Data ownership judgment-oriented feature searching method, device and equipment and medium

Abstract

The invention provides a feature searching method, device and equipment oriented to data ownership judgment and a medium, and relates to the technical field of data ownership judgment, and the method comprises the steps: obtaining multi-dimensional data features corresponding to target data oriented to data ownership judgment; according to the multi-dimensional data features corresponding to the target data and the binary search tree, feature similarity search is carried out, and whether suspicious data are found from the binary search tree or not is determined; wherein the binary search tree is constructed in advance based on feature similarity calculated by an ownership judgment database, and the ownership judgment database is a database used for data ownership judgment; the similarity between the suspicious data and the target data meets a similarity threshold value. According to the technical scheme, the binary search tree is constructed based on the feature similarity to perform suspicious data search on the target data, an ownership judgment database does not need to be traversed, the overhead of the search process is reduced, and the query complexity is reduced.

Description

Technical Field

[0001] This invention relates to the field of data ownership determination technology, and in particular to a feature search method, apparatus, device and medium for data ownership determination. Background Technology

[0002] Determining data ownership is the foundation of data transactions. By empowering data processors, they gain corresponding legal control over their data, thereby excluding others from infringing upon the data to a certain extent or within a certain scope.

[0003] In the process of determining data ownership, the common practice is to extract features from the data and then search for similar data that has already been determined in a data ownership database to determine whether there is any infringement. Existing technologies typically involve a traversal search, meaning that the data must be compared with every feature in the data ownership database to determine if there is any infringement. This method requires traversing the entire database, resulting in excessive retrieval overhead and increased complexity. Summary of the Invention

[0004] This invention provides a feature-based search method, apparatus, device, and medium for determining data ownership, which addresses the shortcomings of existing technologies that require traversing the entire database, resulting in excessive retrieval overhead and increased complexity. It enables efficient searching of suspicious data in target data for determining data ownership based on a binary search tree constructed from feature similarity, without traversing the entire ownership determination database, thus reducing the overhead during the search process and lowering query complexity.

[0005] This invention provides a feature search method for determining data ownership, comprising the following steps.

[0006] Obtain multi-dimensional data features corresponding to target data for determining data ownership.

[0007] Based on the multidimensional data features corresponding to the target data and a binary search tree, feature similarity is searched to determine whether suspicious data is found in the binary search tree. The binary search tree is constructed in advance based on feature similarity calculated from an ownership determination database, which is a database used to determine data ownership. Suspicious data refers to data whose similarity to the target data meets a similarity threshold.

[0008] According to the feature search method for data ownership determination provided by the present invention, the steps for determining the binary search tree are as follows: obtaining an ownership determination database and a unit feature vector; wherein, the ownership determination database includes at least one pre-acquired data feature, and the unit feature vector is a pre-defined feature vector; determining all feature similarities based on all data features and the unit feature vector in the ownership determination database; wherein, all feature similarities include the similarity between each data feature and the unit feature vector; and generating a binary search tree based on all feature similarities.

[0009] According to the present invention, a feature search method for determining data ownership includes performing feature similarity search based on multidimensional data features corresponding to target data and a binary search tree to determine whether suspicious data is found in the binary search tree. The method comprises: determining the similarity of ownership-to-be-determined features based on the multidimensional data features and unit feature vectors corresponding to the target data; performing feature similarity search based on the similarity of ownership-to-be-determined features and the binary search tree to determine whether the similarity of ownership-to-be-determined features meets preset conditions; if the similarity of ownership-to-be-determined features meets the preset conditions, determining that suspicious data is found in the binary search tree; and if the similarity of ownership-to-be-determined features does not meet the preset conditions, determining that no suspicious data is found in the binary search tree.

[0010] According to a feature search method for determining data ownership provided by the present invention, after performing feature similarity search based on the multidimensional data features corresponding to the target data and a binary search tree to determine whether suspicious data is found in the binary search tree, the method further includes: when the suspicious data is found in the binary search tree, performing feature comparison based on the data features corresponding to the suspicious data and the multidimensional data features corresponding to the target data to determine the ownership determination result of the target data.

[0011] According to a feature search method for data ownership determination provided by the present invention, when suspicious data is found in a binary search tree, feature comparison is performed based on the data features corresponding to the suspicious data and the multidimensional data features corresponding to the target data. The method includes: when suspicious data exists in the target data, threshold data filtering is performed on the suspicious data to obtain candidate data; wherein, the threshold data filtering is performed by directly selecting a set selection threshold and selecting the selected threshold number of suspicious data as the candidate data; and feature comparison is performed based on the data features corresponding to the candidate data and the multidimensional data features corresponding to the target data.

[0012] According to the present invention, a feature search method for determining data ownership includes comparing the data features corresponding to suspicious data and the multidimensional data features corresponding to target data to determine the ownership determination result of the target data. The method comprises: if the feature comparison between the data features corresponding to suspicious data and the multidimensional data features corresponding to target data is successful, determining the ownership determination result of the target data as ownership determination failure; and if the feature comparison between the data features corresponding to suspicious data and the multidimensional data features corresponding to target data fails, determining the ownership determination result of the target data as ownership determination success.

[0013] The feature search method for determining data ownership according to the present invention further includes: determining that the ownership of the target data is successfully determined when no suspicious data is found in the binary search tree.

[0014] The present invention also provides a feature search device for determining data ownership, comprising the following modules.

[0015] The feature acquisition module is used to acquire multi-dimensional data features corresponding to target data for data ownership determination.

[0016] The feature search module is used to perform feature similarity search based on the multidimensional data features corresponding to the target data and a binary search tree, and to determine whether suspicious data is found in the binary search tree; wherein, the binary search tree is constructed in advance based on the feature similarity calculated by the ownership determination database, and the ownership determination database is a database used to determine the ownership of data; the suspicious data refers to data whose similarity with the target data meets the similarity threshold.

[0017] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements any of the above-described feature search methods for determining data ownership.

[0018] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the above-described feature search methods for determining data ownership.

[0019] The present invention also provides a computer program product, including a computer program, which, when executed by a processor, implements any of the above-described feature search methods for determining data ownership.

[0020] This invention provides a feature search method, apparatus, device, and medium for data ownership determination. It acquires multi-dimensional data features corresponding to target data for data ownership determination; performs feature similarity search based on the multi-dimensional data features and a binary search tree to determine whether suspicious data is found in the binary search tree; wherein the binary search tree is pre-constructed based on feature similarity calculated from an ownership determination database, which is used for data ownership determination; the suspicious data refers to data whose similarity to the target data meets a similarity threshold. This invention addresses the shortcomings of existing technologies that require traversing the entire database, leading to excessive retrieval overhead and increased complexity. It enables the search for suspicious data in target data for data ownership determination based on a binary search tree constructed from feature similarity, eliminating the need to traverse the entire ownership determination database, thus reducing search overhead and query complexity. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0022] Figure 1 This is a flowchart illustrating the feature search method for determining data ownership provided by the present invention.

[0023] Figure 2 This is a schematic diagram of the feature search device for determining data ownership provided by the present invention.

[0024] Figure 3 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0026] The following is combined with Figure 1The present invention describes a feature search method for determining data ownership. This method is applicable to situations requiring rapid feature search for determining data ownership. The execution subject of this method can be an electronic device or a feature search device for determining data ownership installed in the electronic device. This feature search device for determining data ownership can be implemented through software, hardware, or a combination of both. Figure 1 This is a flowchart illustrating the feature lookup method for data ownership determination provided by the present invention, as shown below. Figure 1 As shown, the method includes the following steps 101 and 102.

[0027] Step 101: Obtain the multi-dimensional data features corresponding to the target data for data ownership determination.

[0028] In this step, the target data is the data for which data ownership determination is required, and the multidimensional data features are the data features obtained by extracting multidimensional features from the target data for data ownership determination. Multidimensional features can be one-dimensional, two-dimensional, three-dimensional, four-dimensional, and five-dimensional, etc., and this embodiment does not limit them.

[0029] Specifically, the process involves acquiring the target data for which data ownership determination is required, then extracting multi-dimensional data features from the target data to obtain the corresponding multi-dimensional data features.

[0030] Step 102: Perform feature similarity search based on the multidimensional data features corresponding to the target data and the binary search tree to determine whether suspicious data is found in the binary search tree.

[0031] In this step, the binary search tree is pre-constructed based on feature similarity calculated from the ownership determination database. The binary search tree transforms a linear search into a logarithmic search. The suspicious data refers to data whose similarity to the target data meets a similarity threshold.

[0032] The ownership determination database is used for determining data ownership, and it is a very large database. When suspicious data exists, there may be multiple suspicious data entries.

[0033] Specifically, after obtaining the multidimensional data features corresponding to the target data, the multidimensional data features corresponding to the target data are input into a binary search tree. Suspicious data is filtered through the similarity threshold set in the binary search tree to determine whether suspicious data is found in the binary search tree.

[0034] In one specific implementation, the steps for determining a binary search tree are as follows: obtaining an ownership determination database and a unit feature vector; wherein, the ownership determination database includes at least one pre-acquired data feature, and the unit feature vector is a pre-defined feature vector; determining the similarity of all features based on all data features and the unit feature vector in the ownership determination database; and generating a binary search tree based on the similarity of all features.

[0035] In this step, all the feature similarities include the similarity between each data feature and the unit feature vector. The unit feature vector is a pre-defined baseline feature vector. For example, if the Euclidean distance between the data feature of data 1 and the unit feature vector is D1, and the Euclidean distance between the data feature of similar data 2 and the unit feature vector is D2, then there is a high probability that D1 is approximately equal to D2. Therefore, the ownership determination features existing in the ownership determination database can be sorted according to their distance from the unit vector. Once the sorting is complete, a logarithmic search can be performed using a binary search tree; this embodiment does not limit this process.

[0036] In one specific implementation, generating a binary search tree based on all feature similarities includes: sorting all feature similarities to obtain a sorted target feature similarity set; and generating a binary search tree based on the distance between all feature similarities in the target feature similarity set.

[0037] In this step, there is at least one binary search tree, such as one, two, three, four, five, or six trees. This embodiment does not limit this.

[0038] The first binary search tree can be, for example, a binary search tree established by the Euclidean distance between the data features of the ownership determination database and the unit feature vector; the second binary search tree can be, for example, a binary search tree established by the cosine similarity between the data features of the ownership determination database and the unit feature vector; the third binary search tree can be, for example, a binary search tree established by the Chernshev distance between the data features of the ownership determination database and the unit feature vector; the fourth binary search tree can be, for example, a binary search tree established by the Manhattan distance between the data features of the ownership determination database and the unit feature vector; the fifth binary search tree can be, for example, a binary search tree established by the Jaccard distance between the data features of the ownership determination database and the unit feature vector; the sixth binary search tree can be, for example, a binary search tree established by the Minkowski distance between the data features of the ownership determination database and the unit feature vector. This embodiment does not limit the specific binary search tree to this type.

[0039] The Euclidean distance is calculated as shown in formula (1).

[0040] (1) In formula (1), Representing data characteristics and unit eigenvectors European distance, Representing data characteristics, Represents a unit eigenvector. Indicates the index of the data feature and the unit eigenvector. This represents the total number of data features and unit eigenvectors.

[0041] The cosine similarity is calculated as shown in formula (2).

[0042] (2) In formula (2), Representing data characteristics and unit eigenvectors The cosine similarity.

[0043] The Chebyshev distance is calculated as shown in formula (3).

[0044] (3) In formula (3), Representing data characteristics and unit eigenvectors Chebyshev distance.

[0045] The Manhattan distance is calculated as shown in formula (4).

[0046] (4) In formula (4), Representing data characteristics and unit eigenvectors Manhattan distance.

[0047] The Jaccard distance is calculated as shown in formula (5).

[0048] (5) In formula (5), Representing data characteristics and unit eigenvectors The distance of Jaccard.

[0049] The Minkowski distance is calculated as shown in formula (6).

[0050] (6) In formula (6), Representing data characteristics and unit eigenvectors Minkowski distance.

[0051] Specifically, the ownership determination database and unit feature vectors are obtained. The distances between all data features in the ownership determination database and the unit vectors are calculated respectively, and the distances are sorted from left to right. After sorting, the target feature similarity set is obtained. Then, a binary search tree is generated by using a balanced binary search tree to generate all the sorted distances in the target feature similarity set.

[0052] For example, taking Euclidean distance as an example, first, calculate the Euclidean distance between all data features and unit feature vectors in the ownership determination database, and then sort them from left to right. After sorting, generate a binary search tree from all sorted distances using a balanced binary search tree approach. The method for constructing binary search trees for other distances is the same.

[0053] In one specific implementation, feature similarity search is performed based on the multidimensional data features corresponding to the target data and a binary search tree to determine whether suspicious data is found in the binary search tree. This includes: determining the similarity of the ownership determination features based on the multidimensional data features and unit feature vectors corresponding to the target data; performing feature similarity search based on the ownership determination features and the binary search tree to determine whether the ownership determination features similarity meets a preset condition; if the ownership determination features similarity meets the preset condition, determining that suspicious data is found in the binary search tree; if the ownership determination features similarity does not meet the preset condition, determining that no suspicious data is found in the binary search tree.

[0054] In this step, the calculation method for the similarity of the features to be determined includes at least one of the following: Euclidean distance, cosine similarity, Chebyshev distance, Manhattan distance, Jaccard distance, and Minkowski distance. This embodiment does not limit the calculation method.

[0055] Similarity of features for determining ownership is a metric used to measure the degree of similarity between data. It is used to assess the proximity of data in a specific dimension or feature space. The higher the similarity, the closer or more similar the two data are.

[0056] The preset condition is a pre-defined feature similarity search condition. For example, the preset condition may be that the absolute value of the difference between the feature similarity value of the binary tree node and the feature similarity value to be determined is less than or equal to the similarity threshold. Different calculation distances may correspond to different similarity thresholds, and this embodiment does not limit this.

[0057] Specifically, the similarity of the features to be determined for ownership determination is determined based on the multidimensional data features and unit feature vectors corresponding to the target data. Then, this similarity is input into a binary search tree for rapid feature similarity searching. Starting from the root node of the binary search tree, if the feature similarity value of a node is greater than the similarity of the features to be determined for ownership determination, the search continues to the left subtree from the current node. If the feature similarity value of a node is less than or equal to the similarity of the features to be determined for ownership determination, the search continues to the right subtree from the current node. If the absolute value of the difference between the feature similarity value of the current node and the similarity of the features to be determined for ownership determination is less than or equal to a similarity threshold, the search stops, and it is determined that the similarity of the features to be determined for ownership determination meets a preset condition. The entire subtree rooted at the current node is then considered as the filtered suspicious data. If, after searching the binary search tree, the similarity of the features to be determined for ownership determination does not meet the preset condition, it is determined that no suspicious data was found in the binary search tree.

[0058] The advantage of this setup is that by using multiple distances and efficiently searching for features, the visibility between features can be evaluated from different angles, improving the accuracy of finding suspicious data. Furthermore, in the subsequent infringement judgment process, the similarity between the target data and the suspicious data can be comprehensively measured, enhancing the robustness of infringement identification. This helps reduce the impact of biases in the data, making the query more objective and accurate, thereby improving the accuracy of data ownership determination and reducing the possibility of misjudgment in infringement judgment.

[0059] In one specific implementation, it further includes: when no suspicious data is found in the binary search tree, determining that the ownership of the target data has been successfully determined.

[0060] Specifically, if no suspicious data is found in the binary search tree, the ownership determination of the target data is successful.

[0061] In one specific implementation, after performing feature similarity search based on the multidimensional data features corresponding to the target data and the binary search tree to determine whether suspicious data is found in the binary search tree, the method further includes: when suspicious data is found in the binary search tree, performing feature comparison based on the data features corresponding to the suspicious data and the multidimensional data features corresponding to the target data to determine the ownership determination result of the target data.

[0062] Specifically, after obtaining suspicious data, the data features corresponding to the suspicious data are compared with the multidimensional data features corresponding to the target data to determine the ownership of the target data.

[0063] The advantage of this setup is that it eliminates the need to compare the multidimensional data features corresponding to the target data with all data features in the ownership determination database, thus reducing computational complexity.

[0064] In one specific implementation, when suspicious data is found in the binary search tree, feature comparison is performed based on the data features corresponding to the suspicious data and the multidimensional data features corresponding to the target data. This includes: when suspicious data exists in the target data, threshold data filtering is performed on the suspicious data to obtain candidate data; wherein, the threshold data filtering is performed by directly selecting the suspicious data as the candidate data through a set selection threshold; and feature comparison is performed based on the data features corresponding to the candidate data and the multidimensional data features corresponding to the target data.

[0065] In this step, the data filtering mode for suspicious data may include a threshold data filtering mode based on a filtering threshold, or a voting data filtering mode based on voting. This embodiment does not limit this to either.

[0066] Among them, the threshold data filtering mode based on the filtering threshold takes the union of the top K data with different similarity as the final candidate data. K is a hyperparameter that needs to be set experimentally. This embodiment does not limit it.

[0067] The voting data filtering mode based on voting is to select data with more than or equal to T metrics that are considered to have the potential for infringement as the final candidate data. T is a hyperparameter that needs to be set experimentally, and this embodiment does not limit it.

[0068] Specifically, when suspicious data is found in the binary search tree, the suspicious data is filtered through different data filtering modes to obtain candidate data; feature comparison is then performed based on the data features corresponding to the candidate data and the multidimensional data features corresponding to the target data.

[0069] The advantage of this setup is that it further reduces computational complexity and improves the efficiency of ownership determination.

[0070] In one specific implementation, feature comparison is performed based on the data features corresponding to the suspicious data and the multidimensional data features corresponding to the target data to determine the ownership determination result of the target data. This includes: if the feature comparison between the data features corresponding to the suspicious data and the multidimensional data features corresponding to the target data is successful, the ownership determination result of the target data is determined to be ownership determination failure; if the feature comparison between the data features corresponding to the suspicious data and the multidimensional data features corresponding to the target data is unsuccessful, the ownership determination result of the target data is determined to be ownership determination success.

[0071] Specifically, the multidimensional data features corresponding to the target data are compared one by one with the data features corresponding to the suspicious data. If the data features corresponding to the suspicious data are successfully compared with the multidimensional data features corresponding to the target data, the ownership determination result of the target data is determined to be ownership determination failure; if the data features corresponding to the suspicious data are not successfully compared with the multidimensional data features corresponding to the target data, the ownership determination result of the target data is determined to be ownership determination success.

[0072] The advantage of this setup is that it avoids comparing each piece of data with all the data in the ownership determination database. Instead, it only compares the data with suspicious data and determines the ownership of the target data, thus reducing the complexity of the search and improving search efficiency.

[0073] This invention provides a feature search method for determining data ownership. It acquires multi-dimensional data features corresponding to target data for ownership determination; performs feature similarity search based on these multi-dimensional data features and a binary search tree; and determines whether suspicious data is found within the binary search tree. The binary search tree is pre-constructed based on feature similarity calculated from an ownership determination database, which is used for data ownership determination. Suspicious data refers to data whose similarity to the target data meets a similarity threshold. This invention addresses the shortcomings of existing technologies that require traversing the entire database, leading to excessive retrieval overhead and increased complexity. It enables the search of target data for ownership determination based on a binary search tree constructed from feature similarity, thereby determining the ownership result of the target data without traversing the entire ownership determination database, reducing overhead in the ownership determination process and lowering query complexity.

[0074] The following describes the feature search device for determining data ownership provided by the present invention. The feature search device for determining data ownership described below can be referred to in correspondence with the feature search method for determining data ownership described above.

[0075] Figure 2 This is a schematic diagram of the feature search device for data ownership determination provided by the present invention, with reference to... Figure 2 As shown, the feature search device 200 for determining data ownership includes a feature acquisition module 201 and a feature search module 202.

[0076] The feature acquisition module 201 is used to acquire multi-dimensional data features corresponding to the target data for data ownership determination.

[0077] The feature search module 202 is used to perform feature similarity search based on the multidimensional data features corresponding to the target data and a binary search tree, and to determine whether suspicious data is found in the binary search tree; wherein, the binary search tree is constructed in advance based on the feature similarity calculated by the ownership determination database, and the ownership determination database is a database used to determine the ownership of data; the suspicious data refers to data whose similarity with the target data meets the similarity threshold.

[0078] In one example embodiment, the apparatus further includes a search tree construction module. The search tree construction module is configured to: acquire an ownership determination database and a unit feature vector; wherein the ownership determination database includes at least one pre-acquired data feature, and the unit feature vector is a pre-defined feature vector; determine all feature similarities based on all data features and the unit feature vector in the ownership determination database; wherein all feature similarities include the similarity between each data feature and the unit feature vector; and generate a binary search tree based on all feature similarities.

[0079] In one example embodiment, the search tree construction module generates a binary search tree based on all feature similarities, specifically by: sorting all feature similarities to obtain a sorted set of target feature similarities; and generating a binary search tree based on the distance between all feature similarities in the set of target feature similarities.

[0080] In one example embodiment, the feature lookup module 202 is specifically used to: determine the similarity of the ownership determination features based on the multidimensional data features and unit feature vectors corresponding to the target data; perform feature similarity lookup based on the ownership determination feature similarity and a binary search tree to determine whether the ownership determination feature similarity meets a preset condition; if the ownership determination feature similarity meets the preset condition, determine that suspicious data has been found in the binary search tree; if the ownership determination feature similarity does not meet the preset condition, determine that no suspicious data has been found in the binary search tree.

[0081] In one example embodiment, the device further includes a result determination module. The result determination module is configured to: after performing feature similarity lookup based on the multidimensional data features corresponding to the target data and a binary search tree to determine whether suspicious data is found in the binary search tree, if the suspicious data is found in the binary search tree, perform feature comparison based on the data features corresponding to the suspicious data and the multidimensional data features corresponding to the target data to determine the ownership determination result of the target data.

[0082] In one example embodiment, the result determination module, when suspicious data is found in the binary search tree, performs feature comparison based on the data features corresponding to the suspicious data and the multidimensional data features corresponding to the target data. Specifically, when suspicious data exists in the target data, the suspicious data is subjected to threshold data filtering to obtain candidate data; wherein, the threshold data filtering is performed by directly selecting the suspicious data as the candidate data through a set selection threshold; and feature comparison is performed based on the data features corresponding to the candidate data and the multidimensional data features corresponding to the target data.

[0083] In one example embodiment, the result determination module performs feature comparison based on the data features corresponding to the suspicious data and the multidimensional data features corresponding to the target data to determine the ownership determination result of the target data. Specifically, it is used to: determine the ownership determination result of the target data as ownership determination failure when the feature comparison between the data features corresponding to the suspicious data and the multidimensional data features corresponding to the target data is successful; and determine the ownership determination result of the target data as ownership determination success when the feature comparison between the data features corresponding to the suspicious data and the multidimensional data features corresponding to the target data is unsuccessful.

[0084] In one example embodiment, the device further includes an ownership determination success module. The ownership determination success module is used to determine that the ownership of the target data has been successfully determined when no suspicious data is found in the binary search tree.

[0085] The apparatus of this embodiment can be used to execute the method of any embodiment in the feature search method for determining data ownership. Its specific implementation process and technical effects are similar to those in the feature search method for determining data ownership. For details, please refer to the detailed description in the feature search method for determining data ownership, which will not be repeated here.

[0086] Figure 3 This is a schematic diagram of the structure of the electronic device provided by the present invention, such as... Figure 3As shown, the electronic device may include a processor 310, a communications interface 320, a memory 330, and a communication bus 340. The processor 310, communications interface 320, and memory 330 communicate with each other via the communication bus 340. The processor 310 can call logical instructions in the memory 330 to execute a feature search method for data ownership determination. This method includes: acquiring multi-dimensional data features corresponding to target data for data ownership determination; performing feature similarity search based on the multi-dimensional data features corresponding to the target data and a binary search tree to determine whether suspicious data is found in the binary search tree; wherein the binary search tree is pre-constructed based on feature similarity calculated using an ownership determination database, and the ownership determination database is a database used for data ownership determination; the suspicious data refers to data whose similarity to the target data meets a similarity threshold.

[0087] Furthermore, the logical instructions in the aforementioned memory 330 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0088] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the feature search method for data ownership determination provided by the above methods. The method includes: acquiring multi-dimensional data features corresponding to target data for data ownership determination; performing feature similarity search based on the multi-dimensional data features corresponding to the target data and a binary search tree to determine whether suspicious data is found in the binary search tree; wherein, the binary search tree is constructed in advance based on feature similarity calculated based on an ownership determination database, and the ownership determination database is a database used for data ownership determination; the suspicious data refers to data whose similarity to the target data meets a similarity threshold.

[0089] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program implements a feature search method for data ownership determination provided by the above methods. The method includes: acquiring multi-dimensional data features corresponding to target data for data ownership determination; performing feature similarity search based on the multi-dimensional data features corresponding to the target data and a binary search tree to determine whether suspicious data is found in the binary search tree; wherein the binary search tree is constructed in advance based on feature similarity calculated from an ownership determination database, and the ownership determination database is a database used for data ownership determination; the suspicious data refers to data whose similarity to the target data meets a similarity threshold.

[0090] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. 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 this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0091] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0092] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A feature search method for determining data ownership, characterized in that, include: Obtain multi-dimensional data features corresponding to target data for data ownership determination; Based on the multidimensional data features corresponding to the target data and a binary search tree, feature similarity is searched to determine whether suspicious data is found in the binary search tree; wherein, the binary search tree is constructed in advance based on feature similarity calculated on an ownership determination database, and the ownership determination database is a database used for determining data ownership; the suspicious data refers to data whose similarity to the target data meets a similarity threshold.

2. The feature search method for data ownership determination according to claim 1, characterized in that, The steps for determining the binary search tree are as follows: Obtain the ownership determination database and the unit feature vector; wherein, the ownership determination database includes at least one pre-acquired data feature, and the unit feature vector is a pre-set feature vector; All feature similarities are determined based on all data features and the unit feature vector in the ownership determination database, respectively; wherein, all feature similarities include the similarity between each data feature and the unit feature vector; The binary search tree is generated based on the similarity of all the stated features.

3. The feature search method for data ownership determination according to claim 2, characterized in that, The step of performing feature similarity search based on the multidimensional data features corresponding to the target data and a binary search tree to determine whether suspicious data is found in the binary search tree includes: The similarity of the features to be determined for ownership is determined based on the multidimensional data features and the unit feature vector corresponding to the target data. Based on the similarity of the features to be determined and the binary search tree, a feature similarity search is performed to determine whether the similarity of the features to be determined meets the preset conditions. If the similarity of the features to be determined to be owned meets the preset conditions, the suspicious data is determined to be found in the binary search tree; If the similarity of the features to be determined to be of ownership does not meet the preset conditions, it is determined that the suspicious data was not found in the binary search tree.

4. The feature search method for data ownership determination according to claim 1, characterized in that, After determining whether suspicious data is found in the binary search tree by performing feature similarity search based on the multidimensional data features corresponding to the target data, the method further includes: When the suspicious data is found in the binary search tree, the ownership of the target data is determined by comparing the data features corresponding to the suspicious data and the multidimensional data features corresponding to the target data.

5. The feature search method for data ownership determination according to claim 4, characterized in that, When the suspicious data is found in the binary search tree, feature comparison is performed based on the data features corresponding to the suspicious data and the multidimensional data features corresponding to the target data, including: When the target data contains the suspicious data, the suspicious data is filtered using a threshold data method to obtain candidate data; wherein, the threshold data filtering is performed by directly selecting a threshold number of the suspicious data as the candidate data using a set selection threshold. Feature comparison is performed based on the data features corresponding to the candidate data and the multidimensional data features corresponding to the target data.

6. The feature search method for data ownership determination according to claim 4, characterized in that, The step of performing feature comparison based on the data features corresponding to the suspicious data and the multidimensional data features corresponding to the target data to determine the ownership determination result of the target data includes: If the data feature corresponding to the suspicious data is successfully matched with the feature of the multidimensional data feature corresponding to the target data, the ownership determination result of the target data is determined to be ownership determination failure. If the feature comparison between the data feature corresponding to the suspicious data and the multidimensional data feature corresponding to the target data fails, the ownership determination result of the target data is determined to be successful.

7. The feature lookup method for determining data ownership according to any one of claims 1-6, characterized in that, Also includes: If the suspicious data is not found in the binary search tree, the ownership determination of the target data is successful.

8. A feature lookup device for determining data ownership, characterized in that, include: The feature acquisition module is used to acquire multi-dimensional data features corresponding to target data for data ownership determination. The feature search module is used to perform feature similarity search based on the multidimensional data features corresponding to the target data and a binary search tree, and to determine whether suspicious data is found in the binary search tree; wherein, the binary search tree is constructed in advance based on feature similarity calculated based on an ownership determination database, and the ownership determination database is a database used for data ownership determination; the suspicious data refers to data whose similarity to the target data meets a similarity threshold.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the feature lookup method for determining data ownership as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the feature search method for determining data ownership as described in any one of claims 1 to 7.

Images