidate object and non-seed objects does not need to be considered. Therefore, for any reference object, if it is a non-seed object, its corresponding attribute information is a value of 0. If it is a seed object, its corresponding attribute information is a non-zero value. Thus, the plaintext correlation between any candidate object and the seed objects among multiple reference objects in the non-closed state is actually a weighted sum of the plaintext correlations between the candidate object and each seed object among the multiple reference objects.

[0068] Furthermore, in some embodiments, the attribute information corresponding to the seed object can be 1 or any other positive number. The attribute information corresponding to the seed object can take only one non-zero value, or the seed objects can be divided into multiple groups, and the attribute information corresponding to each group of seed objects can take different non-zero values ​​(the magnitude of the non-zero value can represent the magnitude of the weight). It is understood that this embodiment does not limit the specific values ​​of the attribute information corresponding to the seed object.

[0069] In step 205, the first device obtains the object number based on the dense state correlation from the second device, and obtains the target information for determining the target object from the second device based on the object number.

[0070] Specifically, the greater the correlation between the candidate object and the seed object in the non-closed state, the closer the candidate object is to the seed object, and the more likely it is to have the same preset characteristics or belong to the same preset category as the seed object. Therefore, the candidate objects can be filtered by judging the magnitude of the non-closed state correlation between the candidate object and the seed object. For example, candidate objects with a non-closed state correlation with the seed object greater than a preset threshold can be selected as target objects.

[0071] In this embodiment, after the second-party device calculates the encrypted correlation between each candidate object and a seed object among multiple reference objects, since the encrypted correlation is encrypted data, the second-party device cannot obtain the plaintext data of the encrypted correlation and therefore cannot perform a filtering operation on the candidate objects based on the encrypted correlation. The first-party device also cannot independently perform a filtering operation on the candidate objects based on the encrypted correlation. Therefore, a secure computation method involving both parties is required to filter the candidate objects.

[0072] Specifically, through secure computation between the two parties, the first-party device can obtain from the second-party device the object IDs corresponding to candidate objects whose non-closed-state correlation with the seed object is greater than a preset threshold. These object IDs are distinct from the object identifiers and can differentiate between different objects. It should be noted that the object identifier cannot be derived from the object ID.

[0073] For example, in one implementation, the second-party device can first obtain a preset threshold and a random positive number, and then generate the judgment ciphertext corresponding to each candidate object using the following formula:

[0074] M i =(D i -t)*r

[0075] Among them, M i D represents the ciphertext for the i-th candidate object. i Let represent the dense state correlation between the i-th candidate object and the seed object among multiple reference objects, t represent the preset threshold, and r be a random positive number.

[0076] The second-party device can send the ciphertext corresponding to each candidate object to the first-party device, as shown in the following equation: Therefore, the first-party device can target M. i Decrypt to obtain The value N i Among them, N i This represents the plaintext of the decision corresponding to the i-th candidate. Since r is a positive number, if N i If greater than or equal to 0, then If the correlation between the i-th candidate object and the seed object is greater than or equal to a preset threshold, it means that the correlation between the i-th candidate object and the seed object is greater than or equal to 0, and the candidate object can be used as the target object. If N i If less than 0, then If the value is less than a preset threshold, it means that the correlation between the i-th candidate object and the seed object is less than 0, and the candidate object cannot be used as the target object. Therefore, after decrypting the ciphertext, the first-party device can determine whether to use the candidate object as the target object by checking whether the value of the obtained plaintext is greater than 0.

[0077] It should be noted that when the second-party device sends the judgment ciphertext corresponding to any candidate object to the first-party device, it also sends the object number of the candidate object to the first-party device in association. Therefore, the first-party device can determine the object number of the target object based on the judgment ciphertext.

[0078] In other implementations, the object ID corresponding to the target object can be obtained through secret sharing or by obfuscating circuits. It is understood that this embodiment does not limit the specific method for obtaining the object ID corresponding to the target object.

[0079] However, since the ciphertext only includes the object number corresponding to the target object (for example, the object number corresponding to the i-th target object can be i, or it can be any other number that can distinguish different objects), and does not include the object identifier of the target object, after the first-party device makes a judgment based on the ciphertext corresponding to each candidate object, it can only obtain the object number of the target object. Therefore, it is also necessary to obtain the target information for determining the target object from the second-party device. This target information can be the object identifier of the target object, or information used to determine the object identifier.

[0080] Specifically, in one implementation, the first-party device can directly send a query request to the second-party device. The query request may include the object number of the target object. This instructs the second-party device to retrieve the object identifier of the target object based on the object number and return the object identifier to the first-party device.

[0081] In another implementation, the first device can also directly send a query request to the second device. The query request may include the object numbers of some or all of the candidate objects, including all target objects. This instructs the second device to query the corresponding object identifier based on the object numbers of the candidate objects and return it to the first device. The first device then determines the object identifier of the target object based on the object identifiers of the candidate objects.

[0082] In another implementation, Private Information Retrieval (PIR) can be used to retrieve target information from a second-party device based on the object ID. Because PIR is used, the second-party device cannot determine which objects among the candidate objects are the target objects, thus preventing it from deducing seed object information from the target objects and further enhancing the security of privacy data.

[0083] This embodiment provides a privacy-preserving method for determining objects. A first-party device acquires attribute information corresponding to multiple reference objects, including seed objects and non-seed objects. The attribute information of any reference object indicates whether it belongs to the seed object. The first-party device homomorphically encrypts each attribute information and sends the resulting encrypted attribute information to a second-party device. Based on the encrypted attribute information, the second-party device calculates the encrypted correlation between each candidate object and the seed object. The first-party device obtains the object number obtained from the second-party device based on the encrypted correlation and, based on the object number, obtains target information from the second-party device for determining the target object. Since the reference objects include both seed and non-seed objects, and the first-party device homomorphically encrypts the attribute information corresponding to the reference objects, there is no need to expose the seed object information to the second-party device, thereby protecting user privacy, reducing the risk of privacy leakage, and improving data security.

[0084] It should be noted that although the operations of the methods of the embodiments of this specification are described in a specific order in the above embodiments, this does not require or imply that these operations must be performed in that specific order, or that all the operations shown must be performed to achieve the desired result. On the contrary, the steps depicted in the flowcharts may be executed in a different order. Additionally or alternatively, certain steps may be omitted, multiple steps may be combined into one step, and / or one step may be broken down into multiple steps.

[0085] The following is a schematic illustration of this solution using a complete application example.

[0086] The application scenario could be as follows: Organization A possesses the phone numbers of multiple members of a known fraud ring, while telecommunications operator B possesses a user graph based on communication interactions between users. The user graph includes nodes representing users, and edges between nodes represent the communication relationships between users (e.g., an edge is constructed between user nodes that have communicated more than a preset number of times). Organization A wants to use the phone numbers of the known fraud ring members and the user graph held by telecommunications operator B to identify high-risk, suspicious individuals closely related to these members.

[0087] Specifically, organization A possesses device a, which can use multiple members of a known fraud ring as seed objects, multiple other individuals unrelated to the fraud ring as non-seed objects, and virtual members as non-seed objects, to obtain a reference object list including both seed and non-seed objects. Device a generates a list of reference objects, which includes the phone number (i.e., object identifier) ​​and attribute information for each reference object. The attribute information for seed objects is 1, and the attribute information for non-seed objects is 0.

[0088] Next, device a homomorphically encrypts the attribute information of each reference object to obtain encrypted attribute information, and sends the reference object's mobile phone number and encrypted attribute information in association to device b held by telecommunications operator B. Device b can use candidate users corresponding to at least some nodes in the user graph as candidate objects, and calculate the encrypted correlation between the candidate users and the fraud gang using a personalized PageRank algorithm based on the user graph, the reference object's mobile phone number, and the encrypted attribute information. Then, through secure computation by both parties, the first-party device obtains the user IDs of suspicious users closely related to the fraud gang among the candidate users. Finally, device a uses a privacy information retrieval method to obtain encrypted data from device b, and determines the mobile phone number of the suspicious user based on the encrypted data.

[0089] Corresponding to the foregoing embodiments of the method for determining objects based on privacy protection, this specification also provides embodiments of the apparatus for determining objects based on privacy protection.

[0090] like Figure 3 As shown, Figure 3 This specification illustrates a system block diagram for determining objects based on privacy protection according to an exemplary embodiment. The system may include: a first-party device 301 and a second-party device 302.

[0091] The first-party device 301 is used to acquire attribute information corresponding to each of the multiple reference objects, perform homomorphic encryption on each attribute information to obtain encrypted attribute information corresponding to each of the multiple reference objects, and send the encrypted attribute information to the second-party device. The multiple reference objects include seed objects and non-seed objects, and the attribute information corresponding to any reference object is used to indicate whether the reference object belongs to the seed object.

[0092] The second-party device 302 is used to calculate the dense state correlation between each candidate object and the seed object among multiple reference objects based on the aforementioned dense state attribute information.

[0093] The first-party device 301 is further configured to obtain an object number obtained from the second-party device 302 based on the aforementioned dense state correlation, and to obtain target information from the second-party device 302 based on the object number, wherein the target information is used to determine the target object.

[0094] In some implementations, the reference objects include a first object belonging to a specified group and a second object not belonging to the specified group. The first object includes a seed object and a non-seed object, and the second object is a non-seed object.

[0095] In other embodiments, the second-party device 302 may calculate the dense state correlation between each candidate object and the seed object among the multiple reference objects based on the aforementioned dense state attribute information in the following manner: calculate the correlation between each candidate object and each reference object among the multiple reference objects; and for each candidate object, calculate the dense state correlation between the candidate object and the seed object among the multiple reference objects based on the correlation between the candidate object and each reference object and the dense state attribute information.

[0096] In other embodiments, the second-party device 302 can calculate the close-state correlation between the candidate object and seed objects among multiple reference objects based on the correlation and close-state attribute information between the candidate object and each reference object: The weighted sum of the correlations between the candidate object and each reference object is calculated as the close-state correlation. The weight of the correlation between the candidate object and any reference object is the close-state attribute information corresponding to that reference object. For any reference object, if the reference object is a seed object, the attribute information corresponding to that reference object is a non-zero value; if the reference object is a non-seed object, the attribute information corresponding to that reference object is a value of 0.

[0097] In other embodiments, the second-party device 302 is further configured to acquire a target graph, which includes multiple nodes representing objects and connecting edges between related nodes. Objects corresponding to at least a subset of the nodes are selected as candidate objects.

[0098] In other embodiments, the first-party device 301 may send the aforementioned encrypted attribute information to the second-party device 302 in the following manner: Each object identifier corresponding to each reference object is sent to the second-party device 302 in association with each encrypted attribute information. The relevance between each candidate object and each of the multiple reference objects can be calculated as follows: based on the object identifiers corresponding to each reference object, a preset node importance algorithm is used to calculate the relevance between each candidate object and each of the multiple reference objects according to the target graph.

[0099] In other embodiments, the first-party device 301 may obtain the object number based on the dense state correlation from the second-party device 302 in the following manner: through a secure computation between the two parties, the first-party device 301 obtains the object number corresponding to the candidate object whose non-dense state correlation with the seed object among a plurality of reference objects is greater than a preset threshold from the second-party device 302.

[0100] In other embodiments, the first-party device 301 may obtain target information from the second-party device 302 based on the object number by means of a privacy information retrieval method, based on the object number, querying the target information from the second-party device 302, wherein the target information includes the encrypted data of the target object corresponding to the object number.

[0101] like Figure 4 As shown, Figure 4 This specification illustrates a device block diagram for determining an object based on privacy protection, according to an exemplary embodiment, applied to a first-party device included in a first system, which also includes a second-party device. The device may include: an acquisition module 401, an encryption module 402, and a determination module 403.

[0102] The acquisition module 401 is used to acquire the attribute information corresponding to each of the multiple reference objects. The multiple reference objects include seed objects and non-seed objects. The attribute information corresponding to any reference object is used to indicate whether the reference object belongs to the seed object.

[0103] The encryption module 402 is used to perform homomorphic encryption on each attribute information to obtain each encrypted attribute information corresponding to each of the multiple reference objects, and send the encrypted attribute information to the second-party device so that the second-party device can calculate the encrypted correlation between each candidate object among the multiple candidate objects and the seed object among the multiple reference objects based on the encrypted attribute information.

[0104] The determination module 403 is used to obtain the object number based on the dense state correlation from the second-party device, and to obtain the target information from the second-party device based on the object number. The target information is used to determine the target object.

[0105] like Figure 5 As shown, Figure 5 This specification illustrates a block diagram of an apparatus for determining an object based on privacy protection, according to an exemplary embodiment, applied to a second-party device included in a first system, the first system also including a first-party device. The apparatus may include: an acquisition module 501, a calculation module 502, and a providing module 503.

[0106] The acquisition module 501 is used to acquire the encrypted attribute information corresponding to each of the multiple reference objects from the first-party device. The encrypted attribute information is obtained by the first-party device through homomorphic encryption of the attribute information corresponding to each of the multiple reference objects. The multiple reference objects include seed objects and non-seed objects, and the attribute information corresponding to any reference object is used to indicate whether the reference object belongs to the seed object.

[0107] The calculation module 502 is used to calculate the dense state correlation between each candidate object among multiple candidate objects and the seed object among multiple reference objects based on the dense state attribute information, so that the first-party device can obtain the object number based on the dense state correlation.

[0108] Module 503 is provided to the first-party device to provide target information for identifying the target object based on the object number.

[0109] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of one or more embodiments of this specification, depending on actual needs. Those skilled in the art can understand and implement this without creative effort.

[0110] This specification also provides one or more embodiments of a computer-readable storage medium storing a computer program that can be used to perform the above-described actions. Figure 2 Any embodiment provides a method for determining objects based on privacy protection.

[0111] One or more embodiments of this specification also provide a computing device, including a memory and a processor, wherein the memory stores executable code, and when the processor executes the executable code, it implements the above-described... Figure 2 Any embodiment provides a method for determining objects based on privacy protection.

[0112] In the 1990s, improvements to a technology could be clearly distinguished as either hardware improvements (e.g., improvements to the circuit structure of diodes, transistors, switches, etc.) or software improvements (improvements to the methodology). However, with technological advancements, many methodological improvements today can be considered direct improvements to the hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved methodology into the hardware circuit. Therefore, it cannot be said that a methodological improvement cannot be implemented using hardware physical modules. For example, a Programmable Logic Device (PLD) (such as a Field Programmable Gate Array (FPGA)) is such an integrated circuit whose logic function is determined by the user programming the device. Designers can program and "integrate" a digital system onto a PLD themselves, without needing chip manufacturers to design and manufacture dedicated integrated circuit chips. Furthermore, nowadays, instead of manually manufacturing integrated circuit chips, this programming is mostly implemented using "logic compiler" software. Similar to the software compiler used in program development, the original code before compilation must be written in a specific programming language, called a Hardware Description Language (HDL). There are many HDLs, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, and RHDL (Ruby Hardware Description Language). Currently, the most commonly used are VHDL (Very-High-Speed ​​Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should understand that by simply performing some logic programming on the method flow using one of these hardware description languages ​​and programming it into an integrated circuit, the hardware circuit implementing the logical method flow can be easily obtained.

[0113] The controller can be implemented in any suitable manner. For example, it can take the form of a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers. Examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicon Labs C8051F320. A memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art will also recognize that, in addition to implementing the controller in purely computer-readable program code form, the same functionality can be achieved by logically programming the method steps to make the controller take the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers. Therefore, such a controller can be considered a hardware component, and the means included therein for implementing various functions can also be considered as structures within the hardware component. Alternatively, the means for implementing various functions can be considered as both software modules implementing the method and structures within the hardware component.

[0114] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or physical entities, or by products with certain functions. A typical implementation device is a server system. Of course, this application does not exclude the possibility that, with the future development of computer technology, the computer implementing the functions of the above embodiments can be, for example, a personal computer, a laptop computer, an in-vehicle human-machine interaction device, a cellular phone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or any combination of these devices.

[0115] While one or more embodiments of this specification provide the operational steps of the methods described in the embodiments or flowcharts, more or fewer operational steps may be included based on conventional or non-inventive means. The order of steps listed in the embodiments is merely one possible order of execution among many steps and does not represent the only possible order. In actual device or end product execution, the methods shown in the embodiments or drawings may be executed sequentially or in parallel (e.g., in a parallel processor or multi-threaded processing environment, or even a distributed data processing environment). The terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, product, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, product, or apparatus. Without further limitations, the presence of other identical or equivalent elements in the process, method, product, or apparatus that includes the elements is not excluded. For example, the use of terms such as "first," "second," etc., is to denote names and does not indicate any particular order.

[0116] For ease of description, the above devices are described in terms of function, divided into various modules. Of course, when implementing one or more of these specifications, the functions of each module can be implemented in one or more software and / or hardware components, or a module that performs the same function can be implemented by a combination of multiple sub-modules or sub-units. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, indirect coupling or communication connection between devices or units, and may be electrical, mechanical, or other forms.

[0117] 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... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0118] 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.

[0119] 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.

[0120] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0121] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0122] Computer-readable media, including both permanent and non-permanent, removable and non-removable media, can store information using any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage, graphene storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0123] Those skilled in the art will understand that one or more embodiments of this specification can be provided as a method, system, or computer program product. Therefore, one or more embodiments of this specification may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, one or more embodiments of this specification may take the form of a computer program product implemented 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.

[0124] One or more embodiments of this specification can be described in the general context of computer-executable instructions, such as program modules, that are executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a particular task or implement a particular abstract data type. One or more embodiments of this specification can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In a distributed computing environment, program modules can reside in local and remote computer storage media, including storage devices.

[0125] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, system embodiments are basically similar to method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments. In the description of this specification, the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this specification. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described can be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0126] The above description is merely an embodiment of one or more embodiments of this specification and is not intended to limit the scope of this specification. Various modifications and variations can be made to the one or more embodiments of this specification by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this specification should be included within the scope of the claims.

Claims

1. A method for determining objects based on privacy protection, applied to a first system; The first system includes a first-party device and a second-party device; the method includes: The first-party device acquires attribute information corresponding to each of multiple reference objects; the multiple reference objects include seed objects and non-seed objects; the attribute information corresponding to any reference object is used to indicate whether the reference object belongs to a seed object; The first-party device performs homomorphic encryption on each attribute information to obtain each encrypted attribute information corresponding to the plurality of reference objects, and sends the encrypted attribute information to the second-party device. Based on the dense state attribute information, the second-party device calculates the dense state correlation between each candidate object among the multiple candidate objects and the seed object among the multiple reference objects; The first-party device obtains an object number based on the dense state correlation from the second-party device, and obtains target information from the second-party device based on the object number; the target information is used to determine the target object. The step of calculating the dense-state correlation between each candidate object and the seed object among the multiple reference objects based on the dense-state attribute information includes: Calculate the correlation between each candidate object and each of the plurality of reference objects; For each candidate object, a weighted sum of the correlation between the candidate object and each of the plurality of reference objects is calculated to obtain the dense state correlation between the candidate object and the seed object among the plurality of reference objects; wherein, the weight of the correlation between the candidate object and any reference object is the dense state attribute information corresponding to the reference object.

2. The method according to claim 1, wherein, The reference objects include a first object belonging to a specified group and a second object not belonging to the specified group; the first object includes seed objects and non-seed objects; the second object is a non-seed object.

3. According to the method described in claim 1, for any reference object, if the reference object belongs to a seed object, the attribute information corresponding to the reference object is a non-zero value; if the reference object belongs to a non-seed object, the attribute information corresponding to the reference object is a value of 0.

4. The method according to claim 1, wherein, Also includes: The second-party device acquires a target image; the target image includes multiple nodes representing objects and connecting edges between related nodes among the multiple nodes; The second-party device selects objects corresponding to at least some of the nodes from the plurality of nodes as the candidate objects.

5. The method according to claim 4, wherein, Sending the encrypted attribute information to the second-party device includes associating each object identifier corresponding to the plurality of reference objects with the encrypted attribute information and sending it to the second-party device. The relevance between each candidate object and each of the multiple reference objects is calculated as follows: based on the object identifiers corresponding to each of the reference objects, a preset node importance algorithm is used to calculate the relevance between each candidate object and each of the multiple reference objects according to the target graph.

6. The method according to claim 1, wherein, The first-party device obtains the object number based on the dense state correlation from the second-party device, including: Through secure computation by both parties, the first device obtains from the second device the object number corresponding to the candidate object whose non-closed-state correlation with the seed object among the plurality of reference objects is greater than a preset threshold.

7. The method according to claim 1, wherein, The step of obtaining target information from the second-party device based on the object number includes: using a privacy information retrieval method, querying the target information from the second-party device based on the object number; the target information includes the encrypted data of the target object corresponding to the object number.

8. A method for determining an object based on privacy protection, applied to a first system including a first-party device; the first system further including a second-party device; the method comprising: Obtain the attribute information corresponding to each of the multiple reference objects; The multiple reference objects include seed objects and non-seed objects; The attribute information corresponding to any reference object is used to indicate whether the reference object belongs to the seed object; The attribute information is homomorphically encrypted to obtain the encrypted attribute information corresponding to each of the multiple reference objects, and the encrypted attribute information is sent to the second-party device; so that the second-party device can calculate the encrypted correlation between each candidate object among the multiple candidate objects and the seed object among the multiple reference objects based on the encrypted attribute information; Obtain the object number based on the dense state correlation from the second-party device, and obtain the target information from the second-party device based on the object number; The target information is used to identify the target object; The step of calculating the dense-state correlation between each candidate object and the seed object among the multiple reference objects based on the dense-state attribute information includes: Calculate the correlation between each candidate object and each of the plurality of reference objects; For each candidate object, a weighted sum of the correlation between the candidate object and each of the plurality of reference objects is calculated to obtain the dense state correlation between the candidate object and the seed object among the plurality of reference objects; wherein, the weight of the correlation between the candidate object and any reference object is the dense state attribute information corresponding to the reference object.

9. A method for determining an object based on privacy protection, applied to a second-party device included in a first system; the first system further includes a first-party device; the method includes: The first party device obtains encrypted attribute information corresponding to each of the multiple reference objects; wherein, the encrypted attribute information is obtained by the first party device through homomorphic encryption of the attribute information corresponding to each of the multiple reference objects; the multiple reference objects include seed objects and non-seed objects; the attribute information corresponding to any reference object is used to indicate whether the reference object belongs to the seed object; Based on the dense state attribute information, the dense state correlation between each candidate object among the multiple candidate objects and the seed object among the multiple reference objects is calculated, so that the first device can obtain the object number based on the dense state correlation. Provide the first-party device with target information for determining the target object based on the object number; The step of calculating the dense-state correlation between each candidate object and the seed object among the multiple reference objects based on the dense-state attribute information includes: Calculate the correlation between each candidate object and each of the plurality of reference objects; For each candidate object, a weighted sum of the correlation between the candidate object and each of the plurality of reference objects is calculated to obtain the dense state correlation between the candidate object and the seed object among the plurality of reference objects; wherein, the weight of the correlation between the candidate object and any reference object is the dense state attribute information corresponding to the reference object.

10. A system for determining objects based on privacy protection; the system includes a first-party device and a second-party device; The first-party device is configured to acquire attribute information corresponding to each of multiple reference objects, perform homomorphic encryption on the attribute information to obtain encrypted attribute information corresponding to each of the multiple reference objects, and send the encrypted attribute information to the second-party device; wherein... The plurality of reference objects include seed objects and non-seed objects; the attribute information corresponding to any reference object is used to indicate whether the reference object belongs to a seed object; The second-party device is used to calculate the dense state correlation between each candidate object among the plurality of candidate objects and the seed object among the plurality of reference objects based on the dense state attribute information; The first-party device is further configured to obtain an object number based on the dense state correlation from the second-party device, and obtain target information from the second-party device based on the object number; the target information is used to determine the target object; The step of calculating the dense-state correlation between each candidate object and the seed object among the multiple reference objects based on the dense-state attribute information includes: Calculate the correlation between each candidate object and each of the plurality of reference objects; For each candidate object, a weighted sum of the correlation between the candidate object and each of the plurality of reference objects is calculated to obtain the dense state correlation between the candidate object and the seed object among the plurality of reference objects; wherein, the weight of the correlation between the candidate object and any reference object is the dense state attribute information corresponding to the reference object.

11. An apparatus for determining an object based on privacy protection, applied to a first-party device included in a first system; the first system further includes a second-party device; the apparatus comprises: The acquisition module is used to obtain the attribute information corresponding to each of multiple reference objects. The multiple reference objects include seed objects and non-seed objects; The attribute information corresponding to any reference object is used to indicate whether the reference object belongs to the seed object; An encryption module is used to perform homomorphic encryption on each attribute information to obtain each encrypted attribute information corresponding to each of the plurality of reference objects, and send the encrypted attribute information to the second-party device; so that the second-party device can calculate the encrypted correlation between each candidate object among the plurality of candidate objects and the seed object among the plurality of reference objects based on the encrypted attribute information; The determination module is used to obtain the object number obtained from the second-party device based on the dense state correlation, and to obtain target information from the second-party device based on the object number; The target information is used to identify the target object; The step of calculating the dense-state correlation between each candidate object and the seed object among the multiple reference objects based on the dense-state attribute information includes: Calculate the correlation between each candidate object and each of the plurality of reference objects; For each candidate object, a weighted sum of the correlation between the candidate object and each of the plurality of reference objects is calculated to obtain the dense state correlation between the candidate object and the seed object among the plurality of reference objects; wherein, the weight of the correlation between the candidate object and any reference object is the dense state attribute information corresponding to the reference object.

12. An apparatus for determining an object based on privacy protection, applied to a second-party device included in a first system; the first system further includes a first-party device; the apparatus includes: The acquisition module is used to acquire each encrypted attribute information corresponding to a plurality of reference objects from the first party device; wherein, each encrypted attribute information is obtained by the first party device through homomorphic encryption of each attribute information corresponding to the plurality of reference objects; the plurality of reference objects include seed objects and non-seed objects; the attribute information corresponding to any reference object is used to indicate whether the reference object belongs to a seed object; The calculation module is used to calculate the dense state correlation between each candidate object among the multiple candidate objects and the seed object among the multiple reference objects based on the dense state attribute information, so that the first-party device can obtain the object number based on the dense state correlation. A module is provided to the first-party device to provide target information for determining the target object based on the object number; The step of calculating the dense-state correlation between each candidate object and the seed object among the multiple reference objects based on the dense-state attribute information includes: Calculate the correlation between each candidate object and each of the plurality of reference objects; For each candidate object, a weighted sum of the correlation between the candidate object and each of the plurality of reference objects is calculated to obtain the dense state correlation between the candidate object and the seed object among the plurality of reference objects; wherein, the weight of the correlation between the candidate object and any reference object is the dense state attribute information corresponding to the reference object.

13. A computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to perform the method of any one of claims 1-9.

14. A computing device comprising a memory and a processor, wherein the memory stores executable code, and the processor, when executing the executable code, implements the method of any one of claims 1-9.

Images