A method, apparatus and system for evidence generation based on multivariate collaborative analysis

By using multivariate collaborative analysis algorithms to filter and process data from distributed systems, efficient and standardized evidence formats are generated. This solves the problems of low evidence generation efficiency and missing information in existing technologies, and improves the accuracy of evidence generation and the ability to support downstream tasks.

CN118277411BActive Publication Date: 2026-11-10HUAZHONG UNIV OF SCI & TECH
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202410313079.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-03-19
Publication Date
2026-11-10
Estimated Expiration
2044-03-19

AI Technical Summary

Technical Problem

In the context of big data, existing evidence generation methods are inefficient, lack key information, have weak generalization ability, and the completeness of the generated evidence information is difficult to guarantee, thus failing to meet the needs of downstream tasks.

Method used

A multivariate collaborative analysis algorithm is used to obtain the data storage structure from a distributed search engine. Data entries are filtered through three-dimensional time rules and IP rules. Correlation analysis is performed by combining grey relational analysis and Pearson correlation coefficient analysis algorithms. Key fields are expanded and standardized to generate a unified evidence format.

Benefits of technology

It improves the efficiency and accuracy of evidence generation, reduces the false alarm rate of the algorithm, realizes efficient screening and standardization of evidence, and supports downstream tasks such as evidence correlation analysis and attack tracing.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118277411B_ABST
    Figure CN118277411B_ABST
Patent Text Reader

Abstract

This invention discloses a method, apparatus, and system for evidence generation based on multivariate collaborative analysis, belonging to the field of database management technology. The evidence generation method includes: obtaining a data storage structure from a distributed search engine; using a multivariate collaborative analysis algorithm to find data entries that meet preset conditions from the data storage structure; performing correlation analysis on the selected data entries and template evidence to filter out key data content; expanding and parsing key fields of the key data content to obtain preliminary evidence; and standardizing the preliminary evidence to generate a unified evidence format. Because the multivariate collaborative analysis algorithm can effectively solve the pain points of low efficiency and missing key information in conventional data query and evidence generation schemes, it improves the accuracy of information query and filtering, the efficiency and universality of evidence generation, and provides strong support for various downstream tasks such as evidence correlation analysis, evidence auditing, evidence chain generation, and cross-certification.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention belongs to the field of database management technology, and more specifically, relates to an evidence generation method, apparatus and system based on multivariate collaborative analysis. Background Technology

[0002] Against the backdrop of large-scale distributed computing clusters and Industry 4.0, artificial intelligence, cloud computing, and the Internet of Things (IoT) technologies have developed rapidly. Correspondingly, numerous system vulnerabilities have been discovered, providing opportunities for attackers. Behind every cybersecurity incident, evidence provides crucial support for post-attack attribution, intelligence analysis, risk assessment, and system remediation, making it an indispensable part of the cybersecurity defense system.

[0003] However, in the context of big data, the amount of log data generated by distributed systems is enormous. Efficiently filtering out massive amounts of noise and accurately extracting key evidence information from this vast amount of data is extremely difficult. If an effective and efficient evidence generation method cannot be designed, it will cause numerous inconveniences for subsequent downstream tasks such as evidence correlation analysis, evidence chain construction, and cross-verification. However, manually generating evidence under massive data conditions is clearly inefficient and impractical. Most existing automated evidence generation methods only follow a specific format and system approach, exhibiting weak generalization ability and being severely limited by prior biases in human cognition. This results in the deletion of much key information while filtering out a large amount of noise, making it difficult to guarantee the completeness of the information in the final generated evidence. Furthermore, these methods do not perform further processing on the filtered evidence, resulting in poor field usability. Since downstream analysis tasks highly depend on the key information contained in the evidence itself, the evidence must possess characteristics such as field diversity, logical integrity, and contextual relevance, which also presents new challenges to evidence generation algorithms.

[0004] In conclusion, it is urgent and necessary to propose an efficient evidence generation method that enables multi-dimensional collaborative analysis, multi-dimensional information filtering, and field expansion. This method can provide strong support for many downstream tasks such as evidence correlation analysis, attack attribution, cross-authentication, and simulated attacks. Summary of the Invention

[0005] To address the aforementioned deficiencies or improvement needs of existing technologies, this invention provides an evidence generation method, apparatus, and system based on multivariate collaborative analysis. Its purpose is to utilize a multivariate collaborative analysis algorithm to identify data entries that meet preset conditions from the data storage structure, perform correlation analysis on the selected data entries and template evidence to filter out key data content, and then expand, parse, and standardize key fields to generate a unified evidence format. This solves the technical problems of missing key information, redundant information, low generation efficiency, and low data filtering accuracy in existing evidence generation methods.

[0006] To achieve the above objectives, according to one aspect of the present invention, an evidence generation method based on multivariate collaborative analysis is provided, comprising:

[0007] S1: Obtain the data storage structure from the distributed search engine;

[0008] S2: Use a multivariate collaborative analysis algorithm to find data entries that meet preset conditions from the data storage structure, perform correlation analysis on the screened data entries and template evidence, and screen out key data content;

[0009] S3: Expand and parse the key fields of the key data content to obtain preliminary evidence;

[0010] S4: Standardize the preliminary evidence to generate a unified evidence format.

[0011] In one embodiment, S1 includes: obtaining the data storage structure from the distributed search engine based on the port IP and index of the distributed search engine.

[0012] In one embodiment, S2 includes:

[0013] S21: The three-dimensional time rule filters out data entries for the specified attack date, and then selects data entries that meet the conditions from the data entries filtered by the three-dimensional time rule according to the subject IP rule and the object IP rule;

[0014] S22: Based on the modification rules, filter out the data entries with more suspicious behaviors from the data entries obtained in S21;

[0015] S23: Perform correlation analysis on the data items and template evidence selected in S22 to identify key data content.

[0016] In one embodiment, S23 includes:

[0017] S231: Use the first correlation analysis algorithm to perform correlation analysis and screening on the data items selected in S22 and the template evidence;

[0018] S232: Using the second correlation analysis algorithm, the data items selected in S231 are compared with the template evidence to obtain the key data content.

[0019] In one embodiment, the first correlation analysis algorithm is a grey correlation analysis algorithm; step S231 includes:

[0020] Using formula Calculate the i-th data entry X selected by S22.i The correlation coefficient ζ with template evidence Y i (k); k is the index of the k-th data field in the corresponding data entry, n is the number of data fields, m is the number of data content entries, ρ is the resolution coefficient, and the intermediate parameter Δ i (k)=|Y(k)′-X′ i (k)|, Y(k) represents the content of the k-th data field in the template evidence. Y(k)′ is the mean of all data fields in the template evidence, Y(k)′ is the normalized result of the k-th data field in the template evidence, and X is the mean of all data fields in the template evidence. i (k) represents the i-th data entry X. i The content of the k-th data field in the data. For the i-th data entry X i The mean of all data fields in X. i (k)′ represents the i-th data entry X i The normalization result of the k-th data field in the dataset;

[0021] Using formula Calculate the i-th data entry X selected by S231 i The correlation degree with template evidence Y is used to filter out data entries whose correlation degree is greater than a first threshold.

[0022] In one embodiment, the second correlation analysis algorithm is a Pearson correlation coefficient analysis algorithm; S232 includes:

[0023] Using formula Calculate the i-th data entry X selected by S231 i The Pearson correlation coefficient with template evidence Y is used to filter out data entries with a Pearson correlation coefficient greater than a second threshold as the key data content.

[0024] in,

[0025] X i Y(k) represents the k-th data field in the i-th data entry, and Y(k) represents the k-th data field in the template evidence. X represents i The sample standard deviation of (k), E(X) i ) represents X i The mean of the data field, where n represents the number of data entries.

[0026] In one embodiment, S3 includes:

[0027] S31. Calculate the one-hot encoding of the key data content, and add the one-hot encoded field to the field of the key data content;

[0028] S32. Calculate the timestamp of the key data content, and add the timestamp key field to the field obtained in S31 to obtain the total evidence and the single piece of evidence.

[0029] S33. The individual piece of evidence is added to the total evidence to generate the preliminary evidence.

[0030] In one embodiment, S4 includes:

[0031] The preliminary evidence is formatted and cleaned, and the cleaned evidence is converted into the standardized evidence format.

[0032] According to another aspect of the present invention, an evidence generation device based on multivariate collaborative analysis is provided, comprising:

[0033] The acquisition module is used to retrieve the data storage structure from the distributed search engine;

[0034] The filtering module is used to use a multivariate collaborative analysis algorithm to find data entries that meet preset conditions from the data storage structure, perform correlation analysis on the filtered data entries and template evidence, and filter out key data content.

[0035] The first generation module is used to expand and parse the key fields of the key data content to obtain preliminary evidence.

[0036] The second generation module is used to standardize the preliminary evidence and generate a unified evidence format.

[0037] According to another aspect of the present invention, an evidence generation system based on multivariate collaborative analysis is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the evidence generation method based on multivariate collaborative analysis.

[0038] According to another aspect of the present invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the evidence generation method based on multivariate collaborative analysis.

[0039] In summary, compared with the prior art, the above-described technical solutions conceived by this invention can achieve the following beneficial effects:

[0040] (1) This invention provides an evidence generation method based on multivariate collaborative analysis. The method utilizes a multivariate collaborative analysis algorithm to find data entries that meet preset conditions from the data storage structure. The filtered data entries and template evidence are then subjected to correlation analysis to identify key data content. Key fields are expanded and parsed, and finally, standardized processing is performed to generate a unified evidence format. Because the multivariate collaborative analysis algorithm effectively addresses the pain points of low efficiency and missing key information in conventional data query and evidence generation schemes, it improves the accuracy of information query and filtering, the efficiency and universality of evidence generation, and provides strong support for various downstream tasks such as evidence correlation analysis, evidence auditing, evidence chain generation, and cross-certification.

[0041] (2) This solution obtains the data storage structure from the distributed search engine based on the port IP and index of the distributed search engine. The operation is simple and can quickly collect the preset data storage structure.

[0042] (3) This scheme uses a three-dimensional time rule to filter out data entries for a specified attack date, and then selects data entries that meet the conditions from the data entries filtered by the three-dimensional time rule according to the subject IP rule and the object IP rule; it can efficiently filter data noise and realize the rapid location of evidence.

[0043] (4) This scheme uses the first correlation analysis algorithm to perform correlation analysis and filtering on the data items selected in S22 and the template evidence; and uses the second correlation analysis algorithm to perform correlation analysis and filtering on the data items selected in S231 and the template evidence to obtain the key data content. By sequentially filtering through the two correlation analysis algorithms, the false alarm rate of the algorithm can be reduced, and accurate correlation analysis can be achieved.

[0044] (5) The first correlation analysis algorithm in this scheme is the grey correlation analysis algorithm; the algorithm has low computational complexity, can effectively identify the correlation under the influence of multiple factors, and realize robust correlation analysis.

[0045] (6) The second correlation analysis algorithm in this scheme is the Pearson correlation coefficient analysis algorithm; this algorithm effectively solves the problem of rating level inflation, can effectively characterize linear correlation, and realize the dual measurement of evidence correlation.

[0046] (7) This solution calculates the one-hot encoding of the key data content, adds the one-hot encoded field to the fields of the key data content; calculates the timestamp of the key data content, adds the timestamp key field to the field obtained in S31 to obtain the total evidence and obtain a single piece of evidence; appends the single piece of evidence to the total evidence to generate the preliminary evidence. The key field expansion and parsing method provided by this solution can effectively construct the digital fingerprint of evidence and achieve universal information coverage of evidence.

[0047] (8) This solution performs formatted cleaning on the preliminary evidence and converts the cleaned evidence into the unified evidence format, which can provide strong support for downstream tasks such as evidence chain analysis and realize the standardized format conversion of evidence. Attached Figure Description

[0048] Figure 1 A flowchart of the efficient evidence generation method based on multivariate collaborative analysis provided in Embodiment 1 of the present invention;

[0049] Figure 2 A detailed flowchart of the efficient evidence generation method based on multivariate collaborative analysis provided in Embodiment 1 of the present invention.

[0050] Figure 3 The flowchart is shown in Embodiment 1 of the present invention.

[0051] Figure 4 This is a flowchart illustrating the expansion and parsing of key information fields provided in Embodiment 1 of the present invention;

[0052] Figure 5 This is a structural diagram of the unified evidence format provided in Embodiment 1 of the present invention. Detailed Implementation

[0053] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0054] Example 1

[0055] This embodiment provides an evidence generation method based on multivariate collaborative analysis, such as Figure 1 and Figure 2 As shown, the process includes: S1: obtaining the data storage structure from the distributed search engine; S2: using a multivariate collaborative analysis algorithm to find data entries that meet preset conditions from the data storage structure, performing correlation analysis on the screened data entries and template evidence, and filtering out key data content; S3: expanding and parsing the key data content to obtain preliminary evidence; S4: standardizing the preliminary evidence to generate a unified evidence format.

[0056] In one embodiment, S1 includes: obtaining the data storage structure in the distributed engine based on the IP address of the Elastic Search port and the index; and segmenting the acquired multi-source logs according to fields in a token format to form an initial log format.

[0057] In one embodiment, S2 includes: S21: using three-dimensional time rules to filter out data entries for a specified attack date, and then selecting data entries that meet the conditions from the data entries filtered by the three-dimensional time rules based on subject IP rules and object IP rules; S22: using modification rules to filter out data entries with more suspicious behavior from the data entries filtered by S21; S23: performing correlation analysis on the data entries filtered by S22 and template evidence to filter out key data content.

[0058] Specifically, for example, step S21 includes: the time format is yy / mm / dd / mm / ss, where each field represents year, month, day, minute, and second, respectively. Data entries for specific dates can be filtered, or a time range can be defined, such as a start time. time and End Time time The poetry selection filters data entries within that time frame. Step S22 specifically includes: (e.g.) Figure 3 In the process, data entries with the subject IP address of 192.168.53.1 are filtered. Data entries with the object IP address of 176.123.63.2 are also filtered. Furthermore, for example, if the attack evidence to be analyzed is sensitive to port changes and frequently uses port number changes for concealment, then data entries with frequently changing port numbers will be the focus. Figure 3 In the specific example given, the port number was changed from 5050 to 80, which is a highly suspicious data entry. Finally, the data entries selected in S22 and the template evidence were analyzed for correlation to identify key data content.

[0059] In one embodiment, S23 includes: S231: using a first correlation analysis algorithm to perform correlation analysis and filtering on the data items selected in S22 and the template evidence; S232: using a second correlation analysis algorithm to perform correlation analysis and filtering on the data items selected in S231 and the template evidence to obtain key data content.

[0060] In one embodiment, the first correlation analysis algorithm is a grey correlation analysis algorithm; S231 includes: using the formula Calculate the i-th data entry X selected by S22. i The correlation coefficient ζ with template evidence Y i (k); k is the index of the data field in the i-th data entry, n is the number of data fields, m is the number of data content entries, ρ is the resolution coefficient, and the intermediate parameter Δ i (k)=|Y(k)′-X′ i (k)|, Y(k) represents the content of the k-th data field in the template evidence. Y(k)′ is the mean of all data fields in the template evidence, Y(k)′ is the normalized result of the k-th data field in the template evidence, and X is the mean of all data fields in the template evidence. i (k) represents the i-th data entry X. i The content of the k-th data field in the data. For the i-th data entry X i The mean of all data fields in X. i (k)′ represents the i-th data entry X i The normalized result of the k-th data field; then using the formula Calculate the i-th data entry X selected by S231 i Based on the correlation with template evidence Y, data entries with a correlation greater than the first threshold are filtered out.

[0061] Step S231 specifically includes: first, vectorizing the data content to be filtered and the template data content respectively, to obtain X respectively. i =X i (k)|k=1,2,...n,i=1,2,...m. Where, X i Let Y represent the i-th data item to be filtered, k be the distinct data fields in the data item to be filtered, n be the number of data fields, and m be the number of data items. Y = Y(k) | k = 1, 2, ..., n. Here, Y is the template data content, and it is necessary to ensure that the number of data fields in the data item to be filtered is consistent with the number of data fields in the template data content.

[0062] Subsequently, mean normalization and dimensionless normalization are performed respectively, as shown below: in Similarly, for X i (k)′, has in Let Δ i (k)=|Y(k)′-X i If (k)′|, then calculate the correlation coefficient: ρ is the resolution coefficient, which can be chosen as 0.5.

[0063] Finally, the final correlation scores are calculated and ranked, as shown below: And filter out the data content that has a high degree of relevance.

[0064] In one embodiment, the second correlation analysis algorithm is the Pearson correlation coefficient analysis algorithm; S232 includes: using the formula Calculate the i-th data entry X selected by S231 iThe Pearson correlation coefficient with template evidence Y is used to filter data entries with a Pearson correlation coefficient greater than a second threshold as key data content; among them, X i Y(k) represents the k-th data field in the i-th data entry, and Y(k) represents the k-th data field in the template evidence. X represents i The sample standard deviation of (k), E(X) i ) represents X i The mean of the data field, where n represents the number of data entries.

[0065] In one embodiment, such as Figure 4 As shown, S3 includes: S31, calculating the one-hot encoding of the key data content and adding the one-hot encoded field to the field of the key data content; S32, calculating the timestamp of the key data content and adding the timestamp key field to the field obtained in S31 to obtain the total evidence and obtain the single piece of evidence; S33, appending the single piece of evidence to the total evidence to generate preliminary evidence.

[0066] Further, step S31 specifically includes: generating one-hot codes, as shown below:

[0067]

[0068] Among them, Onehot Encoding Let n be an identity matrix of size n×n, and let One-hot encoding be used. Encoding Add the field to the fields of the key data content.

[0069] Further, step S32 specifically includes: calculating a timestamp based on the current system time, in the format yy / mm / dd / mm / ss, and adding the timestamp field to the fields of the key data content.

[0070] In one embodiment, S4 includes: formatting and cleaning the preliminary evidence, and converting the cleaned evidence into a unified evidence format. Specifically, the unified evidence format generated using JSON key-value pairs includes the following fields: attack method, attack technique, attack type, attacker IP, event category, event description, associated event ID, event name, credibility, risk level, number of critical vulnerabilities, CVE number, vulnerability description, data source, anomaly score, original alert ID, malware family, threat category, victim IP, virus path, vulnerability impact, one-hot encoding, and generation timestamp. The corresponding unified evidence format structure is as follows: Figure 5 As shown.

[0071] Example 2

[0072] According to another aspect of the present invention, an evidence generation device based on multivariate collaborative analysis is provided, comprising: an acquisition module, a filtering module, a first generation module, and a second generation module. The acquisition module is used to acquire a data storage structure from a distributed search engine; the filtering module is used to use a multivariate collaborative analysis algorithm to find data entries that meet preset conditions from the data storage structure, perform correlation analysis between the filtered data entries and template evidence, and filter out key data content; the first generation module is used to expand and parse key fields of the key data content to obtain preliminary evidence; and the second generation module is used to standardize the preliminary evidence to generate a unified evidence format.

[0073] Example 3

[0074] According to another aspect of the present invention, an evidence generation system based on multivariate collaborative analysis is provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of the evidence generation method based on multivariate collaborative analysis.

[0075] Example 4

[0076] According to another aspect of the present invention, a computer-readable storage medium is provided having a computer program stored thereon, wherein the steps of an evidence generation method based on multivariate collaborative analysis are implemented when the computer program is executed by a processor.

[0077] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can 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. The solutions in the embodiments of the present invention can be implemented using various computer languages, such as the object-oriented programming language Java and the interpreted scripting language JavaScript.

[0078] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0079] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0080] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0081] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0082] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for generating evidence based on multivariate collaborative analysis, characterized in that, include: S1: Obtain the data storage structure from the distributed search engine; S2: Use a multivariate collaborative analysis algorithm to find data entries that meet preset conditions from the data storage structure, perform correlation analysis on the screened data entries and template evidence, and screen out key data content; S3: Expand and parse the key fields of the key data content to obtain preliminary evidence; S4: Standardize the preliminary evidence to generate a unified evidence format; S2 includes: S21: filtering data entries for a specified attack date based on the three-dimensional time rule, and then selecting data entries that meet the conditions from the data entries filtered by the three-dimensional time rule based on the subject IP rule and the object IP rule; S22: filtering data entries with more suspicious behavior from the data entries filtered by S21 based on the modification rule; S23: performing correlation analysis on the data entries filtered by S22 and template evidence to filter out key data content; S23 includes: S231: using a first correlation analysis algorithm to perform correlation analysis and filtering on the data items selected in S22 and the template evidence; S232: using a second correlation analysis algorithm to perform correlation analysis and filtering on the data items selected in S231 and the template evidence to obtain the key data content; S3 includes: S31, calculating the one-hot encoding of the key data content, and adding the one-hot encoding field to the field of the key data content; S32, calculating the timestamp of the key data content, and adding the timestamp key field to the field obtained in S31 to obtain a single piece of evidence; S33, appending the single piece of evidence to the total evidence to generate the preliminary evidence.

2. The evidence generation method based on multivariate collaborative analysis as described in claim 1, characterized in that, S1 includes: obtaining the data storage structure from the distributed search engine based on the port IP and index of the distributed search engine.

3. The evidence generation method based on multivariate collaborative analysis as described in claim 1, characterized in that, The first correlation analysis algorithm is a grey correlation analysis algorithm; S231 includes: Using formula Calculate the i-th data entry selected by S22 With template evidence correlation coefficient ; It is the index of the k-th data field in the corresponding data entry, n is the number of data fields, and m is the number of data content entries. For resolution coefficients, intermediate parameters Y , , , ; The content of the k-th data field in the template evidence. Y is the mean of all data fields in the template evidence. This represents the normalization result of the k-th data field in the template evidence. For the i-th data entry The content of the k-th data field in the data. For the i-th data entry The mean of all data fields in the data. For the i-th data entry The normalization result of the k-th data field in the dataset; Using formula Calculate the i-th data entry selected by S231 With template evidence The correlation degree is used to filter out data entries with a correlation degree greater than a first threshold.

4. The evidence generation method based on multivariate collaborative analysis as described in claim 1, characterized in that, The second correlation analysis algorithm is the Pearson correlation coefficient analysis algorithm; S232 includes: Using formula Calculate the i-th data entry selected by S231 With template evidence The Pearson correlation coefficient is used to filter out data entries with a Pearson correlation coefficient greater than the second threshold as the key data content. in, , ; This represents the k-th data field in the i-th data entry. This represents the k-th data field in the template evidence. express The sample standard deviation express The mean of the data field. Indicates the number of data fields.

5. The evidence generation method based on multivariate collaborative analysis as described in any one of claims 1-4, characterized in that, S4 includes: The preliminary evidence is formatted and cleaned, and the cleaned evidence is converted into the standardized evidence format.

6. An evidence generation device based on multivariate collaborative analysis, characterized in that, The evidence generation method based on multivariate collaborative analysis according to any one of claims 1-5 includes: The acquisition module is used to retrieve the data storage structure from the distributed search engine; The filtering module is used to use a multivariate collaborative analysis algorithm to find data entries that meet preset conditions from the data storage structure, perform correlation analysis on the filtered data entries and template evidence, and filter out key data content. The first generation module is used to expand and parse the key fields of the key data content to obtain preliminary evidence. The second generation module is used to standardize the preliminary evidence and generate a unified evidence format.

7. An evidence generation system based on multivariate collaborative analysis, comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the evidence generation method based on multivariate collaborative analysis as described in any one of claims 1 to 5.

Citation Information

Patent Citations

  • Evidence element identification method for ecological environmental protection cases and related equipment

    CN116824257A

  • System and method for determining dependencies between system resource performance characteristics

    US7031879B1