Data security storage system and method

Abstract

The present invention relates to the field of data storage, and in particular to a data security storage system and method. The system comprises: an evaluation module, which is used to evaluate the risk level of data to be stored within a preset time, and divide the data to be stored into risk data and security data according to the risk level; a storage module, which is used to store risk data and security data; a collection module, which is used to collect storage time and adjustment time; a learning module, which is used to perform time preprocessing on the storage time and adjustment time, generate corresponding preprocessing data, select a number of indicator features to learn the preprocessing data, generate a storage time change graph and an adjustment time change graph, and a scheduling module, which is used to adjust the storage module according to the storage time change graph and the adjustment time change graph. Through machine learning, the accuracy of data risk level evaluation can be effectively improved, the storage space of data is maximized, the security of information storage is improved, and the leakage of important materials is prevented.

Description

Technical Field

[0001] The present invention relates to the field of data storage, and in particular to a data security storage system and method. Background Art

[0002] Data security storage system is a comprehensive data protection solution designed to ensure data integrity, confidentiality and availability.

[0003] Data security storage systems usually include multiple levels of protection measures such as hardware, software, and network. These measures work together to prevent unauthorized access, avoid data loss or damage, and ensure the security of data during storage and transmission.

[0004] China Patent Authorization Announcement No.: CN214507281U discloses a data security storage system, including a cabinet, a server, a controller and a hardware firewall. The cabinet is composed of multiple side panels and multiple columns. Multiple support plates are arranged in the middle of the cabinet from top to bottom. The server, controller and hardware firewall are arranged on the support plates. The side of the columns on the left and right sides of the front of the cabinet are arranged from top to bottom in sequence with multiple slides parallel to the ground. An interface board is arranged between the slides on both sides of the same height. The interface board is slidably arranged in the slide. A push rod motor is also arranged in the slide. The push rod motor is used to drive the interface board to move in the slide. A data interface is arranged on the front of the interface board. An extension line with a data connector is arranged on the back of the interface board. The extension line is connected to the data interface. The data connector is used to insert the corresponding interface of the server. The hardware firewall and the server are in the same network. The controller is connected to the hardware firewall and the push rod motor signal respectively. The system can protect data security from a physical level.

[0005] Chinese Patent Authorization Announcement No.: CN115859394B discloses a data security storage system and method based on edge computing. The data security storage method includes: scanning the edge computing node of the edge computing device to obtain the production equipment corresponding to the edge computing node; monitoring the number of production equipment corresponding to each edge computing node and the amount of data corresponding to the amount of data generated by the production equipment per unit time to obtain data production process information; and using the data production process information to set the data sending parameters of each edge computing node; the edge computing node sends the data to be stored to its corresponding data storage edge computing node according to the data sending parameters; the storage node corresponding to the edge computing node stores the data with the same marking code in the same storage unit according to the marking code of the production equipment. The system includes modules corresponding to the steps of the method.

[0006] However, the above method has the following problems: the data is not confidential, and the data will be destroyed when the memory card is inserted and removed, which affects the accuracy of data analysis. Summary of the invention

[0007] To this end, the present invention provides a data security storage system and method to overcome the problem in the prior art that data has no confidentiality, data is destroyed when a memory card is inserted or removed, and the accuracy of data analysis is affected.

[0008] To achieve the above object, the present invention provides a data security storage system, comprising:

[0009] An evaluation module, which is used to evaluate the risk level of the data to be stored within a preset time, and classify the data to be stored into risk data and safe data according to the risk level;

[0010] A storage module, connected to the evaluation module, for storing the risk data and the safety data;

[0011] A collection module, connected to the storage module, for collecting storage time and adjusting time;

[0012] A learning module, connected to the acquisition module, for performing duration preprocessing on the storage duration and the adjustment duration, generating corresponding preprocessing data, selecting several indicator features of the preprocessing data, learning the preprocessing data according to the indicator features, and generating a storage duration change graph and an adjustment duration change graph;

[0013] a scheduling module connected to the storage module and the learning module, and configured to adjust the storage module according to the storage duration variation graph and the adjustment duration variation graph;

[0014] Wherein, the storage module is provided with a main system and an auxiliary system, the main system is used to store the risk data, and the auxiliary system is used to store the safety data;

[0015] The storage duration is the time required for the risk data to be stored in the main system;

[0016] The adjustment duration is the time required for adjusting the safety data to risk data in the auxiliary system;

[0017] The learning module is provided with a storage duration learning model and an adjustment duration learning model;

[0018] The preprocessing data includes storing the preprocessing data and adjusting the preprocessing data;

[0019] The storage duration learning model is provided with a storage duration threshold, and the adjustment duration learning model is provided with an adjustment duration threshold;

[0020] The storage time threshold is the maximum value of the time required for the risk data to be stored in the main system;

[0021] The adjustment time threshold is the maximum value of the time required for the safety data to be adjusted in the auxiliary system.

[0022] Furthermore, the evaluation module includes:

[0023] An extractor, used for extracting risk points of the data to be stored;

[0024] A simulator connected to the extractor for inputting the risk point and determining the risk level of the data to be stored;

[0025] a classifier, connected to the simulator, for classifying the data to be stored according to the risk level,

[0026] The risk points include the controllability and / or harmfulness of the data to be stored.

[0027] Furthermore, the acquisition module includes:

[0028] A sensor connected to the main system and the auxiliary system for issuing timing instructions;

[0029] A counter, which is connected to the main system and the auxiliary system, and is used to issue a stop timing instruction;

[0030] A timer is connected to the sensor and the counting device, and is used for receiving the timing instruction and the timing stop instruction, and counting the storage time and the adjustment time.

[0031] Furthermore, the extractor determines the risk point according to the data to be stored, inputs the risk point into a risk determination model pre-built in the simulator, determines the risk level of the data to be stored and outputs it, and the classifier labels the risk data and the safety data.

[0032] Wherein, the risk determination model is generated through training based on a training data set;

[0033] The training data set includes all data to be stored within the preset time period.

[0034] Furthermore, when the sensor senses that the risk data enters the main system and / or the safety data enters the auxiliary system, the timing instruction is issued and the timer starts timing. When the risk data is completed stored and / or the safety data is adjusted to risk data, the counter issues a stop timing instruction, the timer stops timing, and outputs the storage time and / or the adjustment time.

[0035] Furthermore, the learning module performs data cleaning on the storage duration and / or adjustment duration to generate corresponding cleaning data;

[0036] The preprocessor selects several indicator features of the storage duration and / or adjustment duration, performs the duration preprocessing, and generates the storage preprocessing data and the adjustment preprocessing data with a sampling rate of the standard sampling rate.

[0037] Wherein, the indicator characteristics include the duration and / or period of the storage duration and / or adjustment duration;

[0038] The duration preprocessing is to segment the cleaned data according to the standard sampling rate;

[0039] The standard sampling rate is a sampling rate that can be identified by the storage duration learning model and / or the adjustment duration learning model, and for a single sampling, the corresponding standard sampling rate is a single sampling rate.

[0040] Furthermore, the storage duration learning model receives the storage preprocessing data and performs learning, generates the storage duration change graph, and labels the storage duration change graph based on the indicator characteristics. The scheduling module receives the storage duration change graph and compares the storage duration change graph with a storage duration threshold. When the storage duration change graph is less than the storage duration threshold, the storage module is not scheduled.

[0041] Furthermore, when the storage duration change graph is greater than the storage duration threshold, the adjustment duration learning model receives the adjustment preprocessing data and performs learning to generate the adjustment duration change graph, and annotates the adjustment duration change graph based on the indicator characteristics. The scheduling module receives the adjustment duration change graph and compares the adjustment duration change graph with the adjustment duration threshold.

[0042] Furthermore, when the adjustment time change graph is smaller than the adjustment time threshold, the storage module reduces the preset time; and when the adjustment time change graph is larger than the adjustment time threshold, the storage module merges part of the auxiliary system into the main system, and uses the adjusted main system as the new main system, and learns the storage time again until the storage time change graph is smaller than the storage time threshold.

[0043] In another aspect, the present invention provides a data security storage method, wherein the training process of the risk determination model includes:

[0044] The evaluation module iteratively optimizes the risk determination model according to the training data set until the loss value converges to a preset value;

[0045] Testing the current risk determination model using the training data set;

[0046] When the test result meets the preset test result, the parameters of the risk determination model are saved.

[0047] Compared with the prior art, the present invention sets an evaluation module to evaluate the risk level of the data to be stored within a preset time, and divides the data to be stored into risk data and safety data according to the risk level; a storage module to store risk data and safety data; a collection module to collect storage time and adjustment time; a learning module to perform time preprocessing on the storage time and adjustment time to generate corresponding preprocessing data, and selects a number of indicator features to learn the preprocessing data to generate a storage time change graph and an adjustment time change graph; a scheduling module to adjust the storage module according to the storage time change graph and the adjustment time change graph. Through machine learning, the accuracy of data risk level assessment can be effectively improved, the storage space of the data is maximized, the security of information storage is improved, and the leakage of important materials is prevented.

[0048] Furthermore, by setting a risk determination model in the assessment module, the data to be stored is divided into risk data and security data, which facilitates the storage module to classify and store the data to be stored, effectively improving the accuracy of the risk level assessment of the data to be stored and improving the storage efficiency of the data to be stored.

[0049] Furthermore, by setting a timer in the acquisition module, it is convenient to receive timing instructions and stop timing instructions, and to count the storage time and adjustment time, so that the machine learning model can learn the storage time and adjustment time, thereby further optimizing the adjustment of the storage module.

[0050] Furthermore, by performing time preprocessing on the storage time and adjustment time, corresponding preprocessed data is obtained, so that effective features of the preprocessed data can be quickly selected, eliminating errors in the learning results caused by incorrect storage time and adjustment time processing during the learning process, making the results of machine learning more accurate.

[0051] Furthermore, by setting a storage time threshold and an adjustment time threshold, the storage module is adjusted to maximize the use of data storage space, improve the security of information storage, and prevent the leakage of important materials. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 This is a schematic diagram of the structure of a data security storage system according to an embodiment of the present invention;

[0053] Figure 2 is a schematic diagram of the structure of an evaluation module according to an embodiment of the present invention;

[0054] Figure 3 This is a schematic diagram of the structure of the acquisition module according to an embodiment of the present invention;

[0055] Figure 4 The figure is a flow chart of the risk determination model training process according to an embodiment of the present invention. DETAILED DESCRIPTION

[0056] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0057] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0058] It should be noted that, in the description of the present invention, terms such as "up", "down", "left", "right", "inside" and "outside" indicating directions or positional relationships are based on the directions or positional relationships shown in the drawings. This is merely for the convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it cannot be understood as a limitation on the present invention.

[0059] In addition, it should be noted that in the description of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0060] See also Figure 1 As shown, it is a structural schematic diagram of a data security storage system of the present invention, including:

[0061] An evaluation module is used to evaluate the risk level of the data to be stored within a preset time, and to classify the data to be stored into risk data and safe data according to the risk level;

[0062] A storage module, which is connected to the assessment module and is used to store risk data and safety data;

[0063] A collection module, which is connected to the storage module and is used to collect storage time and adjust time;

[0064] A learning module, which is connected to the acquisition module, is used to perform duration preprocessing on the storage duration and the adjustment duration, generate corresponding preprocessing data, select several indicator features of the preprocessing data, learn the preprocessing data according to the indicator features, and generate a storage duration change graph and an adjustment duration change graph;

[0065] A scheduling module, which is connected to the storage module and the learning module, and is used to adjust the storage module according to the storage duration change graph and the adjustment duration change graph;

[0066] The storage module is provided with a main system and an auxiliary system, the main system is used to store risk data, and the auxiliary system is used to store safety data;

[0067] The storage duration is the time required for risk data to be stored in the main system;

[0068] The adjustment time is the time required for the safety data to be adjusted to risk data in the auxiliary system;

[0069] The learning module is provided with a storage duration learning model and an adjustment duration learning model;

[0070] Preprocessing data includes storing preprocessed data and adjusting preprocessed data;

[0071] A storage duration threshold is set in the storage duration learning model, and an adjustment duration threshold is set in the adjustment duration learning model;

[0072] The storage time threshold is the maximum time required for risk data to be stored in the main system;

[0073] The adjustment time threshold is the maximum time required for the safety data to be adjusted in the auxiliary system.

[0074] An evaluation module is set up to evaluate the risk level of the data to be stored within a preset time, and the data to be stored is divided into risk data and security data according to the risk level; a storage module is used to store risk data and security data; a collection module is used to collect storage time and adjustment time; a learning module is used to preprocess the storage time and adjustment time to generate corresponding preprocessed data, and a number of indicator features are selected to learn the preprocessed data to generate a storage time change graph and an adjustment time change graph; a scheduling module is used to adjust the storage module according to the storage time change graph and the adjustment time change graph. Through machine learning, the accuracy of data risk level assessment can be effectively improved, the storage space of the data can be maximized, the security of information storage can be improved, and the leakage of important materials can be prevented.

[0075] See also Figure 2 As shown, it is a schematic diagram of the structure of the evaluation module of the present invention, including:

[0076] An extractor, which is used to extract risk points of the data to be stored;

[0077] A simulator, which is connected to the extractor and is used to input risk points and determine the risk level of the data to be stored;

[0078] A classifier, which is connected to the simulator and is used to classify the data to be stored according to the risk level.

[0079] Among them, risk points include the controllability and / or harmfulness of the data to be stored.

[0080] See also Figure 3 As shown, it is a schematic diagram of the structure of the acquisition module of the present invention, including:

[0081] Sensors, which are connected to the main system and auxiliary system to issue timing instructions;

[0082] A counter, which is connected to the main system and the auxiliary system, and is used to issue a stop timing instruction;

[0083] The timer is connected with the sensor and the counter to receive timing instructions and stop timing instructions, and to count and store time and adjust time.

[0084] Embodiment 1:

[0085] When the sensor senses that risk data has entered the main system, it issues a timing instruction and the timer starts timing from 0s. When the risk data is stored, the counter issues a stop timing instruction and the timer stops timing. If the timer displays 10s, the output storage time is 10s.

[0086] Embodiment 2:

[0087] When the sensor senses that the safety data has entered the auxiliary system, it issues a timing instruction and the timer starts timing from 0s. When the safety data is adjusted to risk data, the counter issues a stop timing instruction and the timer stops timing. The timer displays 8s, and the output adjustment duration is 8s.

[0088] By setting a timer in the acquisition module, it is convenient to receive timing instructions and stop timing instructions, and to count the storage time and adjustment time, so that the machine learning model can learn the storage time and adjustment time, and further optimize the adjustment of the storage module.

[0089] Specifically, the extractor determines the risk points based on the data to be stored, inputs the risk points into the risk determination model pre-built in the simulator, determines the risk level of the data to be stored and outputs it, and the classifier labels the risk data and safety data.

[0090] Among them, the risk determination model is trained and generated based on the training data set;

[0091] The training data set includes all the data to be stored within a preset time.

[0092] Embodiment 3:

[0093] The data to be stored in B is "Today's weather is sunny", and the extractor extracts the risk points of the data to be stored in B as "weather, sunny", inputs the risk points into the risk determination model, determines and outputs the risk level of the data to be stored in B as safe data.

[0094] By setting a risk determination model in the assessment module, the data to be stored is divided into risk data and security data, which makes it easier for the storage module to classify and store the data to be stored, effectively improving the accuracy of risk level assessment of the data to be stored and improving the storage efficiency of the data to be stored.

[0095] Specifically, when the sensor senses that risk data enters the main system and / or safety data enters the auxiliary system, it issues a timing instruction and the timer starts timing. When the risk data is stored and / or the safety data is adjusted to risk data, the counter issues a stop timing instruction, the timer stops timing, and outputs the storage time and / or adjustment time.

[0096] Specifically, the learning module performs data cleaning on the storage duration and / or adjustment duration to generate corresponding cleaned data;

[0097] The preprocessor selects several indicator features of storage duration and / or adjustment duration, performs duration preprocessing, and generates storage preprocessing data and adjustment preprocessing data with a sampling rate of a standard sampling rate.

[0098] Among them, the indicator characteristics include the duration and / or period of storage time and / or adjustment time;

[0099] The duration preprocessing is to segment the cleaned data according to the standard sampling rate;

[0100] The standard sampling rate is a sampling rate that can be recognized by the storage duration learning model and / or the adjustment duration learning model, and for a single sampling, the corresponding standard sampling rate is a single sampling rate.

[0101] Preferably, setting the sampling rate to 10 per second has a better learning effect on the preprocessed data, and for the system described in this application, the corresponding learning effect is optimal.

[0102] By preprocessing the storage time and adjustment time, the corresponding preprocessed data is obtained, so that the effective features of the preprocessed data can be quickly selected, eliminating the errors in the learning results caused by incorrect storage time and adjustment time processing during the learning process, making the machine learning results more accurate.

[0103] Specifically, the storage time learning model receives storage preprocessing data and performs learning, generates a storage time change graph, and labels the storage time change graph based on indicator characteristics. The scheduling module receives the storage time change graph and compares the storage time change graph with the storage time threshold. When the storage time change graph is less than the storage time threshold, the storage module is not scheduled.

[0104] Preferably, the storage time threshold is set to 3.5s-4.5s. The storage security of this system is the highest. Usually, the storage time threshold is set to 4s.

[0105] Embodiment 4:

[0106] Assume that the storage duration threshold is 4s, and the value range in the storage duration change graph is 0.5s-4s. If it is less than the storage duration threshold, the storage module is not scheduled.

[0107] Specifically, when the storage time change graph is greater than the storage time threshold, the adjustment time learning model receives the adjustment preprocessing data and performs learning to generate an adjustment time change graph, and annotates the adjustment time change graph based on the indicator characteristics. The scheduling module receives the adjustment time change graph and compares the adjustment time change graph with the adjustment time threshold.

[0108] Embodiment 5:

[0109] Assume that the storage time threshold is 4s, the value range in the storage time change graph is 0.5s-4.5s, and some values ​​are greater than the storage time threshold. The adjustment time learning model receives the adjustment preprocessing data and performs learning, generates an adjustment time change graph, and compares the adjustment time change graph with the adjustment time threshold.

[0110] Specifically, when the adjustment time change graph is less than the adjustment time threshold, the storage module will reduce the preset time. When the adjustment time change graph is greater than the adjustment time threshold, the storage module will merge part of the auxiliary system into the main system, and use the adjusted main system as the new main system to learn the storage time again until the storage time change graph is less than the storage time threshold.

[0111] Preferably, the adjustment time threshold is set to 5.5s-8.5s. The storage security of this system is the highest. Usually, the adjustment time threshold is set to 6s.

[0112] Embodiment 6:

[0113] Assume that the adjustment time threshold is 6s, the preset time is 5s, and the value range in the adjustment time change diagram is 0.5s-4s. If it is less than the adjustment time threshold, the storage module reduces the preset time to 2.5s.

[0114] Embodiment 7:

[0115] Assume that the adjustment time threshold is 6s, the value range in the adjustment time change diagram is 0.5s-7.5s, and some values ​​are greater than the adjustment time threshold. The storage module merges part of the auxiliary system into the main system and uses the adjusted main system as the new main system.

[0116] In a specific implementation, the preset time adopts a recoverable mechanism.

[0117] Embodiment 8:

[0118] Set the preset time to 6s, reduce the preset time to 4s for the first time, reduce the preset time to 2s for the second time, and reduce the preset time to 0s for the third time, triggering the recoverable mechanism and restoring the preset time to the initial value of 6s.

[0119] By setting storage time thresholds and adjustment time thresholds and adjusting storage modules, the storage space for data can be maximized, the security of information storage can be improved, and the leakage of important materials can be prevented.

[0120] See also Figure 4 As shown, it is a flow chart of the risk determination model training process of the present invention, including:

[0121] Step S1, the evaluation module iteratively optimizes the risk determination model according to the training data set until the loss value converges to a preset value;

[0122] Step S2, testing the current risk determination model through a training data set;

[0123] Step S3: When the test result meets the preset test result, the parameters of the risk determination model are saved.

[0124] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

[0125] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A data security storage system, characterized in that: include: An evaluation module, which is used to evaluate the risk level of the data to be stored within a preset time, and classify the data to be stored into risk data and safe data according to the risk level; A storage module, connected to the evaluation module, for storing the risk data and the safety data; A collection module, connected to the storage module, for collecting storage time and adjusting time; A learning module, which is connected to the acquisition module, and is used to perform duration preprocessing on the storage duration and the adjustment duration, generate corresponding preprocessing data, and select several indicator features of the preprocessing data, and learn the preprocessing data according to the indicator features to generate a storage duration change graph and an adjustment duration change graph, wherein the preprocessing data includes storage preprocessing data and adjustment preprocessing data, and the learning module is provided with a storage duration learning model and an adjustment duration learning model; a scheduling module, which is connected to the storage module and the learning module respectively, and is used to adjust the storage module according to the storage duration variation graph and the adjustment duration variation graph; The storage duration is the time required for the risk data to be stored in the main system; The adjustment time is the time required for the safety data to be adjusted to risk data in the auxiliary system; When the storage duration change graph is greater than the storage duration threshold, the adjustment duration learning model receives the adjustment preprocessing data and performs learning to generate the adjustment duration change graph, and annotates the adjustment duration change graph based on the indicator feature, the scheduling module receives the adjustment duration change graph, and compares the adjustment duration change graph with the adjustment duration threshold; Wherein, when the adjustment time variation graph is less than the adjustment time threshold, the storage module reduces the preset time; when the adjustment time variation graph is greater than the adjustment time threshold, the storage module merges part of the auxiliary system into the main system, and uses the adjusted main system as the new main system, and learns the storage time again until the storage time variation graph is less than the storage time threshold; The storage module is provided with the main system and the auxiliary system, the main system is used to store the risk data, and the auxiliary system is used to store the safety data; The storage duration threshold is set in the storage duration learning model, and the adjustment duration threshold is set in the adjustment duration learning model; The storage time threshold is the maximum value of the time required for the risk data to be stored in the main system; The adjustment time threshold is the maximum value of the time required for the safety data to be adjusted in the auxiliary system; The storage duration learning model receives the storage preprocessing data and performs learning, generates the storage duration change graph, and labels the storage duration change graph based on the indicator characteristics. The scheduling module receives the storage duration change graph and compares the storage duration change graph with a storage duration threshold. When the storage duration change graph is less than the storage duration threshold, the storage module is not scheduled.

2. The data security storage system according to claim 1, characterized in that: The evaluation module includes: An extractor, used for extracting risk points of the data to be stored; A simulator connected to the extractor for inputting the risk point and determining the risk level of the data to be stored; a classifier, connected to the simulator, for classifying the data to be stored according to the risk level; The risk points include the controllability and / or harmfulness of the data to be stored.

3. The data security storage system according to claim 2, characterized in that: The acquisition module comprises: A sensor connected to the main system and the auxiliary system for issuing timing instructions; A counter, which is connected to the main system and the auxiliary system, and is used to issue a stop timing instruction; A timer is connected to the sensor and the counting device, and is used for receiving the timing instruction and the timing stop instruction, and counting the storage time and the adjustment time.

4. The data security storage system according to claim 3, characterized in that: The extractor determines the risk point according to the data to be stored, inputs the risk point into the risk determination model pre-built in the simulator, determines the risk level of the data to be stored and outputs it, and the classifier labels the risk data and the safety data. Wherein, the risk determination model is generated through training based on a training data set; The training data set includes all data to be stored within the preset time period.

5. The data security storage system according to claim 4, characterized in that: When the sensor senses that the risk data enters the main system and / or the safety data enters the auxiliary system, the timing instruction is issued and the timer starts timing. When the risk data is completed stored and / or the safety data is adjusted to risk data, the counter issues a stop timing instruction, the timer stops timing, and outputs the storage time and / or the adjustment time.

6. The data security storage system according to claim 5, characterized in that: The learning module performs data cleaning on the storage duration and / or adjustment duration to generate corresponding cleaning data; The preprocessor selects several indicator features of the storage duration and / or adjustment duration, performs the duration preprocessing, and generates the storage preprocessing data and the adjustment preprocessing data with a sampling rate of the standard sampling rate. Wherein, the indicator characteristics include the duration and / or period of the storage duration and / or adjustment duration; The duration preprocessing is to segment the cleaned data according to the standard sampling rate; The standard sampling rate is a sampling rate that can be identified by the storage duration learning model and / or the adjustment duration learning model, and for a single sampling, the corresponding standard sampling rate is a single sampling rate.

Images