Real-time data classification and hierarchical encryption system and method based on deep learning

Abstract

The invention provides a real-time data classification and hierarchical encryption system and method based on deep learning, and the method comprises the steps: dividing real-time data into a plurality of independent data frames, recognizing a data attribute corresponding to each independent data frame, constructing a positive data representation corresponding to the real-time data according to the data attributes, dividing the real-time data into corresponding data sets according to positive data representation, constructing a random ciphertext sequence based on data attributes to perform encryption training on corresponding independent data frames, and constructing initially encrypted real-time data according to an encryption effect corresponding to each independent data frame, according to the method, a plurality of random encrypted modes contained in a data set are selected to locally encrypt the initially encrypted real-time data for a plurality of times, the security level of the real-time data is determined and stored, and different data can be classified and then correspondingly encrypted, so that the security of the data needing to be encrypted is guaranteed, the processing of non-encrypted data is reduced, and the security of the data needing to be encrypted is improved. And the security of the to-be-encrypted data is fully guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly relates to a real-time data classification and hierarchical encryption system and method based on deep learning. Background Art

[0002] With the continuous development of technology leading humanity into the digital age, information technology and digital technology have developed rapidly and profoundly changed all aspects of society. Driven by the data era, humans can obtain the information or data they need in a variety of convenient and fast ways. While enjoying the convenience and intelligence brought by the data era, humans also bear the confusion and risks it brings. For example: a series of problems such as data leakage, data tampering, data disclosure, etc. For individuals, the privacy in life cannot be protected; for enterprises and factories, the interests cannot be maintained. However, with the rapid development of technology, society and individuals are increasingly relying on digital tools, resulting in some people or enterprises being overly dependent on technology and unable to resist risks, nor can they enjoy information technology and digital technology with confidence. Currently, how to protect data that should not be made public and improve the public's trust and acceptance of information technology and digital technology has become an urgent problem to be solved.

[0003] Therefore, the present invention provides a real-time data classification and hierarchical encryption system and method based on deep learning. Summary of the Invention

[0004] The real-time data classification and hierarchical encryption system and method based on deep learning of the present invention can first classify different data and then perform corresponding encryption, which not only ensures the confidentiality of the data to be encrypted, but also reduces the processing of non-encrypted data, and fully guarantees the security of the data to be encrypted.

[0005] The present invention provides a real-time data classification and hierarchical encryption system based on deep learning, including:

[0006] A frame processing module for dividing real-time data into a plurality of independent data frames and identifying the data attributes corresponding to each independent data frame;

[0007] A data classification module for constructing a positive data representation corresponding to the real-time data according to the data attributes and dividing the real-time data into corresponding data sets according to the positive data representation;

[0008] A preliminary encryption module for constructing a random ciphertext sequence based on the data attributes to perform encryption training on the corresponding independent data frames, and constructing initially encrypted real-time data according to the encryption effect corresponding to each independent data frame;

[0009] A deep encryption module, which is used to perform several local encryptions on the initially encrypted real-time data by selecting several random encryption methods included in the dataset, determine the classification level of the real-time data and save it.

[0010] In an implementable manner,

[0011] The frame processing module includes:

[0012] A compression processing unit, which is used to perform compression encoding on the real-time data, screen several blank characters included in the encoding result, and perform corresponding division on the real-time data according to the distribution characteristics of the blank characters in the encoding result to obtain several independent data frames;

[0013] A preliminary screening unit, which is used to randomly select a corresponding number of datasets according to the number of frames of the independent data frames, respectively perform similarity analysis on each independent data frame and different datasets to construct a similarity matrix of the real-time data, and construct an initial matching dataset for each independent data frame according to the elements corresponding to the same matrix position in different similarity matrices;

[0014] An attribute analysis unit, which is used to construct several data matching methods of the real-time data based on the frame number of each independent data frame in the real-time data and the corresponding initial matching datasets, obtain the initial matching data attributes corresponding to each dataset, and respectively verify the attribute exclusion characteristics corresponding to each data matching method;

[0015] An attribute determination unit, which is used to select the target data matching method with the eigenvalue of the attribute exclusion characteristic being 0, and determine the data attribute corresponding to each independent data frame.

[0016] In an implementable manner,

[0017] It further includes:

[0018] An iterative screening unit, which is used to, when all the attribute exclusion characteristics are not 0, randomly select a corresponding number of updated datasets according to the number of frames of the independent data frames, and transmit the updated datasets to the preliminary screening unit for corresponding analysis.

[0019] In an implementable manner,

[0020] The data classification module includes:

[0021] An attribute processing unit, which is used to perform enhancement processing on the data attributes respectively, determine several enhancement items of the real-time data according to the result difference before and after enhancement, find the relevant sub-data corresponding to each enhancement item in the real-time data, assign a value to the enhancement item according to the data value corresponding to the relevant sub-data, and generate a positive data representation of the real-time data.

[0022] A condition generation unit, configured to respectively obtain the in-set features and in-set data disclosure information of each of the data sets, construct a characterization collection condition corresponding to the data set according to the in-set features, and construct an interval collection condition corresponding to the data set according to the in-set data disclosure information;

[0023] A classification execution unit, configured to match the characterization collection condition with the positive data characterization, match the interval collection condition with the data specification of the real-time data, obtain a data set that matches the real-time data, and classify the real-time data according to the set category of the data set.

[0024] In an implementable manner,

[0025] The condition generation unit includes:

[0026] A first generation subunit, configured to respectively obtain the data basic information corresponding to different in-set data in each of the data sets, construct strong sub-features corresponding to the data set according to the same information between the data basic information, obtain the data type corresponding to each in-set data, and combine the different information between the data basic information to construct weak sub-features corresponding to the data set, and generate in-set features corresponding to the data set according to the strong sub-features and the weak sub-features;

[0027] A second generation subunit, configured to respectively obtain the data disclosure information corresponding to different in-set data in each of the data sets, determine the data maximum value and the data minimum value corresponding to the in-set data according to the data disclosure information, obtain a plurality of first test numbers greater than the data maximum value and a plurality of second test numbers less than the data minimum value, and analyze the collectable data range of the data set by using the first test numbers and the second test numbers;

[0028] A third generation subunit, configured to construct a characterization collection condition for the data set according to a plurality of feature key points included in the in-set features, and construct an interval collection condition for the data set according to the collectable data range.

[0029] In an implementable manner,

[0030] The preliminary encryption module includes:

[0031] A ciphertext construction unit, configured to determine an invalid encryption method corresponding to the independent data frame according to the data attribute, delete the corresponding invalid ciphertext, determine a plurality of groups of valid ciphertexts of the real-time data, perform chaos processing on the valid ciphertexts, and generate a chaos ciphertext sequence of the real-time data;

[0032] An encryption training unit is used to select sequence segments from the chaotic ciphertext sequence to obtain random ciphertexts corresponding to each independent data frame, encrypt the corresponding independent data frame with the random ciphertexts, cover the corresponding independent data frame with the obtained encrypted data frame, and perform an encryption effect score on the corresponding random ciphertext according to the coverage rate corresponding to each encrypted data frame;

[0033] An encryption execution unit is used to select the target random ciphertext with the highest score for each independent data frame, sort the target random ciphertexts based on the sorting order of the independent data frames to generate the execution ciphertext of the real-time data, select the effective encryption method of the independent data according to the positive data representation, and encrypt the real-time data with the effective encryption method and the execution ciphertext to generate the initially encrypted real-time data.

[0034] In an implementable manner,

[0035] It further includes:

[0036] Determine several encryption defect positions corresponding to the encrypted data frame according to the encryption effect score;

[0037] Locate the encryption defect positions in the encrypted data frame and hide the encryption defect positions.

[0038] In an implementable manner,

[0039] The deep encryption module includes:

[0040] A method screening unit is used to determine the invalid encryption methods of the real-time data according to the positive data representation, select several random encryption methods except the invalid encryption methods in the data set, and determine the historical usage times corresponding to each random encryption method;

[0041] An encryption preparation unit is used to sort the random encryption methods in ascending order of the historical usage times, set corresponding order levels for the random encryption methods, and determine several encryptable data frames corresponding to each random encryption method according to the encryption attributes corresponding to each random encryption method and the data attributes of the independent data frames;

[0042] A local encryption unit is used to select the corresponding implemented random encryption method for each independent data frame according to the order level of the random encryption method corresponding to each independent data frame, and perform local encryption on the corresponding independent data frame with the implemented random encryption method;

[0043] A classification determination unit is used to determine the classification level of the real-time data according to the data source of the real-time data, add corresponding classification tags to the encrypted real-time data, and input them into the classification storage area of the corresponding dataset for storage.

[0044] The present invention provides a real-time data classification and hierarchical encryption method based on deep learning, including:

[0045] Step 1: Divide the real-time data into several independent data frames, and identify the data attributes corresponding to each independent data frame;

[0046] Step 2: Construct a positive data representation corresponding to the real-time data according to the data attributes, and divide the real-time data into corresponding datasets according to the positive data representation;

[0047] Step 3: Construct a random ciphertext sequence based on the data attributes to perform encryption training on the corresponding independent data frames, and construct initially encrypted real-time data according to the encryption effect corresponding to each independent data frame;

[0048] Step 4: Select several random encryption methods included in the dataset to perform several local encryptions on the initially encrypted real-time data, determine the classification level of the real-time data, and save it.

[0049] In an implementable manner,

[0050] The said Step 2 includes:

[0051] Step 21: Respectively perform enhancement processing on the data attributes, determine several enhancement items of the real-time data according to the result differences before and after enhancement, find the relevant sub-data corresponding to each enhancement item in the real-time data, and assign values to the enhancement items according to the data values corresponding to the relevant sub-data to generate the positive data representation of the real-time data;

[0052] Step 22: Respectively obtain the in-set features and in-set data public information of each dataset, construct a representation collection condition corresponding to the dataset according to the in-set features, and construct an interval collection condition corresponding to the dataset according to the in-set data public information;

[0053] Step 23: Match the representation collection condition with the positive data representation, match the interval collection condition with the data specification of the real-time data, obtain the dataset that matches the real-time data, and classify the real-time data according to the set category of the dataset.

[0054] The achievable beneficial effects of the above technical solution are as follows: In order to better classify and encrypt data, the collected real-time data is divided into several independent data frames, and the positive data representation of the real-time data is constructed by analyzing the data attributes of each independent data frame, so that the data set to which the real-time data belongs can be determined, and it is divided into the corresponding data set to complete the data classification work. Then, the data attributes are used to construct a random ciphertext sequence to encrypt and train the independent data frames, generating the corresponding initially encrypted real-time data. In order to ensure the encryption quality and further guarantee the encryption quality of the real-time data, the initially encrypted real-time data is locally encrypted using the random encryption method used by the existing in-set data in the data set. On the one hand, it can perform compensatory encryption for the weak parts of encryption, and on the other hand, it can perform deep encryption using the encryption method in the data set of the same classification, improving the quality and efficiency of the encryption result, enhancing the overall encryption effect of the real-time data, and achieving the purpose of synchronously generating data and performing complete encryption in a short time.

[0055] Other features and advantages of the present invention will be described in the following specification, and part of them will be obvious from the specification or understood by implementing the present invention. The objectives and other advantages of the present invention can be achieved and obtained by the structures specifically pointed out in the written specification and the drawings.

[0056] The technical solution of the present invention will be further described in detail below through the drawings and embodiments. Description of the Drawings

[0057] The drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation to the present invention. In the drawings:

[0058] Figure 1 It is a schematic diagram of the composition of the real-time data classification and hierarchical encryption system based on deep learning in the embodiment of the present invention;

[0059] Figure 2 It is a schematic diagram of the working process of the real-time data classification and hierarchical encryption method based on deep learning in the embodiment of the present invention. Detailed Embodiments

[0060] The following describes the preferred embodiments of the present invention with reference to the drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention and are not used to limit the present invention.

[0061] Embodiment 1

[0062] This embodiment provides a real-time data classification and hierarchical encryption system based on deep learning, as Figure 1 shown, including:

[0063] The frame processing module is used to divide the real-time data into several independent data frames and identify the data attributes corresponding to each independent data frame;

[0064] The data classification module is used to construct a positive data representation corresponding to the real-time data according to the data attributes, and divide the real-time data into corresponding data sets according to the positive data representation;

[0065] The preliminary encryption module is used to construct a random ciphertext sequence based on the data attributes to perform encryption training on the corresponding independent data frames, and construct the initially encrypted real-time data according to the encryption effect corresponding to each independent data frame;

[0066] The deep encryption module is used to select several random encrypted methods included in the data set to perform several local encryptions on the initially encrypted real-time data, determine the confidentiality level of the real-time data and save it.

[0067] In this example, the independent data frame represents the result of dividing the real-time data into sub-data;

[0068] In this example, the data attribute represents the attribute presented by the independent data frame;

[0069] In this example, the positive data representation represents the combined expression of the characteristics and attributes of the real-time data;

[0070] In this example, the data set represents the data set used to store the real-time data;

[0071] In this example, the initially encrypted real-time data represents the result of encrypting the independent data frames in the real-time data using a random ciphertext sequence;

[0072] In this example, the local encryption represents the process of encrypting a data part in the initially encrypted real-time data using a random encrypted method;

[0073] In this example, the real-time data with corresponding confidentiality levels in the same data set is stored in the same area.

[0074] Working principle and beneficial effects of the above technical solution: In order to better classify and encrypt data, the collected real-time data is divided into several independent data frames. By analyzing the data attributes of each independent data frame, a positive data representation of the real-time data is constructed, so as to determine the data set to which the real-time data belongs and complete the data classification work by dividing it into the corresponding data set. Then, the data attributes are used to construct a random ciphertext sequence to encrypt and train the independent data frames, generating the corresponding initially encrypted real-time data. In order to ensure the encryption quality and further guarantee the encryption quality of real-time data, the initially encrypted real-time data is locally encrypted by using the random encryption method used for the existing in-set data in the data set. On the one hand, it can perform compensatory encryption for the weak parts of encryption, and on the other hand, it can perform deep encryption by using the encryption method in the data set of the same classification, improving the quality and efficiency of the encryption result and enhancing the overall encryption effect of real-time data, achieving the purpose of synchronously generating data and completing the entire encryption in a short time.

[0075] Embodiment 2

[0076] Based on Embodiment 1, for the real-time data classification and hierarchical encryption system based on deep learning, the frame processing module includes:

[0077] A compression processing unit, configured to perform compression encoding on the real-time data, screen a plurality of blank characters included in the encoding result, and perform corresponding division on the real-time data according to the distribution characteristics of the blank characters in the encoding result to obtain a plurality of independent data frames;

[0078] A preliminary screening unit, configured to randomly select a corresponding number of data sets according to the number of frames of the independent data frames, respectively perform similarity analysis on each independent data frame and different data sets to construct a similarity matrix of the real-time data, and construct an initial matching data set for each independent data frame according to the elements corresponding to the same matrix position in different similarity matrices;

[0079] An attribute analysis unit, configured to construct several data matching methods of the real-time data based on the frame number of each independent data frame in the real-time data and the corresponding several initial matching data sets, obtain the initial matching data attributes corresponding to each data set, and respectively verify the attribute exclusion characteristics corresponding to each data matching method;

[0080] An attribute determination unit, configured to select a target data matching method with a characteristic value of 0 for the attribute exclusion characteristic, and determine the data attribute corresponding to each independent data frame.

[0081] In this example, compression encoding represents the process of compressing and re-encoding the real-time data;

[0082] In this example, the number of data sets is the same as the number of frames;

[0083] In this example, the similarity matrix represents a matrix generated by statistically locking the similarity between an independent data frame and different data sets;

[0084] In this example, the initial matching data set means that the independent data frame matches the data set;

[0085] In this example, the data matching method represents the method of matching the independent data frame in the real-time data to the data set generated after combining different initial matching data sets;

[0086] In this example, the attribute exclusion feature represents the feature presented when there is an exclusion between multiple initial matching data attributes included in a data matching method.

[0087] The working principle and beneficial effects of the above technical solution: In order to ensure the accuracy of data classification and encryption, first, it is necessary to determine the data attributes of each independent data frame in the real-time data. First, determine the division position of the independent data frame by means of compression and encoding, so as to divide the real-time data into several independent data frames. Further, select a data set with the same number as the independent data frame, and construct a corresponding similarity matrix by analyzing the similarity between the independent data frame and different data sets. Thus, use the principle that similar matrix elements mean valid data to match the corresponding initial matching data set for the independent data frame. Further, sort the initial matching data sets to construct several data matching methods for the real-time data, and determine the data attributes corresponding to each independent data frame by analyzing the attribute exclusion features in different data matching methods. By determining the data attributes of the independent data frame, the first step of analyzing the real-time data is completed, determining the general direction of the next data classification, ensuring the effectiveness of classifying the real-time data, and effectively reducing errors.

[0088] Embodiment 3

[0089] Based on Embodiment 2, the real-time data classification and hierarchical encryption system based on deep learning further includes:

[0090] An iterative screening unit, configured to randomly select a corresponding number of updated data sets according to the number of frames of the independent data frame when all the attribute exclusion features are not 0, and transmit the updated data sets to the preliminary screening unit for corresponding analysis.

[0091] The working principle and beneficial effects of the above technical solution: When the data set selected for the first time cannot meet the requirements, reselect the data set for iterative analysis to ensure that each independent data frame can match the corresponding data attribute.

[0092] Embodiment 4

[0093] Based on Embodiment 1, in the real-time data classification and hierarchical encryption system based on deep learning, the data classification module includes:

[0094] An attribute processing unit, which is used to enhance the data attributes respectively, determine several enhancement items of the real-time data according to the result difference before and after enhancement, find the relevant sub-data corresponding to each enhancement item in the real-time data, assign a value to the enhancement item according to the data value corresponding to the relevant sub-data, and generate a positive data representation of the real-time data;

[0095] A condition generation unit, which is used to obtain the in-set features and in-set data public information of each data set respectively, construct a representation collection condition corresponding to the data set according to the in-set features, and construct an interval collection condition corresponding to the data set according to the in-set data public information;

[0096] A classification execution unit, which is used to match the representation collection condition with the positive data representation, match the interval collection condition with the data specification of the real-time data, obtain a data set that matches the real-time data, and classify the real-time data according to the set category of the data set.

[0097] In this example, the enhancement process means enhancing the data attributes and strengthening the prominent key attributes among them;

[0098] In this example, the enhancement item means the enhanced data sub-item;

[0099] In this example, the in-set features mean the features presented by the data set, and the in-set data public information means the information that all the in-set data in the data set allows public viewing;

[0100] In this example, the representation collection condition means the basic condition that needs to be met when collecting data into this data set;

[0101] In this example, the interval collection condition means the data specification condition that needs to be met when collecting data into this data set;

[0102] In this example, the set category means the public category to which the data set belongs. For example, the set category of data set A is: financial category.

[0103] Working principle and beneficial effects of the above technical solution: In order to achieve effective classification and divide corresponding categories for real-time data, first, the enhancement items of real-time data are determined by enhancing the data attributes. Further, the relevant sub-data of the enhancement items are used to assign values to the enhancement items to determine the positive data representation of the real-time data. Then, the information of each dataset is processed to construct the representation collection conditions and interval collection conditions for each dataset. Furthermore, it is determined which dataset the real-time data conforms to, and corresponding data classifications are assigned to the real-time data according to the dataset category, thus achieving the purpose of classifying real-time data. In this way, data classification can be completed, and the collection conditions of each dataset can be updated in real time, facilitating the classification of multiple real-time data in a short time and improving the intelligence of the system.

[0104] Embodiment 5

[0105] Based on the real-time data classification and hierarchical encryption system of deep learning in Embodiment 4, the condition generation unit includes:

[0106] The first generation subunit is used to respectively obtain the data basic information corresponding to different in-set data in each dataset, construct the strong sub-features corresponding to the dataset according to the same information among the data basic information, obtain the data type corresponding to each in-set data, and construct the weak sub-features corresponding to the dataset according to the different information among the data basic information, and generate the in-set features corresponding to the dataset according to the strong sub-features and weak sub-features;

[0107] The second generation subunit is used to respectively obtain the data public information corresponding to different in-set data in each dataset, determine the data maximum value and data minimum value corresponding to the in-set data according to the data public information, obtain a number of first test numbers greater than the data maximum value and a number of second test numbers less than the data minimum value, and analyze the range of data that can be included in the dataset by using the first test numbers and the second test numbers;

[0108] The third generation subunit is used to construct the representation collection conditions of the dataset according to several feature key points included in the in-set features, and construct the interval collection conditions of the dataset according to the range of data that can be affected.

[0109] In this example, the strong sub-feature represents the feature that must be realized if a data conforms to the dataset, and the weak sub-feature represents the feature that does not have to be realized if a data conforms to the dataset;

[0110] In this example, the first test number represents the test value used to test and explore the maximum data allowed to be included in the dataset, and the second test number represents the test value used to test and explore the minimum data allowed to be included in the dataset for the dataset to run;

[0111] In this example, the data range that can be included represents the range of data values that the data set allows to include.

[0112] The working principle and beneficial effects of the above technical solution: In order to further ensure the effectiveness of the data set collection conditions and avoid the failure of real-time data classification due to errors, the in-set features and the data range that can be included of the data set are determined by analyzing the basic information and data disclosure information of the data set, so as to construct the corresponding characterization collection conditions and interval collection conditions, laying a foundation for subsequent data classification.

[0113] Example 6

[0114] Based on Example 1, the real-time data classification and hierarchical encryption system based on deep learning, the preliminary encryption module includes:

[0115] The ciphertext construction unit is used to determine the invalid encryption method corresponding to the independent data frame according to the data attribute, delete the corresponding invalid ciphertext, determine several groups of valid ciphertexts of the real-time data, perform chaotic processing on the valid ciphertexts, and generate the chaotic ciphertext sequence of the real-time data;

[0116] The encryption training unit is used to select sequence segments from the chaotic ciphertext sequence to obtain the random ciphertext corresponding to each independent data frame, encrypt the corresponding independent data frame with the random ciphertext, cover the corresponding independent data frame with the obtained encrypted data frame, and perform an encryption effect score on the corresponding random ciphertext according to the coverage rate corresponding to each encrypted data frame;

[0117] The encryption execution unit is used to select the target random ciphertext with the highest score for each independent data frame, sort the target random ciphertexts based on the sorting order of the independent data frames to generate the execution ciphertext of the real-time data, select the effective encryption method of the independent data according to the positive data representation, and encrypt the real-time data with the effective encryption method and the execution ciphertext to generate the preliminarily encrypted real-time data.

[0118] In this example, the invalid encryption method means that this encryption method cannot encrypt the independent data frame of this attribute;

[0119] In this example, chaotic processing means the process of performing behavior analysis and linear analysis on the valid ciphertexts and then generating a digital sequence;

[0120] In this example, the random ciphertext is derived from the chaotic ciphertext sequence and is used to encrypt the independent data frame.

[0121] Working principle and beneficial effects of the above technical solution: To ensure the encryption effect, the system encrypts real-time data twice. When performing the first encryption, invalid encryption methods need to be excluded first, and then the remaining valid ciphertexts are processed by chaos to generate corresponding chaotic ciphertext sequences, ensuring the logic between ciphertexts of different weights and avoiding errors during encryption. Then, corresponding segments are selected from the chaotic ciphertext sequences to construct random ciphertexts to encrypt independent data frames and overwrite the original independent data frames. The target random ciphertext is selected by evaluating the encryption effect of the random ciphertext, and then the real-time data is encrypted to obtain the initially encrypted real-time data. In this way, each independent data frame in the real-time data can be encrypted, and the encryption methods used for different independent data frames may not be the same, greatly improving the encryption quality.

[0122] Example 7

[0123] Based on Example 6, the real-time data classification and hierarchical encryption system based on deep learning further includes:

[0124] Determine several encryption defect positions corresponding to the encrypted data frame according to the encryption effect score;

[0125] Locate the encryption defect positions in the encrypted data frame and hide the encryption defect positions.

[0126] Working principle and beneficial effects of the above technical solution: To ensure the security of real-time data and prevent it from leaking before encryption is completed, the encryption defect positions in the encrypted data frame are hidden to avoid serious consequences caused by data loss.

[0127] Example 8

[0128] Based on Example 1, in the real-time data classification and hierarchical encryption system based on deep learning, the deep encryption module includes:

[0129] A method screening unit for determining invalid encryption methods of the real-time data according to the positive data representation, selecting several random encryption methods from the data set except the invalid encryption methods, and determining the historical usage times corresponding to each random encryption method;

[0130] An encryption preparation unit for sorting the random encryption methods in ascending order of the historical usage times, setting corresponding order levels for the random encryption methods, and determining several encryptable data frames corresponding to each random encryption method according to the encryption attributes corresponding to each random encryption method and the data attributes of the independent data frames;

[0131] A local encryption unit is used to select a corresponding random encryption method for each independent data frame according to the order level of the random encryption method corresponding to each independent data frame, and locally encrypt the corresponding independent data frame by using the random encryption method.

[0132] A confidentiality level determination unit is used to determine the confidentiality level of the real-time data according to the data source of the real-time data, add a corresponding confidentiality level label to the encrypted real-time data, and input it into the confidentiality level storage area of the corresponding data set for storage.

[0133] The working principle and beneficial effects of the above technical solution: When performing the second encryption, first randomly select a valid random encryption method for the data set, and at the same time determine the historical usage times of each random encryption method. Then, perform local encryption on the independent data frames according to the encryptable data frames and their order levels of each random encryption method. Finally, determine the confidentiality level of the real-time data and perform corresponding storage. In this way, the effectiveness of the encryption work can be further improved, and the security of the real-time data is guaranteed.

[0134] Embodiment 9

[0135] This embodiment provides a real-time data classification and hierarchical encryption method based on deep learning, as Figure 2 shown, including:

[0136] Step 1: Divide the real-time data into several independent data frames, and identify the data attributes corresponding to each independent data frame.

[0137] Step 2: Construct a positive data representation corresponding to the real-time data according to the data attributes, and divide the real-time data into the corresponding data sets according to the positive data representation.

[0138] Step 3: Encrypt and train the corresponding independent data frames based on the data attributes to construct a random ciphertext sequence, and construct initially encrypted real-time data according to the encryption effect corresponding to each independent data frame.

[0139] Step 4: Select several random encrypted methods included in the data set to perform local encryption on the initially encrypted real-time data for several times, determine the confidentiality level of the real-time data, and save it.

[0140] In this example, an independent data frame represents the result of dividing the real-time data into sub-data;

[0141] In this example, the data attribute represents the attribute presented by the independent data frame;

[0142] In this example, the positive data representation represents the combined expression of the features and attributes of the real-time data;

[0143] In this example, the data set represents a data set used to store real-time data;

[0144] In this example, the initially encrypted real-time data represents the result of encrypting independent data frames in the real-time data using a random ciphertext sequence;

[0145] In this example, local encryption represents the process of encrypting a data part in the initially encrypted real-time data using a random encryption method;

[0146] In this example, the real-time data corresponding to the security level in the same data set is stored in the same area.

[0147] The working principle and beneficial effects of the above technical solution: In order to better classify and encrypt data, the collected real-time data is divided into several independent data frames. By analyzing the data attributes of each independent data frame, a positive data representation of the real-time data is constructed, so that the data set to which the real-time data belongs can be determined, and it is divided into the corresponding data set to complete the data classification work. Then, the data attributes are used to construct a random ciphertext sequence to encrypt and train the independent data frames, generating the corresponding initially encrypted real-time data. In order to ensure the encryption quality and further guarantee the encryption quality of the real-time data, the initially encrypted real-time data is locally encrypted using the random encryption method used for the existing in-set data in the data set. On the one hand, it can perform compensatory encryption for the weakly encrypted parts, and on the other hand, it can perform deep encryption using the encryption method in the data set of the same classification, improving the quality and efficiency of the encryption result, enhancing the overall encryption effect of the real-time data, and achieving the purpose of synchronously generating data and performing complete encryption in a short time.

[0148] Embodiment 10

[0149] Based on Embodiment 9, for the real-time data classification and hierarchical encryption method based on deep learning, Step 2 includes:

[0150] Step 21: Respectively perform enhancement processing on the data attributes, determine several enhancement items of the real-time data according to the result differences before and after enhancement, find the relevant sub-data corresponding to each enhancement item in the real-time data, and assign values to the enhancement items according to the data values corresponding to the relevant sub-data, generating the positive data representation of the real-time data;

[0151] Step 22: Respectively obtain the in-set features and in-set data public information of each data set, construct a representation collection condition corresponding to the data set according to the in-set features, and construct an interval collection condition corresponding to the data set according to the in-set data public information;

[0152] Step 23: Match the characterization collection conditions with the positive data characterization, and match the interval collection conditions with the data specifications of the real-time data to obtain a data set that matches the real-time data, and classify the real-time data according to the set category of the data set.

[0153] In this example, the enhancement process means enhancing the data attributes and strengthening the prominent key attributes among them.

[0154] In this example, the enhanced item means the data sub that has been enhanced.

[0155] In this example, the in-set features mean the features presented by the data set, and the public information of the in-set data means the information that all the in-set data in the data set allows public viewing.

[0156] In this example, the characterization collection conditions mean the basic conditions that need to be met when collecting data into this data set.

[0157] In this example, the interval collection conditions mean the data specification conditions that need to be met when collecting data into this data set.

[0158] In this example, the set category means the public category to which the data set belongs. For example, the set category of data set A is: financial category.

[0159] The working principle and beneficial effects of the above technical solution: In order to achieve effective classification and divide corresponding categories for real-time data, first, determine the enhanced items of the real-time data by enhancing the data attributes, and further use the relevant sub-data of the enhanced items to assign values to the enhanced items to determine the positive data characterization of the real-time data. Then, process the information of each data set to construct the characterization collection conditions and interval collection conditions for each data set, and then judge which data set the real-time data conforms to. According to the set category of this data set, divide the corresponding data classification for the real-time data, achieving the purpose of classifying the real-time data. In this way, data classification can be completed, and the collection conditions of each data set can be updated in real time, facilitating the classification of multiple real-time data in a short time and improving the intelligence of the system.

[0160] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these changes and modifications.

Claims

1. A real-time data classification and hierarchical encryption system based on deep learning, characterized in that, Including: A frame processing module, configured to divide real-time data into a plurality of independent data frames and identify the data attributes corresponding to each of the independent data frames; A data classification module, configured to construct a positive data representation corresponding to the real-time data according to the data attributes, and divide the real-time data into corresponding data sets according to the positive data representation; A preliminary encryption module, configured to construct a random ciphertext sequence based on the data attributes to perform encryption training on the corresponding independent data frames, and construct initially encrypted real-time data according to the encryption effect corresponding to each of the independent data frames; A deep encryption module, configured to select a plurality of random encrypted methods included in the data set to perform a plurality of local encryptions on the initially encrypted real-time data, determine the confidentiality level of the real-time data and save it.

2. The real-time data classification and hierarchical encryption system based on deep learning according to claim 1, wherein The frame processing module includes: A compression processing unit, configured to perform compression encoding on real-time data, screen a plurality of blank characters included in the encoding result, and perform corresponding division on the real-time data according to the distribution characteristics of the blank characters in the encoding result to obtain a plurality of independent data frames; A preliminary screening unit, configured to randomly select a corresponding number of data sets according to the number of frames of the independent data frames, respectively perform similarity analysis on each of the independent data frames and different data sets to construct a similarity matrix of the real-time data, and construct an initial matching data set for each of the independent data frames according to the elements corresponding to the same matrix position in different similarity matrices; An attribute analysis unit, configured to construct a plurality of data matching methods of the real-time data based on the frame numbers of each of the independent data frames in the real-time data and the corresponding plurality of initial matching data sets, obtain the initial matching data attributes corresponding to each of the data sets, and respectively verify the attribute exclusion characteristics corresponding to each of the data matching methods; An attribute determination unit, configured to select a target data matching method with a feature value of 0 for the attribute exclusion characteristics, and determine the data attributes corresponding to each of the independent data frames.

3. The real-time data classification and hierarchical encryption system based on deep learning according to claim 2, wherein, It further includes: An iterative screening unit, configured to, when all the attribute exclusion characteristics are not 0, randomly select a corresponding number of updated data sets according to the number of frames of the independent data frames, and transmit the updated data sets to the preliminary screening unit for corresponding analysis.

4. The real-time data classification and hierarchical encryption system based on deep learning according to claim 1, characterized in that The data classification module includes: An attribute processing unit, configured to respectively perform enhancement processing on the data attributes, determine a plurality of enhancement items of the real-time data according to the result differences before and after enhancement, find the relevant sub-data corresponding to each of the enhancement items in the real-time data, and assign values to the enhancement items according to the data values corresponding to the relevant sub-data to generate a positive data representation of the real-time data; A condition generation unit, configured to respectively obtain the in-set features and in-set data public information of each of the data sets, construct a representation collection condition corresponding to the data set according to the in-set features, and construct an interval collection condition corresponding to the data set according to the in-set data public information; A classification execution unit is used to match the characterization collection condition with the positive data characterization, match the interval collection condition with the data specification of the real-time data, obtain a data set that matches the real-time data, and classify the real-time data according to the set category of the data set.

5. The real-time data classification and hierarchical encryption system based on deep learning according to claim 4, characterized in that, The condition generation unit includes: A first generation subunit is used to respectively obtain the data basic information corresponding to different in-set data in each data set, construct strong sub-features corresponding to the data set according to the same information between the data basic information, obtain the data type corresponding to each in-set data, and combine the different information between the data basic information to construct weak sub-features corresponding to the data set, and generate in-set features corresponding to the data set according to the strong sub-features and weak sub-features; A second generation subunit is used to respectively obtain the data disclosure information corresponding to different in-set data in each data set, determine the data maximum value and data minimum value corresponding to the in-set data according to the data disclosure information, obtain a number of first test numbers greater than the data maximum value and a number of second test numbers less than the data minimum value, and analyze the collectable data range of the data set by using the first test numbers and the second test numbers; A third generation subunit is used to construct the characterization collection condition of the data set according to a number of feature key points included in the in-set features, and construct the interval collection condition of the data set according to the collectable data range.

6. The real-time data classification and hierarchical encryption system based on deep learning according to claim 1, characterized in that, The preliminary encryption module includes: A ciphertext construction unit is used to determine the invalid encryption method corresponding to the independent data frame according to the data attribute, delete the corresponding invalid ciphertext, determine a number of groups of valid ciphertexts of the real-time data, perform chaotic processing on the valid ciphertexts, and generate a chaotic ciphertext sequence of the real-time data; An encryption training unit is used to select sequence segments in the chaotic ciphertext sequence to obtain random ciphertexts corresponding to each independent data frame, encrypt the corresponding independent data frame by using the random ciphertexts, cover the corresponding independent data frame with the obtained encrypted data frames, and perform an encryption effect score on the corresponding random ciphertexts according to the coverage rate corresponding to each encrypted data frame; An encryption execution unit is used to respectively select the target random ciphertext with the highest score for each independent data frame, sort the target random ciphertexts based on the sorting order of the independent data frames to generate an execution ciphertext of the real-time data, select the effective encryption method of the independent data according to the positive data characterization, and encrypt the real-time data by using the effective encryption method and the execution ciphertext to generate a preliminarily encrypted real-time data.

7. The real-time data classification and hierarchical encryption system based on deep learning according to claim 6, characterized in that, It also includes: Determine a number of encryption defect positions corresponding to the encrypted data frame according to the encryption effect score; Locate the encryption defect positions in the encrypted data frame and hide the encryption defect positions.

8. The real-time data classification and hierarchical encryption system based on deep learning according to claim 1, characterized in that, The deep encryption module includes: A method screening unit, configured to determine an invalid encryption method of the real-time data according to the positive data representation, select several random encryption methods in the data set except the invalid encryption method, and determine the historical usage times corresponding to each random encryption method; An encryption preparation unit, configured to sort the random encryption methods in ascending order of the historical usage times, set corresponding order levels for the random encryption methods, and determine several encryptable data frames corresponding to each random encryption method according to the encryption attributes corresponding to each random encryption method and the data attributes of the independent data frames; A local encryption unit, configured to select a corresponding implemented random encryption method for the independent data frame according to the order level of the random encryption method corresponding to each independent data frame, and perform local encryption on the corresponding independent data frame by using the implemented random encryption method; A security level determination unit, configured to determine the security level of the real-time data according to the data source of the real-time data, add a corresponding security level label to the encrypted real-time data, and input it into the security level storage area in the corresponding data set for storage.

9. A real-time data classification and hierarchical encryption method based on deep learning, characterized in that, It includes: Step 1: Divide the real-time data into several independent data frames, and identify the data attributes corresponding to each independent data frame; Step 2: Construct a positive data representation corresponding to the real-time data according to the data attributes, and divide the real-time data into corresponding data sets according to the positive data representation; Step 3: Construct a random ciphertext sequence based on the data attributes to perform encryption training on the corresponding independent data frames, and construct initially encrypted real-time data according to the encryption effect corresponding to each independent data frame; Step 4: Select several random encrypted methods included in the data set to perform several local encryptions on the initially encrypted real-time data, determine the security level of the real-time data, and save it.

10. The real-time data classification and hierarchical encryption method based on deep learning according to claim 9, characterized in that, The said Step 2 includes: Step 21: Respectively perform enhancement processing on the data attributes, determine several enhancement items of the real-time data according to the result differences before and after enhancement, find the relevant sub-data corresponding to each enhancement item in the real-time data, and assign values to the enhancement items according to the data values corresponding to the relevant sub-data to generate the positive data representation of the real-time data; Step 22: Respectively obtain the in-set features and in-set data public information of each data set, construct a representation collection condition corresponding to the data set according to the in-set features, and construct an interval collection condition corresponding to the data set according to the in-set data public information; Step 23: Match the representation collection condition with the positive data representation, match the interval collection condition with the data specification of the real-time data, obtain a data set that matches the real-time data, and classify the real-time data according to the set category of the data set.

Images