Network traffic security assessment method and system based on big data

By processing network traffic data and network byte data separately and using convolutional gated loop unit model, the problem that traditional methods cannot utilize data characteristics and cope with high-dimensional data is solved, and more efficient network traffic security assessment is achieved, improving Assessment of adaptability and stability.

CN118659934BActive Publication Date: 2024-11-15温州中壹技术研究院有限公司
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411141766.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-20
Publication Date
2024-11-15
Estimated Expiration
2044-08-20

AI Technical Summary

Technical Problem

Traditional network traffic security assessment methods cannot fully utilize the characteristics of different data forms, are difficult to explore deep-level patterns and complex correlation characteristics, and are difficult to cope with large and rapidly evolving network data, affecting the adaptability and reliability of security assessment.

Method used

The network traffic data and network byte data are processed separately, and the network byte data is converted into grayscale image to extract texture features. At the same time, the convolutional gated loop unit model is used to evaluate network traffic security, and automatically learn and adapt to new data modes. and threats.

Benefits of technology

Effectively capturing the data spatial structure and local patterns improves the adaptability and stability of the evaluation, can better cope with the high dimensions and diversity of the data, and improves the accuracy and reliability of security assessments.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118659934B_ABST
    Figure CN118659934B_ABST
Patent Text Reader

Abstract

The present invention discloses a network traffic security assessment method and system based on big data, and the method includes data collection, data preprocessing, constructing security assessment fusion features, building a network traffic security assessment model and network traffic security assessment. The present invention relates to the field of network traffic security assessment technology, and specifically refers to a network traffic security assessment method and system based on big data. The present invention obtains original data through data collection; adopts a data preprocessing method of data labeling, data set segmentation, data encoding and data normalization; adopts a method of processing network traffic data and network byte data separately, converting network byte data into grayscale images and extracting texture features, making full use of features of different data forms; adopts a convolutional gated recurrent unit model for network traffic security assessment, which can effectively mine the time and space information in the data, and improve the adaptability and reliability of the assessment.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of network traffic security assessment, and in particular to a network traffic security assessment method and system based on big data. Background Art

[0002] Network traffic security assessment is a key process that deeply analyzes and evaluates data flows in the network to identify potential security threats and abnormal behaviors, and take necessary defensive measures in a timely manner, thereby protecting information assets from unauthorized access and malicious attacks, while optimizing network performance and meeting compliance requirements.

[0003] However, traditional methods of network traffic security assessment have the problem of treating various data forms as a whole and failing to fully utilize the characteristics of different data forms and finding it difficult to mine deep patterns and complex correlation features in the data. Traditional methods also have the problem of being unable to effectively cope with the high dimensionality and diversity of large amounts of rapidly evolving network data and being unable to adapt to new data patterns and threats, which affects the adaptability and reliability of security assessments. Summary of the invention

[0004] In view of the above situation, in order to overcome the defects of the prior art, the present invention provides a network traffic security assessment method and system based on big data. The traditional network traffic security assessment method has the problem that it faces various data forms and processes them as a whole, cannot fully utilize the characteristics of different data forms, and is difficult to mine deep patterns and complex correlation features in the data. This solution creatively processes network traffic data and network byte data separately, converts the network byte data into grayscale images and extracts texture features, which can intuitively capture the data spatial structure and local patterns and fully utilize the characteristics of different data forms; the traditional network traffic security assessment method has the problem that it is difficult to effectively deal with the high dimensionality and diversity of data when facing a large amount of rapidly evolving network data, and cannot adapt to new data patterns and threats, which affects the adaptability and reliability of security assessment. This solution creatively adopts a convolutional gated recurrent unit model for network traffic security assessment, which can effectively mine the time and space information in the data, and can automatically learn and adapt to new data patterns and threats, thereby improving the adaptability and stability of the assessment.

[0005] The technical solution adopted by the present invention is as follows: The network traffic security assessment method based on big data provided by the present invention comprises the following steps:

[0006] Step S1: data collection;

[0007] Step S2: data preprocessing;

[0008] Step S3: construct security assessment fusion features;

[0009] Step S4: construct a network traffic security assessment model;

[0010] Step S5: Network traffic security assessment.

[0011] Further, in step S1, the data collection is used to collect the original data required for network traffic security assessment, specifically, to obtain the security assessment original data set and the network byte original data set from the network security system through collection;

[0012] The security assessment original data set specifically includes network flow data, address information, port information, protocol type, flow direction and load content; the network flow data specifically includes timestamp and flow value; the network byte original data set specifically includes network byte data.

[0013] Furthermore, in step S2, the data preprocessing is used to preprocess the original data, and specifically includes the following steps:

[0014] Step S21: data labeling, which is used to label the original data, specifically labeling the data of the security assessment original data set as safe or non-safe, and using the data as data labels to obtain a labeled data set;

[0015] Step S22: data set segmentation, which is used to segment the data set, specifically by segmenting the labeled data set to obtain a security assessment data set and a security assessment test set;

[0016] Step S23: data encoding, used to convert the data into digital form, specifically encoding the security assessment data set by a one-hot encoding method to obtain an encoded data set;

[0017] Step S24: data normalization, which is used to normalize the data, specifically, normalizing the encoded data set by a minimum-maximum method to obtain a standardized data set;

[0018] Step S25: preprocessing, specifically, performing data preprocessing through the data labeling, the data set segmentation, the data encoding and the data normalization to obtain the standardized data set and the security assessment test set.

[0019] Furthermore, in step S3, constructing the security assessment fusion feature specifically includes the following steps:

[0020] Step S31: data vectorization, which is used to vectorize the preprocessed data, specifically using a continuous bag of words model to process the standardized data set, converting the data in the standardized data set into a vector form, and obtaining a network traffic feature set;

[0021] Step S32: constructing a grayscale image, specifically converting the data in the network byte original data set into a grayscale image of 128×128 pixels, the steps comprising:

[0022] Step S321: Constructing a grayscale image representation matrix , each element of the grayscale image representation matrix is ​​the grayscale value of the pixel at the corresponding position of the grayscale image;

[0023] Step S322: data standardization, which is used to standardize the network byte data, specifically mapping the network byte data to the range of [0, 255] to obtain a network byte data set;

[0024] Step S323: Generate a grayscale image, specifically using the data in the network byte data set as the grayscale image representation matrix The elements of the grayscale image representation matrix are filled in turn , and generate a grayscale image to obtain a grayscale image set;

[0025] Step S33: texture feature extraction, used to extract texture features of the grayscale image, specifically using a convolutional neural network to process the grayscale image set, extracting texture features in vector form, and obtaining a network byte feature set;

[0026] Step S34: constructing a fusion feature, specifically merging the network traffic feature set and the network traffic feature set to obtain a security assessment fusion feature set.

[0027] Further, in step S4, the network traffic security assessment model is constructed to construct a model required for network traffic security assessment, specifically, a convolutional gated recurrent unit model is constructed and model training and performance verification are performed to obtain a network traffic security assessment model;

[0028] The convolutional gated recurrent unit model is specifically a six-layer structure, wherein the first and second layers are convolutional layers for extracting features, the third and fourth layers are gated recurrent unit layers for processing features and capturing temporal dependencies, the fifth and sixth layers are fully connected layers, the fifth layer is used to reduce the dimension, the sixth layer is used to perform classification output using a softmax function, and a cross entropy loss function is selected as the loss function;

[0029] The construction of the network traffic security assessment model specifically includes the following steps:

[0030] Step S41: Construct a convolutional layer. The formula used is as follows:

[0031] ;

[0032] In the formula, represents the ReLU activation function, represents the input of the first convolutional layer, represents the input of the second convolutional layer, represents the convolution kernel of the first convolutional layer, represents the convolution kernel of the second convolutional layer, represents the convolution operation function, Indicates that the convolution operation is performed on the input of the first convolution layer using the convolution kernel of the first convolution layer. Indicates that the convolution operation is performed on the input of the second convolution layer using the convolution kernel of the second convolution layer. represents the output of the first convolutional layer, Represents the output of the second convolutional layer;

[0033] Step S42: constructing a gated recurrent unit layer, the steps include:

[0034] Step S421: Update gate calculation, the formula used is as follows:

[0035] ;

[0036] In the formula, represents the value of the update gate at time t, represents the sigmoid function, represents the update gate weight matrix, represents the hidden state at time t-1, represents the input at time t;

[0037] Step S422: reset gate calculation, the formula used is as follows:

[0038] ;

[0039] In the formula, represents the value of the reset gate at time t, Represents resetting the gate weight matrix;

[0040] Step S423: Calculate the hidden state using the following formula:

[0041] ;

[0042] In the formula, represents the candidate hidden state at time t, represents the hidden state at time t, represents the weight matrix used to calculate the candidate hidden state;

[0043] Step S43: Construct a fully connected layer. The formula used is as follows:

[0044] ;

[0045] In the formula, represents the output of the fifth fully connected layer, represents the weight matrix of the fifth fully connected layer, represents the input of the fifth fully connected layer, represents the bias term of the fifth fully connected layer, represents the output of the sixth fully connected layer, represents the weight matrix of the sixth fully connected layer, represents the bias term of the sixth fully connected layer, represents the softmax function;

[0046] Step S44: Design a loss function, the formula used is as follows:

[0047] ;

[0048] In the formula, L represents the loss function value, N represents the number of samples, Represents the true category label of the nth sample, with a value of 0 or 1, 1 indicates that the sample category is safe, and 0 indicates that the sample category is non-safe. Indicates the probability of predicting the nth sample category as safe;

[0049] Step S45: constructing a model and training it, specifically, constructing the convolutional gated recurrent unit model by using the constructed convolutional layer, the constructed gated recurrent unit layer, the constructed fully connected layer and the designed loss function, and training the model based on the security assessment fusion feature set, verifying the model performance based on the security assessment test set, and obtaining the network traffic security assessment model.

[0050] Further, in step S5, the network traffic security assessment is specifically to use the network traffic security assessment model to perform a network traffic security assessment on the collected data to obtain security assessment reference data, and evaluate the network traffic security status based on the security assessment reference data.

[0051] The network traffic security assessment system based on big data provided by the present invention comprises a data acquisition module, a data preprocessing module, a security assessment fusion feature construction module, a network traffic security assessment model construction module and a network traffic security assessment module;

[0052] The data acquisition module is used for data acquisition, and obtains a security assessment original data set and a network byte original data set through data acquisition, and sends the security assessment original data set to the data preprocessing module, and sends the network byte original data set to the security assessment fusion feature construction module;

[0053] The data preprocessing module is used for data preprocessing, and obtains a standardized data set and a security assessment test set through data preprocessing, and sends the standardized data set to the security assessment fusion feature construction module, and sends the security assessment test set to the network traffic security assessment model construction module;

[0054] The security assessment fusion feature construction module is used to construct a security assessment fusion feature, obtain a security assessment fusion feature set by constructing the security assessment fusion feature, and send the security assessment fusion feature set to the network traffic security assessment model construction module;

[0055] The network traffic security assessment model construction module is used to construct a network traffic security assessment model, obtain the network traffic security assessment model by constructing a convolutional gated recurrent unit model, and send the network traffic security assessment model to the network traffic security assessment module;

[0056] The network traffic security assessment module is used for network traffic security assessment and obtains security assessment reference data by performing network traffic security assessment.

[0057] The beneficial effects achieved by the present invention using the above scheme are as follows:

[0058] (1) In view of the technical problems that traditional methods of network traffic security assessment face diverse data forms and treat them as a whole, they are unable to fully utilize the characteristics of different data forms and find it difficult to mine deep patterns and complex correlation features in the data. This solution creatively processes network traffic data and network byte data separately, converts network byte data into grayscale images and extracts texture features. It can intuitively capture the spatial structure and local patterns of the data and fully utilize the characteristics of different data forms.

[0059] (2) In view of the technical problems that traditional methods of network traffic security assessment have difficulty in effectively dealing with the high dimensionality and diversity of data when faced with large amounts of rapidly evolving network data, and are unable to adapt to new data patterns and threats, which affects the adaptability and reliability of security assessments, this solution creatively uses a convolutional gated recurrent unit model for network traffic security assessment, which can effectively mine the temporal and spatial information in the data, and can automatically learn and adapt to new data patterns and threats, thereby improving the adaptability and stability of the assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 A schematic diagram of the flow of a network traffic security assessment method based on big data provided by the present invention;

[0061] Figure 2 A schematic diagram of the module flow of the network traffic security assessment system based on big data provided by the present invention;

[0062] Figure 3 This is a schematic diagram of the process of data preprocessing in step S2;

[0063] Figure 4 A flowchart of constructing security assessment fusion features for step S3;

[0064] Figure 5 A flowchart of building a network traffic security assessment model in step S4.

[0065] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention. DETAILED DESCRIPTION

[0066] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0067] In the description of the present invention, it should be understood that terms such as “upper”, “lower”, “front”, “back”, “left”, “right”, “top”, “bottom”, “inside” and “outside” indicating directions or positional relationships are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore should not be understood as limiting the present invention.

[0068] Example 1, see Figure 1 The present invention provides a network traffic security assessment method based on big data, which comprises the following steps:

[0069] Step S1: data collection;

[0070] Step S2: data preprocessing;

[0071] Step S3: construct security assessment fusion features;

[0072] Step S4: construct a network traffic security assessment model;

[0073] Step S5: Network traffic security assessment.

[0074] Example 2, see Figure 1 and Figure 2In step S1, the data collection is used to collect the original data required for network traffic security assessment, specifically, to obtain the security assessment original data set and the network byte original data set from the network security system through collection;

[0075] The security assessment original data set specifically includes network flow data, address information, port information, protocol type, flow direction and load content; the network flow data specifically includes timestamp and flow value; the network byte original data set specifically includes network byte data.

[0076] Example 3, see Figure 1 , Figure 2 and Figure 3 This embodiment is based on the above embodiment. In step S2, the data preprocessing is used to preprocess the original data, and specifically includes the following steps:

[0077] Step S21: data labeling, which is used to label the original data, specifically labeling the data of the security assessment original data set as safe or non-safe, and using the data as data labels to obtain a labeled data set;

[0078] Step S22: data set segmentation, which is used to segment the data set, specifically by segmenting the labeled data set to obtain a security assessment data set and a security assessment test set;

[0079] Step S23: data encoding, used to convert the data into digital form, specifically encoding the security assessment data set by a one-hot encoding method to obtain an encoded data set;

[0080] Step S24: data normalization, which is used to normalize the data, specifically, normalizing the encoded data set by a minimum-maximum method to obtain a standardized data set;

[0081] Step S25: preprocessing, specifically, performing data preprocessing through the data labeling, the data set segmentation, the data encoding and the data normalization to obtain the standardized data set and the security assessment test set.

[0082] By performing the above operations, the traditional motion posture analysis method only simply defines and classifies the motion posture when classifying motion postures, which cannot comprehensively reflect the typical motion postures in daily life and cannot fully capture the technical problem of the complexity of motion postures in the real world. This scheme creatively adopts a method of defining motion postures in detail and classifying them into symmetrical motion postures and asymmetrical motion postures, which can comprehensively reflect the characteristic information of motion postures in daily life, and analyze symmetrical motion postures and asymmetrical motion postures separately, making the analysis more accurate and targeted.

[0083] Example 4, see Figure 1 , Figure 2 and Figure 4 This embodiment is based on the above embodiment. In step S3, the construction of the security assessment fusion feature specifically includes the following steps:

[0084] Step S31: data vectorization, which is used to vectorize the preprocessed data, specifically using a continuous bag of words model to process the standardized data set, converting the data in the standardized data set into a vector form, and obtaining a network traffic feature set;

[0085] Step S32: constructing a grayscale image, specifically converting the data in the network byte original data set into a grayscale image of 128×128 pixels, the steps comprising:

[0086] Step S321: Constructing a grayscale image representation matrix , each element of the grayscale image representation matrix is ​​the grayscale value of the pixel at the corresponding position of the grayscale image;

[0087] Step S322: Data standardization is used to standardize the network byte data, specifically mapping the network byte data to the range of [0, 255] to obtain a network byte data set. The formula used is as follows:

[0088] ;

[0089] In the formula, Represents the normalized value of the k-th byte data, Indicates the kth byte of data, Indicates the minimum value of byte data. Indicates the maximum value of byte data;

[0090] Step S323: Generate a grayscale image, specifically using the data in the network byte data set as the grayscale image representation matrix The elements of the grayscale image representation matrix are filled in turn , and generate a grayscale image to obtain a grayscale image set;

[0091] The grayscale image representation matrix is ​​sequentially filled in , the formula used is as follows:

[0092] ;

[0093] In the formula, Represents the grayscale image representation matrix The element in row i and column j, Represents the normalized value of the 128(i-1)+j byte data;

[0094] Step S33: texture feature extraction, used to extract texture features of the grayscale image, specifically using a convolutional neural network to process the grayscale image set, extracting texture features in vector form, and obtaining a network byte feature set;

[0095] Step S34: constructing a fusion feature, specifically merging the network traffic feature set and the network traffic feature set to obtain a security assessment fusion feature set.

[0096] By performing the above operations, the traditional method of network traffic security assessment faces various data forms and treats them as a whole, which cannot fully utilize the characteristics of different data forms and is difficult to mine deep patterns and complex correlation features in the data. This solution creatively processes network traffic data and network byte data separately, converts network byte data into grayscale images and extracts texture features, which can intuitively capture the spatial structure and local patterns of the data and fully utilize the characteristics of different data forms.

[0097] Example 5, see Figure 1 , Figure 2 and Figure 5 , this embodiment is based on the above embodiment. In step S4, the network traffic security assessment model is constructed to construct the model required for network traffic security assessment, specifically, a convolutional gated recurrent unit model is constructed and model training and performance verification are performed to obtain a network traffic security assessment model;

[0098] The convolutional gated recurrent unit model is specifically a six-layer structure, wherein the first and second layers are convolutional layers for extracting features, the third and fourth layers are gated recurrent unit layers for processing features and capturing temporal dependencies, the fifth and sixth layers are fully connected layers, the fifth layer is used to reduce the dimension, the sixth layer is used to perform classification output using a softmax function, and a cross entropy loss function is selected as the loss function;

[0099] The construction of the network traffic security assessment model specifically includes the following steps:

[0100] Step S41: Construct a convolutional layer. The formula used is as follows:

[0101] ;

[0102] In the formula, represents the ReLU activation function, represents the input of the first convolutional layer, represents the input of the second convolutional layer, represents the convolution kernel of the first convolutional layer, represents the convolution kernel of the second convolutional layer, represents the convolution operation function, Indicates that the convolution operation is performed on the input of the first convolution layer using the convolution kernel of the first convolution layer. Indicates that the convolution operation is performed on the input of the second convolution layer using the convolution kernel of the second convolution layer. represents the output of the first convolutional layer, Represents the output of the second convolutional layer;

[0103] Step S42: constructing a gated recurrent unit layer, the steps include:

[0104] Step S421: Update gate calculation, the formula used is as follows:

[0105] ;

[0106] In the formula, represents the value of the update gate at time t, represents the sigmoid function, represents the update gate weight matrix, represents the hidden state at time t-1, represents the input at time t;

[0107] Step S422: reset gate calculation, the formula used is as follows:

[0108] ;

[0109] In the formula, represents the value of the reset gate at time t, Represents resetting the gate weight matrix;

[0110] Step S423: Calculate the hidden state using the following formula:

[0111] ;

[0112] In the formula, represents the candidate hidden state at time t, represents the hidden state at time t, represents the weight matrix used to calculate the candidate hidden state;

[0113] Step S43: Construct a fully connected layer. The formula used is as follows:

[0114] ;

[0115] In the formula, represents the output of the fifth fully connected layer, represents the weight matrix of the fifth fully connected layer, represents the input of the fifth fully connected layer, represents the bias term of the fifth fully connected layer, represents the output of the sixth fully connected layer, represents the weight matrix of the sixth fully connected layer, represents the bias term of the sixth fully connected layer, represents the softmax function;

[0116] Step S44: Design a loss function, the formula used is as follows:

[0117] ;

[0118] In the formula, L represents the loss function value, N represents the number of samples, Represents the true category label of the nth sample, with a value of 0 or 1, 1 indicates that the sample category is safe, and 0 indicates that the sample category is non-safe. Indicates the probability of predicting the nth sample category as safe;

[0119] Step S45: constructing a model and training it, specifically, constructing the convolutional gated recurrent unit model by using the constructed convolutional layer, the constructed gated recurrent unit layer, the constructed fully connected layer and the designed loss function, and training the model based on the security assessment fusion feature set, verifying the model performance based on the security assessment test set, and obtaining the network traffic security assessment model.

[0120] By performing the above operations, this solution creatively adopts a convolutional gated recurrent unit model for network traffic security assessment, which can effectively mine the temporal and spatial information in the data, and can automatically learn and adapt to new data patterns and threats, thereby improving the adaptability and stability of the assessment. This is to address the technical problems that traditional methods of network traffic security assessment have difficulty in effectively coping with the high dimensionality and diversity of data when faced with large amounts of rapidly evolving network data, and cannot adapt to new data patterns and threats, thus affecting the adaptability and reliability of security assessment.

[0121] Example 6, see Figure 1 and Figure 2 This embodiment is based on the above embodiment. In step S5, the network traffic security assessment is specifically to use the network traffic security assessment model to perform a network traffic security assessment on the collected data to obtain security assessment reference data, and evaluate the network traffic security status based on the security assessment reference data.

[0122] Embodiment 7, see Figure 1 and Figure 2, this embodiment is based on the above embodiment, and the network traffic security assessment system based on big data provided by the present invention includes a data acquisition module, a data preprocessing module, a security assessment fusion feature construction module, a network traffic security assessment model construction module and a network traffic security assessment module;

[0123] The data acquisition module is used for data acquisition, and obtains a security assessment original data set and a network byte original data set through data acquisition, and sends the security assessment original data set to the data preprocessing module, and sends the network byte original data set to the security assessment fusion feature construction module;

[0124] The data preprocessing module is used for data preprocessing, and obtains a standardized data set and a security assessment test set through data preprocessing, and sends the standardized data set to the security assessment fusion feature construction module, and sends the security assessment test set to the network traffic security assessment model construction module;

[0125] The security assessment fusion feature construction module is used to construct a security assessment fusion feature, obtain a security assessment fusion feature set by constructing the security assessment fusion feature, and send the security assessment fusion feature set to the network traffic security assessment model construction module;

[0126] The network traffic security assessment model construction module is used to construct a network traffic security assessment model, obtain the network traffic security assessment model by constructing a convolutional gated recurrent unit model, and send the network traffic security assessment model to the network traffic security assessment module;

[0127] The network traffic security assessment module is used for network traffic security assessment and obtains security assessment reference data by performing network traffic security assessment.

[0128] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0129] While the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that many changes, modifications, substitutions and variations can be made to the embodiments without departing from the principles and spirit of the invention.

[0130] The present invention and its embodiments are described above, and such description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if ordinary technicians in the field are inspired by it, without departing from the purpose of the invention, they can design a structure and embodiment similar to the technical solution without creativity, which should belong to the protection scope of the present invention.

Claims

1. A network traffic security assessment method based on big data, characterized by: The method comprises the following steps: Step S1: data collection, specifically, obtaining a security assessment original data set and a network byte original data set from a network security system through collection, wherein the security assessment original data set specifically includes network flow data, flow direction and load content, the network flow data specifically includes a timestamp and flow value, and the network byte original data set specifically includes network byte data, and the network byte data specifically includes port information, protocol type, address information, data packet header information, byte statistics data and encrypted byte data; Step S2: Data preprocessing: labeling and data set segmentation of the collected data to obtain a security assessment test set; labeling, data set segmentation, data encoding and data normalization to obtain a standardized data set; Step S3: constructing a security assessment fusion feature, specifically, vectorizing the standardized data set by adopting a continuous bag of words model to obtain a network traffic feature set, converting the network byte data into a grayscale image and extracting texture features to obtain a network byte feature set, and merging the network traffic feature set and the network byte feature set to obtain a security assessment fusion feature set; Step S4: constructing a network traffic security assessment model, specifically constructing a convolutional gated recurrent unit model and performing model training and performance verification to obtain a network traffic security assessment model; In step S4, the network traffic security assessment model is constructed to construct a model required for network traffic security assessment, specifically, a convolutional gated recurrent unit model is constructed and model training and performance verification are performed to obtain a network traffic security assessment model; The convolutional gated recurrent unit model is specifically a six-layer structure, wherein the first and second layers are convolutional layers for extracting features, the third and fourth layers are gated recurrent unit layers for processing features and capturing temporal dependencies, the fifth and sixth layers are fully connected layers, the fifth layer is used to reduce the dimension, the sixth layer is used to perform classification output using a softmax function, and a cross entropy loss function is selected as the loss function; The construction of the network traffic security assessment model specifically includes the following steps: Step S41: Construct a convolutional layer. The formula used is as follows: ; In the formula, represents the ReLU activation function, represents the input of the first convolutional layer, represents the input of the second convolutional layer, represents the convolution kernel of the first convolutional layer, represents the convolution kernel of the second convolutional layer, represents the convolution operation function, Indicates that the convolution operation is performed on the input of the first convolution layer using the convolution kernel of the first convolution layer. Indicates that the convolution operation is performed on the input of the second convolution layer using the convolution kernel of the second convolution layer. represents the output of the first convolutional layer, Represents the output of the second convolutional layer; Step S42: constructing a gated recurrent unit layer, the steps include: Step S421: Update gate calculation, the formula used is as follows: ; In the formula, represents the value of the update gate at time t, represents the sigmoid function, represents the update gate weight matrix, represents the hidden state at time t-1, represents the input at time t; Step S422: reset gate calculation, the formula used is as follows: ; In the formula, represents the value of the reset gate at time t, Represents resetting the gate weight matrix; Step S423: Calculate the hidden state using the following formula: ; In the formula, represents the candidate hidden state at time t, represents the hidden state at time t, represents the weight matrix used to calculate the candidate hidden state; Step S43: Construct a fully connected layer. The formula used is as follows: ; In the formula, represents the output of the fifth fully connected layer, represents the weight matrix of the fifth fully connected layer, represents the input of the fifth fully connected layer, represents the bias term of the fifth fully connected layer, represents the output of the sixth fully connected layer, represents the weight matrix of the sixth fully connected layer, represents the bias term of the sixth fully connected layer, represents the softmax function; Step S44: Design a loss function, the formula used is as follows: ; In the formula, L represents the loss function value, N represents the number of samples, Represents the true category label of the nth sample, with a value of 0 or 1, 1 indicates that the sample category is safe, and 0 indicates that the sample category is non-safe. Indicates the probability of predicting the nth sample category as safe; Step S45: constructing a model and training it, specifically, using the constructed convolutional layer, the constructed gated recurrent unit layer, the constructed fully connected layer and the designed loss function to construct the convolutional gated recurrent unit model, and training the model based on the security assessment fusion feature set, verifying the model performance based on the security assessment test set, and obtaining the network traffic security assessment model; Step S5: network traffic security assessment, specifically, using the network traffic security assessment model to perform network traffic security assessment on the collected data to obtain security assessment reference data.

2. The network traffic security assessment method based on big data according to claim 1 is characterized by: In step S3, the construction of the security assessment fusion feature specifically includes the following steps: Step S31: data vectorization, which is used to vectorize the preprocessed data, specifically using a continuous bag of words model to process the standardized data set, converting the data in the standardized data set into a vector form, and obtaining a network traffic feature set; Step S32: constructing a grayscale image, specifically converting the data in the network byte original data set into a grayscale image of 128×128 pixels, the steps comprising: Step S321: Constructing a grayscale image representation matrix , each element of the grayscale image representation matrix is ​​the grayscale value of the pixel at the corresponding position of the grayscale image; Step S322: data standardization, which is used to standardize the network byte data, specifically mapping the network byte data to the range of [0, 255] to obtain a network byte data set; Step S323: Generate a grayscale image, specifically using the data in the network byte data set as the grayscale image representation matrix The elements of the grayscale image representation matrix are filled in turn , and generate a grayscale image to obtain a grayscale image set; Step S33: texture feature extraction, used to extract texture features of the grayscale image, specifically using a convolutional neural network to process the grayscale image set, extracting texture features in vector form, and obtaining a network byte feature set; Step S34: constructing fusion features, specifically, obtaining a network traffic feature set by vectorizing the data, obtaining a network byte feature set by constructing a grayscale image and extracting texture features, and merging the network traffic feature set and the network byte feature set to obtain a security assessment fusion feature set.

3. The network traffic security assessment method based on big data according to claim 1 is characterized by: In step S2, the data preprocessing is used to preprocess the original data, and specifically includes the following steps: Step S21: data labeling, which is used to label the original data, specifically labeling the data of the security assessment original data set as safe or non-safe, and using the data as data labels to obtain a labeled data set; Step S22: data set segmentation, which is used to segment the data set, specifically by segmenting the labeled data set to obtain a security assessment data set and a security assessment test set; Step S23: data encoding, used to convert the data into digital form, specifically encoding the security assessment data set by a one-hot encoding method to obtain an encoded data set; Step S24: data normalization, which is used to normalize the data, specifically, normalizing the encoded data set by a minimum-maximum method to obtain a standardized data set; Step S25: preprocessing, specifically obtaining the security assessment test set through the data labeling and the data set segmentation, and obtaining the standardized data set through the data labeling, the data set segmentation, the data encoding and the data normalization.

4. The network traffic security assessment method based on big data according to claim 1 is characterized by: In step S5, the network traffic security assessment specifically adopts the network traffic security assessment model to perform a network traffic security assessment on the collected data to obtain security assessment reference data, and assesses the network traffic security status based on the security assessment reference data.

5. A network traffic security assessment system based on big data, used to implement the network traffic security assessment method based on big data as described in any one of claims 1 to 4, characterized in that: It includes data collection module, data preprocessing module, security assessment fusion feature construction module, network traffic security assessment model construction module and network traffic security assessment module.

6. The network traffic security assessment system based on big data according to claim 5 is characterized in that: The data acquisition module is used for data acquisition, and obtains a security assessment original data set and a network byte original data set through data acquisition, and sends the security assessment original data set to the data preprocessing module, and sends the network byte original data set to the security assessment fusion feature construction module; The data preprocessing module is used for data preprocessing, and obtains a standardized data set and a security assessment test set through data preprocessing, and sends the standardized data set to the security assessment fusion feature construction module, and sends the security assessment test set to the network traffic security assessment model construction module; The security assessment fusion feature construction module is used to construct a security assessment fusion feature, obtain a security assessment fusion feature set by constructing the security assessment fusion feature, and send the security assessment fusion feature set to the network traffic security assessment model construction module; The network traffic security assessment model construction module is used to construct a network traffic security assessment model, obtain the network traffic security assessment model by constructing a convolutional gated recurrent unit model, and send the network traffic security assessment model to the network traffic security assessment module; The network traffic security assessment module is used for network traffic security assessment and obtains security assessment reference data by performing network traffic security assessment.