Characteristic extraction method and device oriented to ownership judgment, electronic equipment and medium

Abstract

The invention provides a feature extraction method and device for ownership judgment, electronic equipment and a medium, and relates to the technical field of ownership judgment, and the method comprises the steps: obtaining to-be-judged data for ownership judgment; performing multi-dimensional feature extraction on the to-be-judged data according to the feature extraction tool to obtain multi-dimensional data features of the to-be-judged data; wherein the feature extraction tool is used for performing feature extraction of five dimensions of data integrity, metadata, data flow, data semantics and high-order attributes on to-be-judged data; the multi-dimensional feature extraction comprises data integrity feature extraction, metadata feature extraction, data flow feature extraction, data semantic feature extraction and high-order attribute feature extraction. According to the technical scheme, the multi-dimensional data features are obtained by performing multi-dimensional feature extraction on the to-be-judged data, so that the accuracy of subsequent ownership judgment is realized, the similarity precision is improved, and the method can be used for data ownership judgment in data element circulation.

Description

Technical Field

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

[0002] Determining data ownership is fundamental to data trading, circulation, and sharing, and data feature extraction is arguably the most crucial component of ownership determination. Current data feature extraction primarily utilizes the International Standard Content Code (ISCC), mainly describing ISCC metadata and its integration with other schemes. ISCC is applicable to specific digital assets and is a data descriptor deterministically constructed from multiple hash digests using the algorithms and rules outlined in this document. However, ISCC feature extraction methods do not support similarity calculations based on high-dimensional data features, resulting in low accuracy in ownership determination calculations based on extracted features. Summary of the Invention

[0003] This invention provides a feature extraction method, apparatus, electronic device, and medium for ownership determination, which addresses the shortcomings of existing ISCC feature extraction methods that do not support the extraction of high-dimensional features from data, resulting in low accuracy in ownership determination based on extracted features. This invention achieves multi-dimensional data features by extracting multi-dimensional features from the data to be determined, thereby improving the accuracy of subsequent data display and similarity accuracy. It can be used for data ownership determination in the circulation of data elements.

[0004] This invention provides a feature extraction method for ownership determination, comprising the following steps.

[0005] Obtain the data to be determined for ownership determination.

[0006] The feature extraction tool performs multi-dimensional feature extraction on the data to be judged, resulting in multi-dimensional data features. This feature extraction tool extracts features from the data to be judged across five dimensions: data integrity, metadata, data flow, data semantics, and higher-order attributes. Multi-dimensional feature extraction includes data integrity feature extraction, metadata feature extraction, data flow feature extraction, data semantic feature extraction, and higher-order attribute feature extraction. The multi-dimensional data features include integrity data features obtained from data integrity feature extraction, metadata data features obtained from metadata feature extraction, data flow data features obtained from data flow feature extraction, semantic data features obtained from data semantic feature extraction, and attribute data features obtained from higher-order attribute feature extraction.

[0007] According to the present invention, a feature extraction method for ownership determination is provided, which extracts data stream features from the data to be determined using a feature extraction tool to obtain data stream data features. The method includes: dividing the data file containing the data to be determined into blocks using the feature extraction tool to obtain block data files; performing hash calculations on the block data files using block hashing and fuzzy hashing techniques to obtain a fuzzy hash set; and encoding the fuzzy hash set to obtain the data stream data features corresponding to the data to be determined.

[0008] According to the present invention, a feature extraction method for ownership determination is provided, which extracts semantic data features from the data to be determined using a feature extraction tool, including: determining the data modality of the data to be determined and the fuzzy hash algorithm corresponding to the data modality based on the feature extraction tool; performing hash processing on the data to be determined corresponding to different data modalities based on the fuzzy hash algorithm to obtain the semantic data features corresponding to the data to be determined.

[0009] According to the present invention, a feature extraction method for ownership determination includes attribute data features obtained by extracting high-order attribute features from the data to be determined using a feature extraction tool, comprising: The data modality of the data to be judged and the corresponding learning algorithm are determined based on the feature extraction tool; the learning algorithm includes machine learning algorithm, deep learning algorithm and statistical analysis algorithm; features are extracted from the data to be judged corresponding to different data modalities based on the machine learning algorithm, deep learning algorithm and statistical analysis algorithm to obtain the attribute data features corresponding to the data to be judged.

[0010] According to the present invention, a feature extraction method for ownership determination is provided, which extracts integrity features from the data to be determined using a feature extraction tool to obtain integrity data features, including: performing integrity hash calculation on the data to be determined using the feature extraction tool to obtain the integrity data features corresponding to the data to be determined.

[0011] According to the present invention, a feature extraction method for ownership determination is provided, which extracts metadata features from the data to be determined using a feature extraction tool to obtain metadata data features, including: extracting metadata from the data to be determined using a feature extraction tool to obtain metadata data features corresponding to the data to be determined.

[0012] According to the feature extraction method for ownership determination provided by the present invention, after performing multi-dimensional feature extraction on the data to be determined using a feature extraction tool to obtain multi-dimensional data features of the data to be determined, the method further includes: performing correlation analysis based on integrity data features, metadata data features, data flow data features, semantic data features and attribute data features to obtain the ownership knowledge graph corresponding to the data to be determined, and performing infringement determination on the data to be determined based on the ownership knowledge graph.

[0013] The present invention also provides a feature extraction device for ownership determination, comprising the following modules.

[0014] The data acquisition module is used to acquire the data to be determined for ownership determination.

[0015] The feature extraction module is used to extract multi-dimensional features from the data to be judged using feature extraction tools, thereby obtaining multi-dimensional data features of the data to be judged. The feature extraction tools refer to tools that extract features from the data to be judged in five dimensions: data integrity, metadata, data flow, data semantics, and higher-order attributes. Multi-dimensional feature extraction includes data integrity feature extraction, metadata feature extraction, data flow feature extraction, data semantic feature extraction, and higher-order attribute feature extraction. Multi-dimensional data features include integrity data features obtained from data integrity feature extraction, metadata data features obtained from metadata feature extraction, data flow data features obtained from data flow feature extraction, semantic data features obtained from data semantic feature extraction, and attribute data features obtained from higher-order attribute feature extraction.

[0016] 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 extraction methods for ownership determination.

[0017] 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 extraction methods for ownership determination.

[0018] 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 extraction methods for ownership determination.

[0019] This invention provides a feature extraction method, apparatus, electronic device, and medium for ownership determination. The method involves acquiring data to be determined for ownership determination; performing multi-dimensional feature extraction on the data to be determined using a feature extraction tool to obtain multi-dimensional data features of the data to be determined; wherein the feature extraction tool refers to a tool for extracting features from the data to be determined in five dimensions: data integrity, metadata, data flow, data semantics, and higher-order attributes; multi-dimensional feature extraction includes data integrity feature extraction, metadata feature extraction, data flow feature extraction, data semantic feature extraction, and higher-order attribute feature extraction; and multi-dimensional data features include integrity data features obtained from data integrity feature extraction, metadata data features obtained from metadata feature extraction, data flow data features obtained from data flow feature extraction, semantic data features obtained from data semantic feature extraction, and attribute data features obtained from higher-order attribute feature extraction. The technical solution of this invention addresses the shortcomings of existing ISCC feature extraction methods, which do not support the extraction of high-dimensional features from data, resulting in low accuracy in ownership determination based on extracted features. This invention achieves multi-dimensional data features by extracting multi-dimensional features from the data to be determined, thereby improving the accuracy of subsequent data display and similarity accuracy. It can be used for data ownership determination in the circulation of data elements. Attached Figure Description

[0020] 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.

[0021] Figure 1 This is a flowchart illustrating the feature extraction method for ownership determination provided by the present invention.

[0022] Figure 2 This is a schematic diagram of the feature extraction device for ownership determination provided by the present invention.

[0023] Figure 3 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0024] 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.

[0025] The following is combined with Figure 1 The present invention describes a feature extraction method for ownership determination. This method is applicable to multi-dimensional feature extraction for ownership determination. The execution subject of this method can be an electronic device or a feature extraction device for ownership determination installed in the electronic device. The feature extraction device for ownership determination can be implemented by software, hardware, or a combination of both. Figure 1 This is a flowchart illustrating the feature extraction process for ownership determination provided by the present invention, as shown below. Figure 1 As shown, the method includes the following steps 101 and 102.

[0026] Step 101: Obtain the data to be determined for ownership determination.

[0027] In this step, the data to be determined is the data whose ownership needs to be determined to ascertain whether any infringement has occurred. The data to be determined can be multimodal data, such as text, images, audio, structured data, and combinations of multiple modalities. This embodiment does not limit this.

[0028] The modality combining multiple modalities can be, for example, a data report in the form of a Portable Document Format (PDF), including text and images, but this embodiment is not limited to this.

[0029] Specifically, obtain the data to be determined for ownership determination.

[0030] Step 102: Use the feature extraction tool to extract multi-dimensional features from the data to be judged, and obtain the multi-dimensional data features of the data to be judged.

[0031] Multidimensional feature extraction refers to extracting features from multiple different dimensions. Multidimensional data features refer to features in multiple dimensions, that is, data features are composed of features in multiple different dimensions.

[0032] In this step, the feature extraction tool refers to the tool that extracts features from the data to be judged in five dimensions: data integrity, metadata, data flow, data semantics, and higher-order attributes. Multi-dimensional feature extraction includes data integrity feature extraction, metadata feature extraction, data flow feature extraction, data semantic feature extraction, and higher-order attribute feature extraction. Multi-dimensional data features include integrity data features obtained from data integrity feature extraction, metadata data features obtained from metadata feature extraction, data flow data features obtained from data flow feature extraction, semantic data features obtained from data semantic feature extraction, and attribute data features obtained from higher-order attribute feature extraction.

[0033] Specifically, after obtaining the data to be judged, the data is input into the feature extraction tool. The feature extraction tool performs multi-dimensional feature extraction on the data to be judged, that is, it starts from five dimensions: data integrity, metadata, data flow, data semantics, and higher-order attributes. These dimensions characterize the data from low to high dimensions, thereby obtaining the multi-dimensional data features corresponding to the data to be judged.

[0034] In one specific implementation, the integrity data features are obtained by extracting integrity features from the data to be judged using a feature extraction tool, including: performing integrity hash calculation on the data to be judged using the feature extraction tool to obtain the integrity data features corresponding to the data to be judged.

[0035] In this step, the integrity data feature is also called the first-dimensional data feature obtained by extracting the first-dimensional feature from the data to be judged.

[0036] Specifically, after obtaining the data to be judged, the data to be judged is input into the feature extraction tool after the first dimension feature extraction is performed. The feature extraction tool performs integrity hash calculation on the data to be judged to obtain a digital digest, and the digital digest is determined as the integrity data feature corresponding to the data to be judged.

[0037] In one specific implementation, the metadata data features are obtained by extracting metadata features from the data to be judged using a feature extraction tool, including: extracting metadata from the data to be judged using a feature extraction tool to obtain the metadata data features corresponding to the data to be judged.

[0038] In this step, the metadata data features are also referred to as the second-dimensional data features obtained by extracting the second-dimensional features from the data to be judged.

[0039] Metadata extracted during metadata feature extraction includes data name, data size, data type, data format, data industry, application scenario, data description, data keywords, information format, data length, precautions, creation time, modification time, owner, and ownership application submission time, etc. This embodiment does not limit these.

[0040] Specifically, after obtaining the data to be judged, the second-dimensional feature extraction is performed. The data to be judged is then input into the feature extraction tool, which extracts the original data, including data name, data size, data type, data format, data industry, application scenario, data description, data keywords, information format, data length, precautions, creation time, modification time, owner, and ownership application submission time. All extracted metadata is then identified as the metadata data features corresponding to the data to be judged.

[0041] In one specific implementation, the data stream features are obtained by extracting data stream features from the data to be judged using a feature extraction tool, including: dividing the data file containing the data to be judged into blocks using the feature extraction tool to obtain block data files; performing hash calculations on the block data files using block hashing and fuzzy hashing techniques to obtain a fuzzy hash set; and encoding the fuzzy hash set to obtain the data stream features corresponding to the data to be judged.

[0042] In this step, the data stream data features are also referred to as the third-dimensional data features obtained by extracting the third-dimensional features from the data to be judged.

[0043] Block hashing is a technique for calculating hash values ​​on large files without loading the entire file at once. It involves dividing the file into smaller blocks, calculating a hash value for each block, and then combining the hash values ​​from all blocks to form the final hash result. Fuzzy hashing, on the other hand, is a hashing technique that fragments the content. Its main principle is to use a weak hash to calculate the hash value of a local part of the file, fragment the file under specific conditions, and then use a strong hash to calculate the hash value of each fragment to obtain the final hash result.

[0044] Specifically, after obtaining the data to be judged, the data to be judged is input into the feature extraction tool after the third-dimensional feature extraction is performed. The feature extraction tool uses block hashing and fuzzy hashing techniques to perform hash calculations on the block data file to obtain a fuzzy hash set. Then, all hash values ​​in the fuzzy hash set are encoded to obtain the data stream features corresponding to the data to be judged.

[0045] In one specific implementation, semantic data features are obtained by extracting semantic features from the data to be judged using a feature extraction tool, including: determining the data modality of the data to be judged and the fuzzy hash algorithm corresponding to the data modality based on the feature extraction tool; and performing hash processing on the data to be judged corresponding to different data modalities based on the fuzzy hash algorithm to obtain the semantic data features corresponding to the data to be judged.

[0046] In this step, semantic data features are also referred to as fourth-dimensional data features obtained by extracting fourth-dimensional features from the data to be judged.

[0047] Fuzzy hashing algorithms refer to algorithms that differ depending on the data modality. For example, if the data modality is image, the corresponding fuzzy hashing algorithm is suitable for image modality; if the data modality is text, the corresponding fuzzy hashing algorithm is suitable for text modality. Supported data modalities include, but are not limited to, text, images, video, audio, structured tables, and database types. For mixed types, the data to be determined in the mixed data modality needs to be extracted into individual data modalities for separate ownership determination; this embodiment does not limit this.

[0048] Specifically, after obtaining the data to be judged, the data to be judged is input into the feature extraction tool while performing fourth-dimensional feature extraction. The feature extraction tool determines the data modality of the data to be judged and the corresponding fuzzy hash algorithm. Then, the data to be judged corresponding to different data modalities is hashed according to the fuzzy hash algorithm corresponding to the data modality to obtain the semantic data features corresponding to the data to be judged.

[0049] In one specific implementation, the attribute data features obtained by extracting high-order attribute features from the data to be judged using a feature extraction tool include: determining the data modality of the data to be judged and the learning algorithm corresponding to the data modality based on the feature extraction tool; wherein, the learning algorithm includes machine learning algorithm, deep learning algorithm and statistical analysis algorithm; and extracting features from the data to be judged corresponding to different data modalities based on the machine learning algorithm, deep learning algorithm and statistical analysis algorithm to obtain the attribute data features corresponding to the data to be judged.

[0050] In this step, the attribute data features are also referred to as the fifth-dimensional data features obtained by extracting the fifth-dimensional features from the data to be judged.

[0051] The learning algorithms include machine learning algorithms, deep learning algorithms, and statistical analysis algorithms. These algorithms are all suitable for high-dimensional feature extraction of their respective data modalities, and this embodiment does not impose any limitations on them. Specifically, if the data modality is an image modality, the learning algorithm will be a high-dimensional feature extraction algorithm suitable for image modalities; if the data modality is a text modality, the learning algorithm will be a high-dimensional feature extraction algorithm suitable for text modalities. Supported modalities include, but are not limited to, text, images, video, audio, structured tables, and database types. For mixed types, the data to be determined in the mixed data modality needs to be extracted into individual data modalities for separate weight determination; this embodiment does not impose any limitations on this.

[0052] Specifically, after obtaining the data to be judged, the data to be judged is input into the feature extraction tool after the fifth dimension feature extraction is performed. The feature extraction tool determines the data mode of the data to be judged and the corresponding learning algorithm. The learning algorithm corresponding to the data mode is used to extract features from the data to be judged corresponding to different data modes to obtain the attribute data features corresponding to the data to be judged.

[0053] In one specific implementation, after extracting multi-dimensional features from the data to be judged using a feature extraction tool to obtain multi-dimensional data features of the data to be judged, the method further includes: performing correlation analysis based on integrity data features, metadata data features, data flow data features, semantic data features, and attribute data features to obtain the ownership knowledge graph corresponding to the data to be judged, and making an infringement judgment on the data to be judged based on the ownership knowledge graph.

[0054] In this step, during the infringement determination process, the first dimension of feature extraction, the integrity data feature, is a cryptographic hash, which has an avalanche effect and can be directly compared with the data to be judged to determine the infringement determination result. The second dimension of feature extraction, the metadata data feature, directly compares each metadata with the data to be judged, or compares the distance between each metadata and the data to be judged to determine the infringement determination result. The third dimension of feature extraction, the data stream data feature, compares the fuzzy hash of the binary data stream with the data to be judged using Hamming distance to determine the infringement determination result. The fourth dimension of feature extraction, the semantic data feature, compares the fuzzy hash of the data content with the data to be judged using Hamming distance and Jaccard distance to determine the infringement determination result. The fifth dimension of feature extraction, the attribute data feature, performs high-dimensional feature comparison, comparing the attribute data feature with the data to be judged using six distances: Euclidean distance, cosine similarity, Chebyshev distance, Manhattan distance, Jaccard distance, and Minkowski distance to determine the infringement determination result.

[0055] The ownership knowledge graph contains the relationships between various features.

[0056] Specifically, in the process of determining the infringement of data to be judged based on integrity data features, metadata data features, data flow data features, semantic data features, and attribute data features, the extracted integrity data features, metadata data features, data flow data features, semantic data features, and attribute data features can be correlated to obtain the ownership knowledge graph corresponding to the data to be judged, and the infringement of the data to be judged can be determined based on the correlation between the features contained in the ownership knowledge graph.

[0057] The advantage of this setup is that it extracts features from five dimensions: data integrity, metadata, data flow, data semantics, and higher-order attributes. This allows for infringement determination based on these five features, improving the accuracy and precision of infringement assessment. Furthermore, it can process data of any data modality. This addresses the issue that ISCC feature extraction does not support feature extraction from structured data, which constitutes a significant portion of data transactions / circulation, and also resolves the problem that ISCC feature extraction does not support the extraction of high-dimensional features.

[0058] This invention provides a feature extraction method for ownership determination, which involves acquiring data to be determined for ownership determination; performing multi-dimensional feature extraction on the data to be determined using a feature extraction tool to obtain multi-dimensional data features of the data to be determined; wherein, the feature extraction tool refers to a tool for extracting features from the data to be determined in five dimensions: data integrity, metadata, data flow, data semantics, and higher-order attributes; multi-dimensional feature extraction includes data integrity feature extraction, metadata feature extraction, data flow feature extraction, data semantic feature extraction, and higher-order attribute feature extraction; multi-dimensional data features include integrity data features obtained from data integrity feature extraction, metadata data features obtained from metadata feature extraction, data flow data features obtained from data flow feature extraction, semantic data features obtained from data semantic feature extraction, and attribute data features obtained from higher-order attribute feature extraction. Based on the above embodiments, the technical solution of the present invention is used to solve the defect that the existing ISCC feature extraction method does not support the feature extraction of high-dimensional features of data, resulting in low accuracy of ownership determination based on the extracted features. It realizes the extraction of multi-dimensional features from the data to be determined to obtain multi-dimensional data features, thereby improving the accuracy of subsequent data display and similarity accuracy. It can be used for data ownership determination in the circulation of data elements.

[0059] The following describes the feature extraction device for ownership determination provided by the present invention. The feature extraction device for ownership determination described below and the feature extraction method for ownership determination described above can be referred to in correspondence.

[0060] Figure 2 This is a schematic diagram of the feature extraction device for ownership determination provided by the present invention, with reference to... Figure 2 As shown, the feature extraction device 200 for ownership determination includes: a data acquisition module 201 and a feature extraction module 202.

[0061] The data acquisition module 201 is used to acquire the data to be determined for ownership determination.

[0062] The feature extraction module 202 is used to perform multi-dimensional feature extraction on the data to be judged using a feature extraction tool, thereby obtaining multi-dimensional data features of the data to be judged. The feature extraction tool refers to a tool for extracting features from the data to be judged in five dimensions: data integrity, metadata, data flow, data semantics, and higher-order attributes. Multi-dimensional feature extraction includes data integrity feature extraction, metadata feature extraction, data flow feature extraction, data semantic feature extraction, and higher-order attribute feature extraction. The multi-dimensional data features include integrity data features obtained from data integrity feature extraction, metadata data features obtained from metadata feature extraction, data flow data features obtained from data flow feature extraction, semantic data features obtained from data semantic feature extraction, and attribute data features obtained from higher-order attribute feature extraction.

[0063] In one example embodiment, the feature extraction module 202 extracts data stream features from the data to be judged using a feature extraction tool. Specifically, it performs the following steps: it divides the data file containing the data to be judged into blocks using the feature extraction tool to obtain a block data file; it performs hash calculations on the block data file using block hashing and fuzzy hashing techniques to obtain a fuzzy hash set; and it encodes the fuzzy hash set to obtain the data stream features corresponding to the data to be judged.

[0064] In one example embodiment, the feature extraction module 202 extracts semantic data features from the data to be judged using a feature extraction tool. Specifically, it is used to: determine the data modality of the data to be judged and the fuzzy hash algorithm corresponding to the data modality based on the feature extraction tool; and perform hash processing on the data to be judged corresponding to different data modalities based on the fuzzy hash algorithm to obtain the semantic data features corresponding to the data to be judged.

[0065] In one example embodiment, the feature extraction module 202, based on the attribute data features obtained by the feature extraction tool through high-order attribute feature extraction of the data to be judged, is specifically used for: determining the data modality of the data to be judged and the learning algorithm corresponding to the data modality based on the feature extraction tool; wherein, the learning algorithm includes machine learning algorithm, deep learning algorithm and statistical analysis algorithm; and performing feature extraction on the data to be judged corresponding to different data modalities based on the machine learning algorithm, deep learning algorithm and statistical analysis algorithm to obtain the attribute data features corresponding to the data to be judged.

[0066] In one example embodiment, the feature extraction module 202 extracts integrity features from the data to be judged using a feature extraction tool, specifically by performing integrity hash calculations on the data to be judged using the feature extraction tool to obtain the integrity features corresponding to the data to be judged.

[0067] In one example embodiment, the feature extraction module 202 extracts metadata features from the data to be judged using a feature extraction tool to obtain metadata data features. Specifically, it is used to: extract metadata from the data to be judged using a feature extraction tool to obtain metadata data features corresponding to the data to be judged.

[0068] In one example embodiment, the device further includes an infringement determination module. The infringement determination module is configured to: after extracting multi-dimensional features from the data to be determined using a feature extraction tool to obtain multi-dimensional data features of the data to be determined, perform correlation analysis based on integrity data features, metadata data features, data flow data features, semantic data features, and attribute data features to obtain an ownership knowledge graph corresponding to the data to be determined, and perform infringement determination on the data to be determined based on the ownership knowledge graph.

[0069] The apparatus of this embodiment can be used to execute the method of any embodiment in the side embodiment of the feature extraction method for ownership determination. Its specific implementation process and technical effects are similar to those in the side embodiment of the feature extraction method for ownership determination. For details, please refer to the detailed description in the side embodiment of the feature extraction method for ownership determination, which will not be repeated here.

[0070] 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 communications bus 340, wherein the processor 310, the communications interface 320, and the memory 330 communicate with each other through the communications bus 340. The processor 310 can call logical instructions in the memory 330 to execute a feature extraction method for ownership determination. This method includes: acquiring data to be determined for ownership determination; performing multi-dimensional feature extraction on the data to be determined using a feature extraction tool to obtain multi-dimensional data features of the data to be determined; wherein the feature extraction tool refers to a tool for extracting features from the data to be determined in five dimensions: data integrity, metadata, data flow, data semantics, and higher-order attributes; multi-dimensional feature extraction includes data integrity feature extraction, metadata feature extraction, data flow feature extraction, data semantic feature extraction, and higher-order attribute feature extraction; multi-dimensional data features include integrity data features obtained from data integrity feature extraction, metadata data features obtained from metadata feature extraction, data flow data features obtained from data flow feature extraction, semantic data features obtained from data semantic feature extraction, and attribute data features obtained from higher-order attribute feature extraction.

[0071] 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.

[0072] 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 extraction method for ownership determination provided by the above methods. The method includes: acquiring data to be determined for ownership determination; performing multi-dimensional feature extraction on the data to be determined using a feature extraction tool to obtain multi-dimensional data features of the data to be determined; wherein, the feature extraction tool refers to a tool for extracting features of five dimensions: data integrity, metadata, data flow, data semantics, and higher-order attributes from the data to be determined; multi-dimensional feature extraction includes data integrity feature extraction, metadata feature extraction, data flow feature extraction, data semantic feature extraction, and higher-order attribute feature extraction; multi-dimensional data features include integrity data features obtained by extracting data integrity features from the data to be determined, metadata data features obtained by extracting metadata features from the data to be determined, data flow data features obtained by extracting data flow features from the data to be determined, semantic data features obtained by extracting data semantic features from the data to be determined, and attribute data features obtained by extracting higher-order attribute features from the data to be determined.

[0073] 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 extraction method for ownership determination provided by the above-described methods. The method includes: acquiring ownership-determining data to be determined; performing multi-dimensional feature extraction on the data to be determined using a feature extraction tool to obtain multi-dimensional data features of the data to be determined; wherein, the feature extraction tool refers to a tool for extracting features from the data to be determined in five dimensions: data integrity, metadata, data flow, data semantics, and higher-order attributes; multi-dimensional feature extraction includes data integrity feature extraction, metadata feature extraction, data flow feature extraction, data semantic feature extraction, and higher-order attribute feature extraction; and multi-dimensional data features include integrity data features obtained by extracting data integrity features from the data to be determined, metadata data features obtained by extracting metadata features from the data to be determined, data flow data features obtained by extracting data flow features from the data to be determined, semantic data features obtained by extracting data semantic features from the data to be determined, and attribute data features obtained by extracting higher-order attribute features from the data to be determined.

[0074] 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.

[0075] 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.

[0076] 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 extraction method for ownership determination, characterized in that, include: Obtain the data to be determined for ownership assessment; The data to be judged is subjected to multi-dimensional feature extraction using a feature extraction tool to obtain multi-dimensional data features of the data to be judged. The feature extraction tool refers to a tool that extracts features from the data to be judged in five dimensions: data integrity, metadata, data flow, data semantics, and higher-order attributes. The multi-dimensional feature extraction includes data integrity feature extraction, metadata feature extraction, data flow feature extraction, data semantic feature extraction, and higher-order attribute feature extraction. The multi-dimensional data features include integrity data features obtained from data integrity feature extraction, metadata data features obtained from metadata feature extraction, data flow data features obtained from data flow feature extraction, semantic data features obtained from data semantic feature extraction, and attribute data features obtained from higher-order attribute feature extraction.

2. The feature extraction method for ownership determination according to claim 1, characterized in that, The data stream features are obtained by performing data stream feature extraction on the data to be determined using the feature extraction tool, including: The feature extraction tool is used to divide the data file containing the data to be determined into blocks to obtain the block data file. The block data file is hashed using block hashing and fuzzy hashing techniques to obtain a fuzzy hash set. The fuzzy hash set is encoded to obtain the data stream data features corresponding to the data to be determined.

3. The feature extraction method for ownership determination according to claim 1, characterized in that, The semantic data features are obtained by extracting the semantic features of the data to be determined using the feature extraction tool, including: The feature extraction tool is used to determine the data modality of the data to be judged and the corresponding fuzzy hash algorithm. The fuzzy hash algorithm is used to hash the data to be determined corresponding to different data modalities to obtain the semantic data features corresponding to the data to be determined.

4. The feature extraction method for ownership determination according to claim 1, characterized in that, The attribute data features obtained by extracting higher-order attribute features from the data to be determined using the feature extraction tool include: The feature extraction tool is used to determine the data modality of the data to be judged and the learning algorithm corresponding to the data modality; wherein, the learning algorithm includes machine learning algorithm, deep learning algorithm and statistical analysis algorithm; Based on the machine learning algorithm, the deep learning algorithm, and the statistical analysis algorithm, feature extraction is performed on the data to be judged corresponding to different data modalities to obtain the attribute data features corresponding to the data to be judged.

5. The feature extraction method for ownership determination according to claim 1, characterized in that, The integrity data features are obtained by extracting integrity features from the data to be determined using the feature extraction tool, including: The integrity hash calculation is performed on the data to be judged using the feature extraction tool to obtain the integrity data feature corresponding to the data to be judged.

6. The feature extraction method for ownership determination according to claim 1, characterized in that, The metadata data features are obtained by performing metadata feature extraction on the data to be determined using the feature extraction tool, including: The metadata of the data to be judged is extracted using the feature extraction tool to obtain the metadata features corresponding to the data to be judged.

7. The feature extraction method for ownership determination according to any one of claims 1-6, characterized in that, After performing multi-dimensional feature extraction on the data to be judged using a feature extraction tool to obtain the multi-dimensional data features of the data to be judged, the method further includes: Based on the integrity data features, metadata data features, data flow data features, semantic data features, and attribute data features, a correlation analysis is performed to obtain the ownership knowledge graph corresponding to the data to be judged, and infringement judgment is made on the data to be judged based on the ownership knowledge graph.

8. A feature extraction device for ownership determination, characterized in that, include: The data acquisition module is used to acquire the data to be determined for ownership determination. The feature extraction module is used to perform multi-dimensional feature extraction on the data to be judged using a feature extraction tool, thereby obtaining multi-dimensional data features of the data to be judged. The feature extraction tool refers to a tool that extracts features from the data to be judged in five dimensions: data integrity, metadata, data flow, data semantics, and higher-order attributes. The multi-dimensional feature extraction includes data integrity feature extraction, metadata feature extraction, data flow feature extraction, data semantic feature extraction, and higher-order attribute feature extraction. The multi-dimensional data features include integrity data features obtained by extracting data integrity features from the data to be judged, metadata data features obtained by extracting metadata features from the data to be judged, data flow data features obtained by extracting data flow features from the data to be judged, semantic data features obtained by extracting data semantic features from the data to be judged, and attribute data features obtained by extracting higher-order attribute features from the data to be judged.

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 extraction method for ownership determination 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 extraction method for ownership determination as described in any one of claims 1 to 7.

Images