Data processing method, device, equipment and storage medium based on privacy protection

Abstract

The present application discloses a data processing method, device, equipment and storage medium based on privacy protection, which obtains pre-processed feature data, calls a trained deep learning model to identify the feature data, determines the privacy information and the type information corresponding to the privacy information; determines the information processing method according to the type information, and encrypts the privacy information based on the privacy information processing method to obtain the encrypted information; in response to the user's viewing instruction, obtains the data access rights corresponding to the user's viewing instruction through a smart contract; decrypts the encrypted information according to the data access rights to obtain the privacy data. The present application can effectively identify and classify privacy information through a trained deep learning model; at the same time, the data access rules are automatically executed through smart contracts to improve the efficiency and security of data processing.

Description

Technical Field

[0001] The present application relates to the field of data security technology, and specifically to a data processing method, device, equipment and storage medium based on privacy protection. Background Art

[0002] Existing sensitive information processing technologies mainly rely on keyword matching and rule setting to identify and filter sensitive information to prevent the spread and abuse of inappropriate information. Among them, keyword matching is the most common way to process sensitive information. It mainly compares the keywords in the text with the preset sensitive word library. If they are matched keywords, the text is processed accordingly, such as filtering, blocking or warning. Rule setting is a means of processing sensitive information based on certain logic and rules, which can be customized for specific situations, contexts or needs. For example, based on comprehensive factors such as IP address, user behavior, and access frequency, it can be determined whether a user is a robot, thereby restricting their access to content.

[0003] Because it is based on preset rules and patterns, it is difficult to cope with dynamically changing sensitive information content. When new sensitive words or variant expressions appear, the above methods may miss or misjudge, resulting in low efficiency and accuracy. In addition, as network attack methods continue to escalate, traditional data protection methods may not be able to effectively deal with advanced network attacks. For example, traditional encryption methods may be cracked, and traditional intrusion detection systems may not be able to identify new attack patterns. Summary of the invention

[0004] In view of this, the present application provides a data processing method, apparatus, device and storage medium based on privacy protection, which can effectively identify and classify privacy information; at the same time, the data access rules are automatically executed through smart contracts to improve the efficiency and security of data processing.

[0005] The present application provides a data processing method based on privacy protection, the method comprising:

[0006] Acquire preprocessed feature data, call the trained deep learning model to identify the feature data, and determine the privacy information and type information corresponding to the privacy information;

[0007] Determining an information processing method according to the type information, and encrypting the private information based on the private information processing method to obtain encrypted information;

[0008] In response to a user viewing instruction, obtaining data access rights corresponding to the user viewing instruction through a smart contract;

[0009] The encrypted information is decrypted according to the data access permission to obtain private data.

[0010] Optionally, the step of obtaining the preprocessed feature data comprises:

[0011] The user's initial data is obtained, natural language is obtained using natural language processing methods, and natural language is converted into machine language through part-of-speech quantification.

[0012] Optionally, the calling of a trained deep learning model to identify the feature data, the deep learning model includes:

[0013] The deep learning model adopts a recurrent neural network model, wherein the recurrent neural network model includes a traditional recurrent neural network, a long short-term memory network or a gated recurrent unit.

[0014] Optionally, determining the information processing method according to the type information, and encrypting the sensitive information based on the sensitive information processing method to obtain encrypted information includes:

[0015] Determine the target encryption method in a preset encryption mapping table according to the type information;

[0016] Based on the target encryption method, the private information is encrypted to obtain encrypted information.

[0017] Optionally, in response to the user viewing instruction, obtaining data access rights corresponding to the user viewing instruction through a smart contract, the method further includes:

[0018] Constructing a blockchain framework for data storage and access, wherein the blockchain framework receives an access request for target data in the user viewing instruction from the user demand side;

[0019] Executing the smart contract of the blockchain framework in a trusted execution environment to determine whether the user viewing instruction satisfies the access permission condition of the target data;

[0020] If satisfied, configure the corresponding data access permissions for the user.

[0021] Optionally, the method further comprises: obtaining the preprocessed feature data, calling the trained deep learning model to identify the feature data, and determining the privacy information and type information corresponding to the privacy information.

[0022] The privacy information selects a logical expression corresponding to the hierarchical modal logic as representation according to the type information.

[0023] Optionally, the determining the information processing method according to the type information, and encrypting the private information based on the private information processing method to obtain encrypted information, the method further includes:

[0024] Acquire the privacy information, and use a 3D sparse convolutional network to perform feature extraction to obtain 3D sparse features;

[0025] Determining a target encryption method in a preset encryption map based on the 3D sparse features;

[0026] The 3D sparse features are combined to generate an encryption key, and the private information is encrypted to obtain encrypted information.

[0027] Correspondingly, the present application provides a data processing device based on privacy protection, comprising:

[0028] A data acquisition module, used to acquire preprocessed feature data, call the trained deep learning model to identify the feature data, and determine the privacy information and the type information corresponding to the privacy information;

[0029] A data encryption module, used to determine an information processing method according to the type information, and encrypt the private information based on the private information processing method to obtain encrypted information;

[0030] An access permission module, used to obtain data access permission corresponding to a user viewing instruction through a smart contract in response to the user viewing instruction;

[0031] The data decryption module is used to decrypt the encrypted information according to the data access permission to obtain the private data.

[0032] In addition, the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the data processing method based on privacy protection as described in any of the above items are implemented.

[0033] On this basis, the present application also provides a computer-readable storage medium on which a computer program is stored, wherein when the computer program is executed by a processor, the steps of the data processing method based on privacy protection as described in any of the above items are implemented.

[0034] The present application provides a data processing method, device, equipment and storage medium based on privacy protection, which obtains pre-processed feature data, calls a trained deep learning model to identify the feature data, determines the privacy information and the type information corresponding to the privacy information; determines the information processing method according to the type information, and encrypts the privacy information based on the privacy information processing method to obtain encrypted information; in response to the user's viewing instruction, obtains the data access rights corresponding to the user's viewing instruction through a smart contract; decrypts the encrypted information according to the data access rights to obtain the privacy data. The present application can effectively identify and classify privacy information through a trained deep learning model; at the same time, the data access rules are automatically executed through smart contracts to improve the efficiency and security of data processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0036] Figure 1 A flowchart of a data processing method based on privacy protection provided in an embodiment of the present application;

[0037] Figure 2 A schematic diagram of a data processing device based on privacy protection provided in an embodiment of the present application

[0038] Figure 3 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0039] The technical solutions in the embodiments of the present application are clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application. In the absence of conflict, the following embodiments and their technical features can be combined with each other.

[0040] See also Figure 1 , Figure 1 A scenario diagram of a data processing method based on privacy protection provided in an embodiment of the present application.

[0041] exist Figure 1In the application scenario, the computing device 101 can obtain preprocessed feature data, call the trained deep learning model to identify the feature data, and determine the privacy information and the type information corresponding to the privacy information; the computing device 101 can determine the information processing method according to the type information, and encrypt the privacy information based on the privacy information processing method to obtain encrypted information; the computing device 101 can respond to the user viewing instruction and obtain the data access right corresponding to the user viewing instruction through the smart contract; the computing device 101 can decrypt the encrypted information according to the data access right to obtain the privacy data.

[0042] It should be noted that the computing device 101 can be hardware or software. When the computing device 101 is hardware, it can be implemented as a distributed cluster consisting of multiple servers or terminal devices, or it can be implemented as a single server or a single terminal device. When the computing device 101 is embodied as software, it can be installed in the hardware devices listed above. It can be implemented as multiple software or software modules for providing distributed services, for example, or it can be implemented as a single software or software module. No specific limitation is made here.

[0043] In addition, this application can be applied to the identification and protection of privacy information of medical and health data, customer information protection in financial services, or the security management of internal enterprise documents. For example, in the field of medical and health care, this technical solution can be used to identify and encrypt sensitive information in patients' medical records, identify sensitive data in personal health information (such as medical history, drug reactions, etc.) through deep learning models, and use blockchain technology to ensure the security and compliance of data access.

[0044] See also Figure 2 , Figure 2 A flowchart of a data processing method based on privacy protection provided according to an embodiment of the present specification is shown.

[0045] S1. Obtain preprocessed feature data, call the trained deep learning model to identify the feature data, and determine the privacy information and type information corresponding to the privacy information.

[0046] In a possible implementation, the user's initial data is obtained, natural language is obtained using a natural language processing method, and the natural language is converted into machine language through part-of-speech quantification; wherein the natural language processing method includes data preprocessing, feature extraction, key information identification, context understanding and association analysis.

[0047] Specifically, the natural language processing method includes the following steps:

[0048] S101. Data preprocessing: Clean and format the user's initial data, including removing irrelevant characters (such as punctuation marks, special symbols), unifying text formats, and segmenting vocabulary. Cleaning and formatting are intended to reduce the complexity of model processing and improve the accuracy and efficiency of subsequent analysis.

[0049] S102. Use TF-IDF, Word2Vec and other methods to convert the text of the initial data into a numerical form, namely a feature vector, so as to convert the semantic information of the text into a form recognizable by the model and facilitate the subsequent identification of sensitive information.

[0050] S103. Use NLP models (BERT, GPT pre-trained models) to analyze feature vectors and identify key information in feature vectors, such as personal name, address, and phone number, to quickly and accurately extract private information from large amounts of data and provide a basis for subsequent information classification and processing.

[0051] S104. Based on the key information, analyze the contextual information of the text, understand the relationship and meaning between the information, and ensure that the identification of private information does not only rely on the matching of keywords, but also combines the contextual environment to improve the accuracy and reliability of identification.

[0052] In one possible implementation, a trained deep learning model is called to identify feature data and determine privacy information and type information corresponding to the privacy information, including: calling the trained deep learning model to determine whether the feature data contains privacy information; and when the feature data contains target privacy information, calling the trained deep learning model to determine the type information of the privacy information.

[0053] Specifically, when training and optimizing deep learning models, the key steps include preprocessing and feature engineering, model training, evaluation and verification, hyperparameter adjustment and model optimization, hyperparameter adjustment and model optimization. The above steps need to comprehensively consider data characteristics, model architecture, hyperparameter adjustment, performance evaluation and continuous learning to ensure that the model can show good performance and accuracy in both training and actual application. Among them, the specific steps of deep learning model training and optimization include:

[0054] Preprocessing and feature engineering: Preprocess the data, such as text cleaning, standardization, and word embedding, and perform necessary feature engineering to extract features that are beneficial to model training. Preprocessing and feature engineering improve data quality so that the model can learn more effectively and extract useful information.

[0055] Model training: Use the prepared dataset to train the deep learning model. During the training process, the model parameters (such as learning rate, number of hidden layer units, batch size, etc.) are continuously adjusted to optimize the training process. Through the training process, the model learns to identify and classify private information from the dataset.

[0056] Evaluation and Validation: Use cross-validation to evaluate model performance and monitor key metrics such as accuracy, recall, and F1 score to ensure that the model has good generalization capabilities and maintains stable performance on different datasets.

[0057] Hyperparameter adjustment and model optimization: Based on the performance evaluation results, the model's hyperparameters are adjusted. Possible adjustments may include changing the network architecture, optimizing the regularization strategy, adjusting the learning rate, etc. Through continuous adjustment and optimization, the accuracy and efficiency of the model can be further improved.

[0058] Model deployment and continuous learning: Regularly retrain and update the model with new data to adapt to new data patterns and changes, maintain the timeliness and accuracy of the model, and adapt to the ever-changing data environment and needs.

[0059] In a possible implementation, a trained deep learning model is called to identify feature data. The deep learning model includes:

[0060] The deep learning model uses a recurrent neural network model, which includes a traditional recurrent neural network, a long short-term memory network, or a gated recurrent unit. The choice of the recurrent neural network model needs to be based on data characteristics and requirements. Different recurrent neural network architectures have different characteristics and applicable scenarios.

[0061] For example, for processing long sequence data, such as long text or time series data, long short-term memory networks and gated recurrent units are better choices, which can effectively solve the gradient vanishing problem and allow the network to learn long-term dependencies on longer sequences. Among them, the long short-term memory network avoids the gradient vanishing problem by introducing a "gate" structure to control the flow of information. The gating mechanism allows information to be transmitted between time steps, while forgetting unimportant information, making the long short-term memory network particularly suitable for processing data with long-term dependencies. The gated recurrent unit is another commonly used recurrent neural network variant. Its internal structure is similar to that of the long short-term memory network, but its structure is relatively simple. It improves efficiency by reducing the number of parameters and simplifying the calculation process, while maintaining the good performance of the long short-term memory network. In addition, for tasks such as short sequence data or text classification, the simple structure of the traditional recurrent neural network makes it easier to complete training and works well when processing short sequences.

[0062] In a possible implementation, type information is used to define different levels of privacy, which can be divided into high privacy, medium privacy and general privacy, clarify the classification standards of privacy information, ensure the accuracy and consistency of subsequent processing, and determine the corresponding marking rules according to the privacy level. Specifically, information involving personal privacy is marked as high privacy, and information involving internal company data is marked as medium privacy.

[0063] In another possible implementation, based on the automatic labeling and classification of the above-mentioned deep learning model, manual review can be performed to ensure the accuracy and rationality of the labeling. The manual review content may involve checking the labeling results, correcting erroneous labels, and adjusting the classification rules to prevent model misjudgment and ensure the ultimate accuracy and reliability of information classification.

[0064] S2. Determine an information processing method according to the type information, and encrypt the private information based on the private information processing method to obtain encrypted information.

[0065] In a possible implementation, an information processing method is determined according to type information, and private information is encrypted based on the private information processing method to obtain encrypted information, including: determining a target encryption method in a preset encryption mapping table according to the type information; and encrypting the private information based on the target encryption method to obtain encrypted information.

[0066] For example, different encryption methods can be used for different types of private information. For example, algorithms such as AES can be used to encrypt private information such as personal or company information. The specific mode of using the AES encryption algorithm can be determined according to the type of private information. The AES encryption algorithm includes AES-128, AES-192 or AES-256. The main difference between the above modes is the key length. The longer the key, the higher the encryption strength. For highly private private information, AES-256 can be used to encrypt it. The encrypted sensitive information is stored in a database or blockchain framework, such as using a database management system such as MySQL or Oracle. When storing, it should be noted that the encrypted information needs to be associated with the corresponding user ID for subsequent data query and processing.

[0067] S3. In response to the user's viewing instruction, the data access permission corresponding to the user's viewing instruction is obtained through the smart contract.

[0068] In one possible implementation, in response to a user viewing instruction, obtaining user identity information corresponding to the user viewing instruction, and determining data access rights based on the user identity information, it also includes: constructing a blockchain framework for data storage and access, the blockchain framework receiving an access request for target data in the user viewing instruction from the user demander; executing a smart contract of the blockchain framework in a trusted execution environment to determine whether the user viewing instruction meets the access rights conditions for the target data; if so, configuring the corresponding data access rights for the user demander.

[0069] Specifically, the blockchain framework provides an unalterable and highly transparent environment for data access and auditing, which helps to improve the security, authenticity and accuracy of data, reduce audit costs, improve audit efficiency, and strengthen internal control and external supervision. When building a blockchain framework suitable for data storage and access, select the appropriate blockchain type during the construction process, such as public chain, private chain or consortium chain, and determine the optimal blockchain type based on data security and sharing requirements.

[0070] Specifically, smart contracts are deployed on the blockchain framework. Smart contracts are used to automatically manage data access rights. Smart contracts automatically execute data access rules to improve the efficiency and security of data processing. After the smart contracts are deployed, a secure access control mechanism is implemented, including identity authentication and authorization. Technologies such as digital signatures and key management are used to ensure that only authorized users can access data, and all data access behaviors are recorded on the blockchain, including access time, visitor identity, and accessed data content. In addition, the blockchain framework is regularly maintained and updated, including upgrading smart contracts and optimizing blockchain framework performance, to ensure that the blockchain framework can adapt to changing technical and business needs and maintain the security and efficiency of the blockchain framework.

[0071] S4. Decrypt the encrypted information according to the data access rights to obtain the private data.

[0072] In actual applications, for users who have the right to access private information, it is necessary to decrypt the corresponding private information to generate the original private data. The decryption method can be based on the specific desensitization algorithm and encryption method, such as AES encryption algorithm, symmetric encryption algorithm, asymmetric encryption algorithm, etc.

[0073] In a possible implementation, preprocessed feature data is obtained, a trained deep learning model is called to identify the feature data, and private information and type information corresponding to the private information are determined. The method further includes:

[0074] The privacy information is represented by a logical expression corresponding to the hierarchical modal logic according to the type information. Through the multi-level structure of the hierarchical modal logic, the privacy information can be described and processed more carefully, thereby improving the accuracy of recognition.

[0075] Specifically, graded modal logic is an extended modal logic that introduces the concept of "level" or "quantity" on the basis of traditional modal logic. In traditional modal logic, modal operators usually represent necessity (□) and possibility (◇), while in graded modal logic, these modal operators are given quantitative meanings and can express the concepts of "at least how many" or "no more than how many", and the logical expressions in graded modal logic are associated with a specific level or quantity. For example, a graded necessity operator may mean "at least n instances satisfy a certain condition". In this embodiment, the identified privacy information (such as personal data, company information, etc.) is converted into a logical expression in graded modal logic, and the expressive power of graded modal logic is used to accurately describe the characteristics and level of privacy information. For example, different logical expressions are defined to represent the privacy level of information. For example, □1 can represent general privacy, □2 represents moderate privacy, and □3 represents high privacy.

[0076] In a possible implementation, an information processing method is determined according to the type information, and the private information is encrypted based on the private information processing method to obtain encrypted information. The method further includes:

[0077] Acquire privacy information, use a 3D sparse convolutional network to extract features, and obtain 3D sparse features; determine a target encryption method in a preset encryption mapping table based on the 3D sparse features; combine the 3D sparse features to generate an encryption key, encrypt the privacy information, and obtain encrypted information.

[0078] Through the 3D representation of private information, 3D sparse convolution can be applied for feature extraction. Based on the extracted features, the corresponding encryption algorithm can be designed, and the data can be effectively encrypted in combination with the features extracted by 3D sparse convolution. Through the customized encryption algorithm design, the security and uniqueness of the encryption process can be improved. Specific application scenarios include financial institutions, healthcare institutions, and government agencies that require customized protection of large amounts of diverse data. By designing encryption algorithms in combination with the features extracted by 3D sparse convolution, the security of sensitive data during storage and transmission can be ensured, and the encryption strategy can be dynamically adjusted according to the characteristics and risk level of the data to achieve both efficient and secure data protection.

[0079] Specifically, the 3D representation of privacy information is to convert the identified sensitive information (such as text, images, etc.) into a three-dimensional data format. In the above process, the information is converted into a three-dimensional matrix, and each dimension represents different characteristics of the information. By converting it into a three-dimensional data format, the 3D sparse convolutional network can be used more effectively to process the data, thereby improving the efficiency and effectiveness of subsequent encryption steps.

[0080] The present application provides a data processing method, device, equipment and storage medium based on privacy protection, which obtains pre-processed feature data, calls a trained deep learning model to identify the feature data, determines the privacy information and the type information corresponding to the privacy information; determines the information processing method according to the type information, and encrypts the privacy information based on the privacy information processing method to obtain encrypted information; in response to the user's viewing instruction, obtains the data access rights corresponding to the user's viewing instruction through a smart contract; decrypts the encrypted information according to the data access rights to obtain the privacy data. The present application can effectively identify and classify privacy information through a trained deep learning model; at the same time, the data access rules are automatically executed through smart contracts to improve the efficiency and security of data processing.

[0081] like Figure 2 As shown, the present application also provides a data processing device based on privacy protection, comprising:

[0082] The data acquisition module 101 is used to acquire the preprocessed feature data, call the trained deep learning model to identify the feature data, and determine the privacy information and the type information corresponding to the privacy information;

[0083] The data encryption module 102 is used to determine the information processing method according to the type information, and encrypt the private information based on the private information processing method to obtain encrypted information;

[0084] The access permission module 103 is used to obtain the data access permission corresponding to the user viewing instruction through the smart contract in response to the user viewing instruction;

[0085] The data decryption module 104 is used to decrypt the encrypted information according to the data access permission to obtain the private data.

[0086] In addition, the present application also provides an electronic device, such as Figure 3 As shown, it shows a schematic diagram of the structure of the electronic device involved in the embodiment of the present application, specifically:

[0087] The electronic device may include components such as a processor 301 with one or more processing cores, a memory 302 with one or more computer-readable storage media, a power supply 303, and an input unit 304. Those skilled in the art will appreciate that Figure 3The electronic device structure shown in the figure does not constitute a limitation on the electronic device, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0088] The processor 301 is the control center of the electronic device. It uses various interfaces and lines to connect various parts of the entire electronic device. By running or executing software programs and / or modules stored in the memory 302, and calling data stored in the memory 302, it executes various functions of the electronic device and processes data, thereby monitoring the electronic device as a whole. Optionally, the processor 301 may include one or more processing cores; preferably, the processor 301 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface, and application programs, and the modem processor mainly processes wireless communications. It is understandable that the above-mentioned modem processor may not be integrated into the processor 301.

[0089] The memory 302 can be used to store software programs and modules. The processor 301 executes various functional applications and image information communication methods by running the software programs and modules stored in the memory 302. The memory 302 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area may store data created according to the use of the electronic device, etc. In addition, the memory 302 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices. Accordingly, the memory 302 may also include a memory controller to provide the processor 301 with access to the memory 302.

[0090] The electronic device also includes a power supply 303 for supplying power to each component. Preferably, the power supply 303 can be logically connected to the processor 301 through a power management system, so that the power management system can manage charging, discharging, power consumption and other functions. The power supply 303 can also include one or more DC or AC power supplies, recharging systems, power failure detection circuits, power converters or inverters, power status indicators and other arbitrary components.

[0091] The electronic device may further include an input unit 304, which may be used to receive input digital or character information and generate keyboard, mouse, joystick, optical or trackball signal input related to user settings and function control.

[0092] Although not shown, the electronic device may further include a display unit, etc., which will not be described in detail herein. Specifically in this embodiment, the processor 301 in the electronic device will load the executable files corresponding to the processes of one or more application programs into the memory 302 according to the following instructions, and the processor 301 will run the application programs stored in the memory 302, thereby realizing various functions, as follows:

[0093] Acquire preprocessed feature data, call the trained deep learning model to identify the feature data, determine the privacy information and the type information corresponding to the privacy information; determine the information processing method according to the type information, and encrypt the privacy information based on the privacy information processing method to obtain encrypted information; in response to a user viewing instruction, obtain the data access right corresponding to the user viewing instruction through a smart contract; decrypt the encrypted information according to the data access right to obtain the privacy data.

[0094] The specific implementation of the above operations can be found in the previous embodiments, which will not be described in detail here.

[0095] The present application provides an electronic device, which obtains pre-processed feature data, calls a trained deep learning model to identify the feature data, determines the privacy information and the type information corresponding to the privacy information; determines the information processing method according to the type information, and encrypts the privacy information based on the privacy information processing method to obtain the encrypted information; in response to the user viewing instruction, obtains the data access rights corresponding to the user viewing instruction through a smart contract; decrypts the encrypted information according to the data access rights to obtain the privacy data. The present application can effectively identify and classify privacy information through the trained deep learning model; at the same time, the data access rules are automatically executed through the smart contract to improve the efficiency and security of data processing.

[0096] A person of ordinary skill in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be completed by instructions, or by controlling related hardware through instructions. The instructions may be stored in a computer-readable storage medium and loaded and executed by a processor.

[0097] To this end, an embodiment of the present application provides a storage medium in which a plurality of instructions are stored, and the instructions can be loaded by a processor to execute the steps in any one of the data processing methods based on privacy protection provided in the embodiment of the present application. For example, the instructions can execute the following steps:

[0098] Acquire preprocessed feature data, call the trained deep learning model to identify the feature data, determine the privacy information and the type information corresponding to the privacy information; determine the information processing method according to the type information, and encrypt the privacy information based on the privacy information processing method to obtain encrypted information; in response to a user viewing instruction, obtain the data access right corresponding to the user viewing instruction through a smart contract; decrypt the encrypted information according to the data access right to obtain the privacy data.

[0099] The specific implementation of the above operations can be found in the previous embodiments, which will not be described in detail here.

[0100] The storage medium may include: a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0101] Since the instructions stored in the storage medium can execute the steps of any one of the privacy protection-based data processing methods provided in the embodiments of the present application, the beneficial effects that can be achieved by any one of the privacy protection-based data processing methods provided in the embodiments of the present application can be achieved. Please refer to the previous embodiments for details and will not be repeated here.

[0102] Although the present application has been shown and described with respect to one or more implementations, those skilled in the art will recognize equivalent variations and modifications based on reading and understanding this specification and the accompanying drawings. The present application includes all such modifications and variations and is limited only by the scope of the appended claims.

[0103] That is, the above description is only an embodiment of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the specification and drawings of the present application, such as the mutual combination of technical features between the embodiments, or direct or indirect application in other related technical fields, are also included in the patent protection scope of the present application.

[0104] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0105] In addition, for structural elements with the same or similar characteristics, the present application may use the same or different reference numerals for identification. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more features. In the description of the present application, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0106] In the present application, in order to enable any person skilled in the art to implement and use the present application, the present application provides the above description. In the above description, various details are listed for the purpose of explanation. It should be understood that those of ordinary skill in the art can recognize that the present application can also be implemented without using these specific details. In other embodiments, well-known structures and processes will not be elaborated in detail to avoid unnecessary details that make the description of the present application obscure. Therefore, the present application is not intended to be limited to the embodiments shown, but is consistent with the widest range of principles and features disclosed in the present application.

Claims

1. A data processing method based on privacy protection, characterized in that: The method comprises: Acquire preprocessed feature data, call the trained deep learning model to identify the feature data, and determine the privacy information and type information corresponding to the privacy information; The method further comprises: obtaining the preprocessed feature data, calling the trained deep learning model to identify the feature data, determining the privacy information and the type information corresponding to the privacy information; and: The privacy information selects a logical expression corresponding to the hierarchical modal logic as representation according to the type information; Determining an information processing method according to the type information, and encrypting the private information based on the private information processing method to obtain encrypted information; Wherein, the information processing method is determined according to the type information, and the private information is encrypted based on the private information processing method to obtain encrypted information, and the method further includes: Acquire the privacy information, and use a 3D sparse convolutional network to perform feature extraction to obtain 3D sparse features; Determining a target encryption method in a preset encryption map based on the 3D sparse features; Combining the 3D sparse features to generate an encryption key, encrypting the private information to obtain encrypted information; The 3D representation of privacy information is to convert the identified sensitive information into a three-dimensional data format; In response to a user viewing instruction, obtaining data access rights corresponding to the user viewing instruction through a smart contract; The encrypted information is decrypted according to the data access permission to obtain private data.

2. The data processing method based on privacy protection according to claim 1, characterized in that: The specific steps of obtaining the preprocessed feature data include: The user's initial data is obtained, natural language is obtained using natural language processing methods, and natural language is converted into machine language through part-of-speech quantification.

3. The data processing method based on privacy protection according to claim 1, characterized in that: The calling of the trained deep learning model to identify the feature data, the deep learning model comprising: The deep learning model adopts a recurrent neural network model, wherein the recurrent neural network model includes a traditional recurrent neural network, a long short-term memory network or a gated recurrent unit.

4. The data processing method based on privacy protection according to claim 1, characterized in that: The determining of the information processing method according to the type information, and encrypting the private information based on the private information processing method to obtain encrypted information, includes: Determine the target encryption method in a preset encryption mapping table according to the type information; Based on the target encryption method, the private information is encrypted to obtain encrypted information.

5. The data processing method based on privacy protection according to claim 1, characterized in that: In response to the user viewing instruction, obtaining the data access permission corresponding to the user viewing instruction through the smart contract, the method further includes: Constructing a blockchain framework for data storage and access, wherein the blockchain framework receives an access request for target data in the user viewing instruction from the user demand side; Executing the smart contract of the blockchain framework in a trusted execution environment to determine whether the user viewing instruction satisfies the access permission condition of the target data; If satisfied, configure the corresponding data access permissions for the user.

6. A data processing device based on privacy protection, characterized in that: include: A data acquisition module, used to acquire preprocessed feature data, call the trained deep learning model to identify the feature data, and determine the privacy information and type information corresponding to the privacy information; The data acquisition module is specifically used to select a logical expression corresponding to hierarchical modal logic as representation of the privacy information according to the type information; A data encryption module, used to determine an information processing method according to the type information, and encrypt the private information based on the private information processing method to obtain encrypted information; The data encryption module is specifically used to obtain the private information, use a 3D sparse convolutional network to extract features, and obtain 3D sparse features; determine a target encryption method in a preset encryption mapping table based on the 3D sparse features; combine the 3D sparse features to generate an encryption key, encrypt the private information, and obtain encrypted information; wherein the 3D representation of the private information is to convert the identified sensitive information into a three-dimensional data format; An access permission module, used to obtain data access permission corresponding to a user viewing instruction through a smart contract in response to the user viewing instruction; The data decryption module is used to decrypt the encrypted information according to the data access permission to obtain the private data.

7. An electronic device, characterized in that: It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the data processing method based on privacy protection as described in any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium, characterized in that: A computer program is stored thereon, wherein when the computer program is executed by a processor, the steps of the data processing method based on privacy protection as described in any one of claims 1 to 5 are implemented.

Images