Multivariate data structured analysis processing method based on neural network

Abstract

The invention relates to the field of multivariate data structured analysis processing, in particular to a neural network-based multivariate data structured analysis processing method, which comprises the following steps of: establishing data characteristics of real-time multivariate data by utilizing the real-time multivariate data; training based on a neural network according to the data features of the real-time multivariate data to obtain a multivariate data structured correlation analysis model; carrying out discriminant analysis processing by utilizing the multivariate data structured correlation analysis model to obtain a multivariate data structured analysis processing result, abstracting multivariate data into a body for consideration, taking the multivariate data as a whole, respectively determining characteristics for a data source, a structured type and a data capacity, and carrying out analysis to obtain a multivariate data structured correlation analysis result; according to the method, the influence of a single factor on overall analysis processing is avoided, then, after an error is detected, the problem can be quickly positioned, and the accuracy and efficiency of the scheme are greatly improved.

Description

Technical Field

[0001] The present invention relates to the field of multivariate data structured analysis and processing, and in particular to a multivariate data structured analysis and processing method based on a neural network. Background Art

[0002] Multivariate data refers to a collection of data from different sources, formats, and types. These data can include text, images, audio, video, numerical values, and other forms. They are usually used in complex data analysis and decision support systems. After integrating and analyzing multivariate data, more comprehensive and in-depth information insights can be obtained, thereby improving the accuracy and efficiency of decision-making. In the process of integration and acceptance, multivariate data has a high degree of integrity and correlation. Multivariate data with lost content will cause deviations in subsequent use and processing. Therefore, it is crucial to obtain and verify multivariate data in the early stage. Summary of the invention

[0003] In view of the shortcomings of the prior art, the present invention provides a multivariate data structured analysis and processing method based on a neural network. By performing large-scale data analysis and processing through a neural network, an independent and highly correlated multi-level model is established, which is helpful for the structured analysis and verification of multivariate data.

[0004] To achieve the above object, the present invention provides a multi-data structured analysis and processing method based on a neural network, comprising: S1. Using real-time multivariate data to establish data features of real-time multivariate data; S2. Training a multivariate data structured association analysis model based on a neural network according to the data features of the real-time multivariate data; S3. Using the multivariate data structured association analysis model to perform discriminant analysis processing to obtain multivariate data structured analysis processing results.

[0005] Preferably, the data features of the real-time multivariate data established by using the real-time multivariate data include: S1-1, collect real-time multivariate data; S1-2, acquiring corresponding data sources and link nodes as data source features according to the real-time multivariate data; S1-3, obtaining the data volume corresponding to the real-time multivariate data as a data volume feature according to the real-time multivariate data; S1-4, obtaining the data type corresponding to the real-time multivariate data as a data class feature; S1-5. Utilize the data source characteristics, data volume characteristics and data category characteristics as data characteristics of real-time multivariate data.

[0006] Furthermore, the multivariate data structured association analysis model is obtained by training based on a neural network according to the data characteristics of the real-time multivariate data, including: S2-1, using the data features of the real-time multivariate data to train based on a neural network to obtain a multivariate data structured extraction and screening model; S2-2, using the data features of the real-time multivariate data to perform training based on a neural network to obtain a multivariate data structured association classification model; S2-3, obtaining a multivariate data structured security assessment model based on a neural network according to the multivariate data structured association classification model; S2-4. Utilize the multivariate data structured extraction and screening model, the multivariate data structured association classification model and the multivariate data structured safety assessment model as a multivariate data structured association analysis model.

[0007] Furthermore, the multivariate data structured extraction and screening model is obtained by training based on a neural network using the data features of the real-time multivariate data, including: S2-1-1, acquiring corresponding historical multivariate data according to the real-time multivariate data to establish a first data set; S2-1-2, obtaining data source characteristics corresponding to the historical multivariate data to establish a second data set; S2-1-3, obtaining the data volume characteristics corresponding to the historical multivariate data to establish a third data set; S2-1-4, obtaining data class features corresponding to the historical multivariate data to establish a fourth data set; S2-1-5. Using the first data set as input and the second data set as output, establishing a hidden layer according to the number of link nodes, and obtaining a basic data source extraction and screening model based on neural network training; S2-1-6. Using the first data set as input and the third data set as output, training is performed based on a neural network to obtain a basic data volume extraction and screening model; S2-1-7, using the first data set as input and the fourth data set as output, training based on a neural network according to the number of data types to obtain a basic data class extraction and screening model; S2-1-8. Utilize the basic data source extraction and screening model, the basic data volume extraction and screening model and the basic data class extraction and screening model as a multivariate data structured extraction and screening model.

[0008] Furthermore, the multivariate data structured association classification model is obtained by training based on a neural network using the data features of the real-time multivariate data, including: S2-2-1. Using the data characteristics of the real-time multivariate data to correspond to the data volume characteristics and the data class characteristics, a real-time internal mapping is established as a fifth data set; S2-2-2, according to the data features corresponding to the data volume features and the data class features of the real-time multivariate data, obtain the data features corresponding to the historical multivariate data, the data volume features and the data class features, and establish a historical internal mapping as a sixth data set; S2-2-3, using the real-time multivariate data as input, the real-time multivariate data corresponding to the fifth data set as output, and obtaining a first association classification model through training based on a neural network; S2-2-4, using the historical multivariate data as input, the historical multivariate data corresponding to the sixth data set as output, and obtaining a second association classification model through training based on a neural network; S2-2-5, using the historical multivariate data to input a first association classification model to obtain a first association classification test output; S2-2-6, using the real-time multivariate data to input a second association classification model to obtain a second association classification test output; S2-2-7, determine whether the first association classification test output corresponds to the fifth data set, if so, execute S2-2-8, otherwise, update the fifth data set using the data volume feature and data class feature corresponding to the data feature of the non-corresponding historical multivariate data, and return to S2-2-3; S2-2-8. Determine whether the second association classification test output corresponds to the sixth data set. If so, use the first association classification model and the second association classification model to perform model merging processing to obtain a multivariate data structured association classification model. Otherwise, use the data features of the non-corresponding real-time multivariate data to correspond to the data quantity features and data class features to update the sixth data set, and return to S2-2-4.

[0009] Furthermore, the multivariate data structured security assessment model is obtained based on the neural network according to the multivariate data structured association classification model, including: S2-3-1. Establish a seventh data set using the model output of the multivariate data structured association classification model; S2-3-2, acquiring a data source corresponding to the data source characteristics according to the seventh data set to establish an eighth data set; S2-3-3, acquiring link nodes corresponding to data source characteristics according to the seventh data set to establish a ninth data set; S2-3-4, using the seventh data set as input, and the eighth data set corresponding to the seventh data set as output, training based on a neural network to obtain a first safety assessment model; S2-3-5, using the eighth data set as input, and the eighth data set corresponding to the ninth data set as output, training based on a neural network to obtain a second safety assessment model; S2-3-6. Use the first security assessment model and the second security assessment model to establish a chain security assessment model as a multivariate data structured security assessment model.

[0010] Furthermore, the multivariate data structured association analysis model is used to perform discriminant analysis to obtain multivariate data structured analysis processing results including: S3-1, using the multivariate data structured association analysis model to perform internal cross-validation processing to obtain an internal cross-validation result of the multivariate data structured association analysis model; S3-2. Perform discriminant analysis using the internal cross-validation results to obtain multivariate data structured analysis results.

[0011] Furthermore, the internal cross-validation processing is performed using the multivariate data structured association analysis model to obtain the internal cross-validation result of the multivariate data structured association analysis model, including: S3-1-1, obtaining the model output of the multivariate data structured association analysis model corresponding to the multivariate data structured extraction and screening model as the first internal verification output; S3-1-2, obtaining the model output of the multivariate data structured association analysis model corresponding to the multivariate data structured association classification model as the second internal verification output; S3-1-3, obtaining the model output of the multivariate data structured association analysis model corresponding to the multivariate data structured security assessment model as the third internal verification output; S3-1-4, determine whether the third internal verification output corresponds to an existing data source and a link node, if so, execute S3-1-5, otherwise, return to S2-3-1; S3-1-5, determine whether the second internal verification output corresponds to the third internal verification output, if so, execute S3-1-6, otherwise, return to S2-2-1; S3-1-6, determine whether the model output of the basic data source extraction and screening model corresponding to the first internal verification output corresponds to the third internal verification output, if so, execute S3-1-7, otherwise, return to S1-2; S3-1-7. Determine whether the model output of the basic data source extraction and screening model corresponding to the first internal verification output corresponds to the second internal verification output. If so, use the multivariate data structured association analysis model as the internal cross-validation result of the multivariate data structured association analysis model. Otherwise, return to S1-3.

[0012] Furthermore, the discriminant analysis processing is performed using the internal cross-validation results to obtain multivariate data structured analysis processing results including: S3-2-1. Using the data features of the real-time multivariate data to input the internal cross-validation results into the multivariate data structured association analysis model, respectively obtain the multivariate data structured extraction and screening results, the multivariate data structured association classification results and the multivariate data structured security assessment results of the real-time multivariate data; S3-2-2, establishing a link line for real-time multivariate data using link nodes whose data characteristics correspond to data source characteristics of the real-time multivariate data; S3-2-3, judging whether the multivariate data structured extraction and screening results correspond to the link lines of the real-time multivariate data, if so, the multivariate data structured extraction and screening results are normal, and directly executing S3-2-6, otherwise, executing S3-2-4; S3-2-4, determine whether the basic data source extraction and screening result of the multivariate data structured extraction and screening result corresponds to the link line of the real-time multivariate data, if so, directly execute S3-2-5, otherwise, the multivariate data structured extraction and screening result is abnormal, output the basic data source extraction and screening result, and execute S3-2-5; S3-2-5, determine whether the basic data volume extraction and screening result of the multivariate data structured extraction and screening result corresponds to the link line of the real-time multivariate data, if so, the multivariate data structured extraction and screening result is abnormal, and the basic data class extraction and screening result is output; otherwise, the multivariate data structured extraction and screening result is abnormal, and the basic data volume extraction and screening result is output; S3-2-6, judging whether the multivariate data structured extraction and screening result is normal, if so, executing S3-2-7, otherwise, directly executing S3-2-8; S3-2-7, determine whether the multivariate data structured association classification result corresponds to the multivariate data structured extraction and screening result, if so, the multivariate data structured association classification result is normal, and directly execute S3-2-9, otherwise, the multivariate data structured association classification result is abnormal, output the multivariate data structured association classification result, and directly execute S3-2-9; S3-2-8, determining whether the multivariate data structured association classification result corresponds to the link line of the real-time multivariate data, if so, the multivariate data structured association classification result is normal, otherwise, the multivariate data structured association classification result is abnormal, and outputting the multivariate data structured association classification result; S3-2-9, determine whether the multivariate data structured association classification result is normal, if so, execute S3-2-10, otherwise, directly execute S3-2-11; S3-2-10, determine whether the multivariate data structured security assessment result corresponds to the multivariate data structured association classification result, if so, the multivariate data structured security assessment result is normal, and directly execute S3-2-12, otherwise, the multivariate data structured security assessment result is abnormal, output the multivariate data structured security assessment result, and directly execute S3-2-12; S3-2-11, determine whether the multivariate data structured security assessment result corresponds to the link line of the real-time multivariate data, if so, the multivariate data structured security assessment result is normal, otherwise, the multivariate data structured security assessment result is abnormal, and output the multivariate data structured security assessment result; S3-2-12, determine whether the multivariate data structured extraction and screening result is normal, if so, directly execute S3-2-13, otherwise, retain the multivariate data structured extraction and screening result, and execute S3-2-13; S3-2-13, determine whether the multivariate data structured association classification result is normal, if so, directly execute S3-2-14, otherwise, retain the multivariate data structured association classification result and execute S3-2-14; S3-2-14. Determine whether the multivariate data structured security assessment result is normal. If so, the multivariate data structured analysis and processing result is normal; otherwise, the multivariate data structured analysis and processing result is abnormal.

[0013] Compared with the closest prior art, the present invention has the following beneficial effects: With the help of neural networks, we can classify, screen, categorize, verify and analyze large quantities of data, establish multi-category and multi-level order models, and further ensure the reliability and stability of the overall model through iterative verification during model building and cross-screening based on model output after model building. At the same time, we can abstract multivariate data into entities and consider them as a whole. We can establish characteristics for their data sources, structured types and data capacities, and analyze them respectively, thus avoiding the influence of a single factor on the overall analysis and processing. Secondly, when an error is detected, the problem can be quickly located and processed, which greatly improves the accuracy and efficiency of the solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 It is a flowchart of a multi-data structured analysis and processing method based on a neural network provided by the present invention. DETAILED DESCRIPTION

[0015] The specific implementation modes of the present invention will be further described in detail below in conjunction with the accompanying drawings.

[0016] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution 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 part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] Embodiment 1:

[0018] The present invention provides a multi-data structured analysis and processing method based on a neural network, such as Figure 1 As shown, including: S1. Using real-time multivariate data to establish data features of real-time multivariate data; S2. Training a multivariate data structured association analysis model based on a neural network according to the data features of the real-time multivariate data; S3. Using the multivariate data structured association analysis model to perform discriminant analysis processing to obtain multivariate data structured analysis processing results.

[0019] S1 specifically includes: S1-1, collect real-time multivariate data; S1-2, acquiring corresponding data sources and link nodes as data source features according to the real-time multivariate data; S1-3, obtaining the data volume corresponding to the real-time multivariate data as a data volume feature according to the real-time multivariate data; S1-4, obtaining the data type corresponding to the real-time multivariate data as a data class feature; S1-5. Utilize the data source characteristics, data volume characteristics and data category characteristics as data characteristics of real-time multivariate data.

[0020] In this embodiment, a multivariate data structured analysis and processing method based on a neural network is provided, wherein the data source is the source of the multivariate data, and in actual processing, the IP address, data sending object, etc. may be selected for definition. At the same time, the link nodes serve as intermediate nodes of the multivariate data receiving line to provide data transfer and linking functions.

[0021] S2 specifically includes: S2-1, using the data features of the real-time multivariate data to train based on a neural network to obtain a multivariate data structured extraction and screening model; S2-2, using the data features of the real-time multivariate data to perform training based on a neural network to obtain a multivariate data structured association classification model; S2-3, obtaining a multivariate data structured security assessment model based on a neural network according to the multivariate data structured association classification model; S2-4. Utilize the multivariate data structured extraction and screening model, the multivariate data structured association classification model and the multivariate data structured safety assessment model as a multivariate data structured association analysis model.

[0022] S2-1 specifically includes: S2-1-1, acquiring corresponding historical multivariate data according to the real-time multivariate data to establish a first data set; S2-1-2, obtaining data source characteristics corresponding to the historical multivariate data to establish a second data set; S2-1-3, obtaining the data volume characteristics corresponding to the historical multivariate data to establish a third data set; S2-1-4, obtaining data class features corresponding to the historical multivariate data to establish a fourth data set; S2-1-5. Using the first data set as input and the second data set as output, establishing a hidden layer according to the number of link nodes, and obtaining a basic data source extraction and screening model based on neural network training; S2-1-6. Using the first data set as input and the third data set as output, training is performed based on a neural network to obtain a basic data volume extraction and screening model; S2-1-7, using the first data set as input and the fourth data set as output, training based on a neural network according to the number of data types to obtain a basic data class extraction and screening model; S2-1-8. Utilize the basic data source extraction and screening model, the basic data volume extraction and screening model and the basic data class extraction and screening model as a multivariate data structured extraction and screening model.

[0023] In this embodiment, a multivariate data structured analysis and processing method based on a neural network selects a large amount of historical data as a reference to ensure the accuracy and stability of the current model training. However, a large amount of data puts a lot of pressure on model training. Therefore, feature-related data is selected to establish a neural network hidden layer. Different features correspond to different numbers of hidden layers to improve training efficiency and accuracy.

[0024] S2-2 specifically includes: S2-2-1. Using the data characteristics of the real-time multivariate data to correspond to the data volume characteristics and the data class characteristics, a real-time internal mapping is established as a fifth data set; S2-2-2, according to the data features corresponding to the data volume features and the data class features of the real-time multivariate data, obtain the data features corresponding to the historical multivariate data, the data volume features and the data class features, and establish a historical internal mapping as a sixth data set; S2-2-3, using the real-time multivariate data as input, the real-time multivariate data corresponding to the fifth data set as output, and obtaining a first association classification model through training based on a neural network; S2-2-4, using the historical multivariate data as input, the historical multivariate data corresponding to the sixth data set as output, and obtaining a second association classification model through training based on a neural network; S2-2-5, using the historical multivariate data to input a first association classification model to obtain a first association classification test output; S2-2-6, using the real-time multivariate data to input a second association classification model to obtain a second association classification test output; S2-2-7, determine whether the first association classification test output corresponds to the fifth data set, if so, execute S2-2-8, otherwise, update the fifth data set using the data volume feature and data class feature corresponding to the data feature of the non-corresponding historical multivariate data, and return to S2-2-3; S2-2-8. Determine whether the second association classification test output corresponds to the sixth data set. If so, use the first association classification model and the second association classification model to perform model merging processing to obtain a multivariate data structured association classification model. Otherwise, use the data features of the non-corresponding real-time multivariate data to correspond to the data quantity features and data class features to update the sixth data set, and return to S2-2-4.

[0025] S2-3 specifically includes: S2-3-1. Establish a seventh data set using the model output of the multivariate data structured association classification model; S2-3-2, acquiring a data source corresponding to the data source characteristics according to the seventh data set to establish an eighth data set; S2-3-3, acquiring link nodes corresponding to data source characteristics according to the seventh data set to establish a ninth data set; S2-3-4, using the seventh data set as input, and the eighth data set corresponding to the seventh data set as output, training based on a neural network to obtain a first safety assessment model; S2-3-5, using the eighth data set as input, and the eighth data set corresponding to the ninth data set as output, training based on a neural network to obtain a second safety assessment model; S2-3-6. Use the first security assessment model and the second security assessment model to establish a chain security assessment model as a multivariate data structured security assessment model.

[0026] S3 specifically includes: S3-1, using the multivariate data structured association analysis model to perform internal cross-validation processing to obtain an internal cross-validation result of the multivariate data structured association analysis model; S3-2. Perform discriminant analysis using the internal cross-validation results to obtain multivariate data structured analysis results.

[0027] S3-1 specifically includes: S3-1-1, obtaining the model output of the multivariate data structured association analysis model corresponding to the multivariate data structured extraction and screening model as the first internal verification output; S3-1-2, obtaining the model output of the multivariate data structured association analysis model corresponding to the multivariate data structured association classification model as the second internal verification output; S3-1-3, obtaining the model output of the multivariate data structured association analysis model corresponding to the multivariate data structured security assessment model as the third internal verification output; S3-1-4, determine whether the third internal verification output corresponds to an existing data source and a link node, if so, execute S3-1-5, otherwise, return to S2-3-1; S3-1-5, determine whether the second internal verification output corresponds to the third internal verification output, if so, execute S3-1-6, otherwise, return to S2-2-1; S3-1-6, determine whether the model output of the basic data source extraction and screening model corresponding to the first internal verification output corresponds to the third internal verification output, if so, execute S3-1-7, otherwise, return to S1-2; S3-1-7. Determine whether the model output of the basic data source extraction and screening model corresponding to the first internal verification output corresponds to the second internal verification output. If so, use the multivariate data structured association analysis model as the internal cross-validation result of the multivariate data structured association analysis model. Otherwise, return to S1-3.

[0028] S3-2 specifically includes: S3-2-1. Using the data features of the real-time multivariate data to input the internal cross-validation results into the multivariate data structured association analysis model, respectively obtain the multivariate data structured extraction and screening results, the multivariate data structured association classification results and the multivariate data structured security assessment results of the real-time multivariate data; S3-2-2, establishing a link line for real-time multivariate data using link nodes whose data characteristics correspond to data source characteristics of the real-time multivariate data; S3-2-3, judging whether the multivariate data structured extraction and screening results correspond to the link lines of the real-time multivariate data, if so, the multivariate data structured extraction and screening results are normal, and directly executing S3-2-6, otherwise, executing S3-2-4; S3-2-4, determine whether the basic data source extraction and screening result of the multivariate data structured extraction and screening result corresponds to the link line of the real-time multivariate data, if so, directly execute S3-2-5, otherwise, the multivariate data structured extraction and screening result is abnormal, output the basic data source extraction and screening result, and execute S3-2-5; S3-2-5, determine whether the basic data volume extraction and screening result of the multivariate data structured extraction and screening result corresponds to the link line of the real-time multivariate data, if so, the multivariate data structured extraction and screening result is abnormal, and the basic data class extraction and screening result is output; otherwise, the multivariate data structured extraction and screening result is abnormal, and the basic data volume extraction and screening result is output; S3-2-6, judging whether the multivariate data structured extraction and screening result is normal, if so, executing S3-2-7, otherwise, directly executing S3-2-8; S3-2-7, determine whether the multivariate data structured association classification result corresponds to the multivariate data structured extraction and screening result, if so, the multivariate data structured association classification result is normal, and directly execute S3-2-9, otherwise, the multivariate data structured association classification result is abnormal, output the multivariate data structured association classification result, and directly execute S3-2-9; S3-2-8, determining whether the multivariate data structured association classification result corresponds to the link line of the real-time multivariate data, if so, the multivariate data structured association classification result is normal, otherwise, the multivariate data structured association classification result is abnormal, and outputting the multivariate data structured association classification result; S3-2-9, determine whether the multivariate data structured association classification result is normal, if so, execute S3-2-10, otherwise, directly execute S3-2-11; S3-2-10, determine whether the multivariate data structured security assessment result corresponds to the multivariate data structured association classification result, if so, the multivariate data structured security assessment result is normal, and directly execute S3-2-12, otherwise, the multivariate data structured security assessment result is abnormal, output the multivariate data structured security assessment result, and directly execute S3-2-12; S3-2-11, determine whether the multivariate data structured security assessment result corresponds to the link line of the real-time multivariate data, if so, the multivariate data structured security assessment result is normal, otherwise, the multivariate data structured security assessment result is abnormal, and output the multivariate data structured security assessment result; S3-2-12, determine whether the multivariate data structured extraction and screening result is normal, if so, directly execute S3-2-13, otherwise, retain the multivariate data structured extraction and screening result, and execute S3-2-13; S3-2-13, determine whether the multivariate data structured association classification result is normal, if so, directly execute S3-2-14, otherwise, retain the multivariate data structured association classification result and execute S3-2-14; S3-2-14. Determine whether the multivariate data structured security assessment result is normal. If so, the multivariate data structured analysis and processing result is normal; otherwise, the multivariate data structured analysis and processing result is abnormal.

[0029] In this embodiment, a multi-data structured analysis and processing method based on a neural network can consider the link lines and link nodes as a whole when deployed in actual applications, establish a virtual distributed structure, fill in the data flow direction, quantity, type, etc., and then perform the current level-by-level verification operation.

[0030] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0031] The present invention is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0032] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0033] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0034] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A multivariate data structured analysis and processing method based on a neural network, characterized in that: include: S1. Using real-time multivariate data to establish data features of real-time multivariate data; S2. Training a multivariate data structured association analysis model based on a neural network according to the data features of the real-time multivariate data; S3. Using the multivariate data structured association analysis model to perform discriminant analysis processing to obtain multivariate data structured analysis processing results.

2. A multivariate data structured analysis and processing method based on a neural network as claimed in claim 1, characterized in that: The data features of the real-time multivariate data established by using the real-time multivariate data include: S1-1, collect real-time multivariate data; S1-2, acquiring corresponding data sources and link nodes as data source features according to the real-time multivariate data; S1-3, obtaining the data volume corresponding to the real-time multivariate data as a data volume feature according to the real-time multivariate data; S1-4, obtaining the data type corresponding to the real-time multivariate data as a data class feature; S1-5. Utilize the data source characteristics, data volume characteristics and data category characteristics as data characteristics of real-time multivariate data.

3. A multivariate data structured analysis and processing method based on a neural network as claimed in claim 2, characterized in that: The multivariate data structured association analysis model is obtained by training based on a neural network according to the data characteristics of the real-time multivariate data, including: S2-1, using the data features of the real-time multivariate data to train based on a neural network to obtain a multivariate data structured extraction and screening model; S2-2, using the data features of the real-time multivariate data to perform training based on a neural network to obtain a multivariate data structured association classification model; S2-3, obtaining a multivariate data structured security assessment model based on a neural network according to the multivariate data structured association classification model; S2-4. Utilize the multivariate data structured extraction and screening model, the multivariate data structured association classification model and the multivariate data structured safety assessment model as a multivariate data structured association analysis model.

4. A multivariate data structured analysis and processing method based on a neural network as claimed in claim 3, characterized in that: The multivariate data structured extraction and screening model is obtained by training based on a neural network using the data features of the real-time multivariate data, including: S2-1-1, acquiring corresponding historical multivariate data according to the real-time multivariate data to establish a first data set; S2-1-2, obtaining data source characteristics corresponding to the historical multivariate data to establish a second data set; S2-1-3, obtaining the data volume characteristics corresponding to the historical multivariate data to establish a third data set; S2-1-4, obtaining data class features corresponding to the historical multivariate data to establish a fourth data set; S2-1-5. Using the first data set as input and the second data set as output, establishing a hidden layer according to the number of link nodes, and obtaining a basic data source extraction and screening model based on neural network training; S2-1-6. Using the first data set as input and the third data set as output, training is performed based on a neural network to obtain a basic data volume extraction and screening model; S2-1-7, using the first data set as input and the fourth data set as output, training based on a neural network according to the number of data types to obtain a basic data class extraction and screening model; S2-1-8. Utilize the basic data source extraction and screening model, the basic data volume extraction and screening model and the basic data class extraction and screening model as a multivariate data structured extraction and screening model.

5. The multivariate data structured analysis and processing method based on a neural network as claimed in claim 3, characterized in that: The multivariate data structured association classification model is obtained by training based on a neural network using the data features of the real-time multivariate data, including: S2-2-1. Using the data characteristics of the real-time multivariate data to correspond to the data volume characteristics and the data class characteristics, a real-time internal mapping is established as a fifth data set; S2-2-2, according to the data features corresponding to the data volume features and the data class features of the real-time multivariate data, obtain the data features corresponding to the historical multivariate data, the data volume features and the data class features, and establish a historical internal mapping as a sixth data set; S2-2-3, using the real-time multivariate data as input, the real-time multivariate data corresponding to the fifth data set as output, and obtaining a first association classification model through training based on a neural network; S2-2-4, using the historical multivariate data as input, the historical multivariate data corresponding to the sixth data set as output, and obtaining a second association classification model through training based on a neural network; S2-2-5, using the historical multivariate data to input a first association classification model to obtain a first association classification test output; S2-2-6, using the real-time multivariate data to input a second association classification model to obtain a second association classification test output; S2-2-7, determine whether the first association classification test output corresponds to the fifth data set, if so, execute S2-2-8, otherwise, update the fifth data set using the data volume feature and data class feature corresponding to the data feature of the non-corresponding historical multivariate data, and return to S2-2-3; S2-2-8. Determine whether the second association classification test output corresponds to the sixth data set. If so, use the first association classification model and the second association classification model to perform model merging processing to obtain a multivariate data structured association classification model. Otherwise, use the data features of the non-corresponding real-time multivariate data to correspond to the data quantity features and data class features to update the sixth data set, and return to S2-2-4.

6. A multivariate data structured analysis and processing method based on a neural network as claimed in claim 5, characterized in that: The multivariate data structured security assessment model obtained based on the neural network according to the multivariate data structured association classification model includes: S2-3-1. Establish a seventh data set using the model output of the multivariate data structured association classification model; S2-3-2, acquiring a data source corresponding to the data source characteristics according to the seventh data set to establish an eighth data set; S2-3-3, acquiring link nodes corresponding to data source characteristics according to the seventh data set to establish a ninth data set; S2-3-4, using the seventh data set as input, and the eighth data set corresponding to the seventh data set as output, training based on a neural network to obtain a first safety assessment model; S2-3-5, using the eighth data set as input, and the eighth data set corresponding to the ninth data set as output, training based on a neural network to obtain a second safety assessment model; S2-3-6. Use the first security assessment model and the second security assessment model to establish a chain security assessment model as a multivariate data structured security assessment model.

7. A multivariate data structured analysis and processing method based on a neural network as claimed in claim 6, characterized in that: The multivariate data structured association analysis model is used to perform discriminant analysis to obtain multivariate data structured analysis results, including: S3-1, using the multivariate data structured association analysis model to perform internal cross-validation processing to obtain an internal cross-validation result of the multivariate data structured association analysis model; S3-2. Perform discriminant analysis using the internal cross-validation results to obtain multivariate data structured analysis results.

8. The multivariate data structured analysis and processing method based on a neural network as claimed in claim 7, characterized in that: The internal cross-validation results of the multivariate data structured association analysis model obtained by performing internal cross-validation processing on the multivariate data structured association analysis model include: S3-1-1, obtaining the model output of the multivariate data structured association analysis model corresponding to the multivariate data structured extraction and screening model as the first internal verification output; S3-1-2, obtaining the model output of the multivariate data structured association analysis model corresponding to the multivariate data structured association classification model as the second internal verification output; S3-1-3, obtaining the model output of the multivariate data structured association analysis model corresponding to the multivariate data structured security assessment model as the third internal verification output; S3-1-4, determine whether the third internal verification output corresponds to an existing data source and a link node, if so, execute S3-1-5, otherwise, return to S2-3-1; S3-1-5, determine whether the second internal verification output corresponds to the third internal verification output, if so, execute S3-1-6, otherwise, return to S2-2-1; S3-1-6, determine whether the model output of the basic data source extraction and screening model corresponding to the first internal verification output corresponds to the third internal verification output, if so, execute S3-1-7, otherwise, return to S1-2; S3-1-7. Determine whether the model output of the basic data source extraction and screening model corresponding to the first internal verification output corresponds to the second internal verification output. If so, use the multivariate data structured association analysis model as the internal cross-validation result of the multivariate data structured association analysis model. Otherwise, return to S1-3.

9. A multivariate data structured analysis and processing method based on a neural network as claimed in claim 8, characterized in that: The discriminant analysis processing using the internal cross-validation results obtains multivariate data structured analysis processing results including: S3-2-1. Using the data features of the real-time multivariate data to input the internal cross-validation results into the multivariate data structured association analysis model, respectively obtain the multivariate data structured extraction and screening results, the multivariate data structured association classification results and the multivariate data structured security assessment results of the real-time multivariate data; S3-2-2, establishing a link line for real-time multivariate data using link nodes whose data characteristics correspond to data source characteristics of the real-time multivariate data; S3-2-3, judging whether the multivariate data structured extraction and screening results correspond to the link lines of the real-time multivariate data, if so, the multivariate data structured extraction and screening results are normal, and directly executing S3-2-6, otherwise, executing S3-2-4; S3-2-4, determine whether the basic data source extraction and screening result of the multivariate data structured extraction and screening result corresponds to the link line of the real-time multivariate data, if so, directly execute S3-2-5, otherwise, the multivariate data structured extraction and screening result is abnormal, output the basic data source extraction and screening result, and execute S3-2-5; S3-2-5, determine whether the basic data volume extraction and screening result of the multivariate data structured extraction and screening result corresponds to the link line of the real-time multivariate data, if so, the multivariate data structured extraction and screening result is abnormal, and the basic data class extraction and screening result is output; otherwise, the multivariate data structured extraction and screening result is abnormal, and the basic data volume extraction and screening result is output; S3-2-6, judging whether the multivariate data structured extraction and screening result is normal, if so, executing S3-2-7, otherwise, directly executing S3-2-8; S3-2-7, determine whether the multivariate data structured association classification result corresponds to the multivariate data structured extraction and screening result, if so, the multivariate data structured association classification result is normal, and directly execute S3-2-9, otherwise, the multivariate data structured association classification result is abnormal, output the multivariate data structured association classification result, and directly execute S3-2-9; S3-2-8, determining whether the multivariate data structured association classification result corresponds to the link line of the real-time multivariate data, if so, the multivariate data structured association classification result is normal, otherwise, the multivariate data structured association classification result is abnormal, and outputting the multivariate data structured association classification result; S3-2-9, determine whether the multivariate data structured association classification result is normal, if so, execute S3-2-10, otherwise, directly execute S3-2-11; S3-2-10, determine whether the multivariate data structured security assessment result corresponds to the multivariate data structured association classification result, if so, the multivariate data structured security assessment result is normal, and directly execute S3-2-12, otherwise, the multivariate data structured security assessment result is abnormal, output the multivariate data structured security assessment result, and directly execute S3-2-12; S3-2-11, determine whether the multivariate data structured security assessment result corresponds to the link line of the real-time multivariate data, if so, the multivariate data structured security assessment result is normal, otherwise, the multivariate data structured security assessment result is abnormal, and output the multivariate data structured security assessment result; S3-2-12, determine whether the multivariate data structured extraction and screening result is normal, if so, directly execute S3-2-13, otherwise, retain the multivariate data structured extraction and screening result, and execute S3-2-13; S3-2-13, determine whether the multivariate data structured association classification result is normal, if so, directly execute S3-2-14, otherwise, retain the multivariate data structured association classification result and execute S3-2-14; S3-2-14. Determine whether the multivariate data structured security assessment result is normal. If so, the multivariate data structured analysis and processing result is normal; otherwise, the multivariate data structured analysis and processing result is abnormal.

Images