Multi-source threat intelligence fusion analysis method and system for network security
By acquiring, deduplication, clustering, sorting and fusion of threat intelligence data, the problem of potentially high correlation intelligence data omission in the existing technology is solved, and network security protection capabilities are improved.
Patent Information
- Application Number
- CN202410975394.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-19
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2044-07-19
AI Technical Summary
The existing threat intelligence fusion method leads to the omission of potentially highly correlated intelligence data, affecting network security protection capabilities, and leading to false alarms.
By acquiring and deduplication, clustering and sorting based on semantic similarity, calculating the correlation after positional relationship, and correlating and fusion based on preset thresholds and weights, using the field fusion rule table to generate fusion threat intelligence data .
It improves the efficiency of threat intelligence integration, avoids potentially highly correlated intelligence data omissions, and enhances network security protection capabilities.
Smart Images

Figure CN119030741A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of threat intelligence fusion, and in particular relates to a multi-source threat intelligence fusion analysis method and system for network security. Background Art
[0002] Threat intelligence is evidence-based knowledge that includes scenarios, mechanisms, indicators, impacts, and operational recommendations. Threat intelligence describes existing or upcoming threats or dangers to assets and can be used to inform the subject to take some response to the relevant threats or dangers.
[0003] At present, with the diversification of threat intelligence data sources, intelligence fusion is an important means to explore the value of security intelligence and a key way to achieve multi-source intelligence acquisition and integration in a big data environment. However, in terms of intelligence fusion, the current method is to directly fuse intelligence data with high correlation or similarity. The fusion granularity of this fusion method is relatively coarse, which will lead to the omission of some intelligence data with high potential correlation, which will cause false alarms to a certain extent and affect network security protection capabilities. Summary of the invention
[0004] The present invention provides a multi-source threat intelligence fusion analysis method and system for network security, which is used to solve the technical problem that the existing fusion method may cause the omission of some intelligence data with high potential correlation, thereby causing false alarms to a certain extent and affecting the network security protection capability.
[0005] In a first aspect, the present invention provides a multi-source threat intelligence fusion analysis method for network security, comprising:
[0006] Acquire at least one threat intelligence data within the same preset time period, and perform deduplication processing on the at least one threat intelligence data according to a preset screening rule to obtain at least one target threat intelligence data;
[0007] Determining a first degree of association between each target threat intelligence data based on semantic similarity, and clustering the at least one target threat intelligence data according to the first degree of association to obtain at least one target threat intelligence data set, wherein any two target threat intelligence data sets do not contain the same target threat intelligence data;
[0008] sorting the target threat intelligence data in the at least one target threat intelligence data set according to the magnitude of the first correlation degree, and obtaining at least one target threat intelligence data sequence corresponding to the cluster center of the at least one target threat intelligence data set, wherein the cluster center is a target threat intelligence data in the target threat intelligence data set that has a first correlation degree with all target threat intelligence data other than itself;
[0009] Aligning target threat intelligence data in two target threat intelligence data sequences based on a positional relationship, and calculating a second correlation degree between each group of aligned target threat intelligence data, wherein a group of target threat intelligence data includes first target threat intelligence data and second target threat intelligence data, the first target threat intelligence data is target threat intelligence data at a certain position in one target threat intelligence data sequence, and the second target threat intelligence data is target threat intelligence data at a certain position in another target threat intelligence data sequence;
[0010] Determine the target correlation between the two target threat intelligence data sequences according to the second correlation between the groups of target threat intelligence data and the target weights corresponding to the groups of target threat intelligence data;
[0011] Determine whether the target correlation between two target threat intelligence data sequences is greater than a preset threshold;
[0012] If it is greater than a preset threshold, two cluster centers corresponding to the two target threat intelligence data sequences are associated;
[0013] Based on the preset field fusion rule table, each target threat intelligence data set in a target threat intelligence data set and other cluster centers associated with the cluster center of the target threat intelligence data set are fused, and the fused threat intelligence data is written into a preset database.
[0014] In a second aspect, the present invention provides a multi-source threat intelligence fusion analysis system for network security, comprising:
[0015] An acquisition module is configured to acquire at least one threat intelligence data within the same preset time period, and perform deduplication processing on the at least one threat intelligence data according to a preset screening rule to obtain at least one target threat intelligence data;
[0016] A clustering module is configured to determine a first degree of association between each target threat intelligence data based on semantic similarity, and cluster the at least one target threat intelligence data according to the first degree of association to obtain at least one target threat intelligence data set, wherein any two target threat intelligence data sets do not contain the same target threat intelligence data;
[0017] A sorting module is configured to sort each target threat intelligence data in the at least one target threat intelligence data set according to the size of the first correlation degree, and obtain at least one target threat intelligence data sequence corresponding to the cluster center of the at least one target threat intelligence data set, wherein the cluster center is a target threat intelligence data in the target threat intelligence data set that has a first correlation degree with all target threat intelligence data other than itself;
[0018] A calculation module is configured to align the target threat intelligence data in two target threat intelligence data sequences based on a position relationship, and calculate a second correlation degree between each group of aligned target threat intelligence data, wherein a group of target threat intelligence data includes first target threat intelligence data and second target threat intelligence data, the first target threat intelligence data is target threat intelligence data at a certain position in one target threat intelligence data sequence, and the second target threat intelligence data is target threat intelligence data at a certain position in another target threat intelligence data sequence;
[0019] a determination module configured to determine a target correlation degree between the two target threat intelligence data sequences according to a second correlation degree between each group of target threat intelligence data and a target weight corresponding to each group of target threat intelligence data;
[0020] A judgment module is configured to judge whether the target correlation between two target threat intelligence data sequences is greater than a preset threshold;
[0021] an association module, configured to associate two cluster centers corresponding to the two target threat intelligence data sequences if the value is greater than a preset threshold;
[0022] The fusion module is configured to fuse each target threat intelligence data set in a target threat intelligence data set and other cluster centers associated with the cluster center of the target threat intelligence data set based on a preset field fusion rule table, and write the fused threat intelligence data into a preset database.
[0023] According to a third aspect, an electronic device is provided, comprising: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can perform the steps of the multi-source threat intelligence fusion analysis method for network security of any embodiment of the present invention.
[0024] In a fourth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor executes the steps of the multi-source threat intelligence fusion analysis method for network security of any embodiment of the present invention.
[0025] The multi-source threat intelligence fusion analysis method and system for network security of this application, BRIEF DESCRIPTION OF THE DRAWINGS
[0026] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0027] Figure 1 A flowchart of a multi-source threat intelligence fusion analysis method for network security provided by one embodiment of the present invention;
[0028] Figure 2 A structural block diagram of a multi-source threat intelligence fusion analysis system for network security provided by one embodiment of the present invention;
[0029] Figure 3 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0030] 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.
[0031] See also Figure 1 , which shows a flowchart of a multi-source threat intelligence fusion analysis method for network security in the present application.
[0032] like Figure 1 As shown, the multi-source threat intelligence fusion analysis method for network security specifically includes the following steps:
[0033] Step S101, obtaining at least one target threat intelligence data within the same preset time period.
[0034] In this step, the at least one target threat intelligence data is at least one target threat intelligence data from the same source and / or at least one target threat intelligence data from different sources. The acquisition method may be, for example, data analysis, paid purchase, etc.
[0035] Further preferably, the target threat intelligence data obtained from each intelligence source is converted into standard format data to facilitate subsequent data processing.
[0036] Step S102, based on semantic similarity, determines the first correlation between each target threat intelligence data, and clusters the at least one target threat intelligence data according to the first correlation to obtain at least one target threat intelligence data set, wherein any two target threat intelligence data sets do not contain the same target threat intelligence data.
[0037] In this step, a first correlation between two target threat intelligence data is determined based on semantic similarity, and it is determined whether the first correlation is greater than a preset first correlation threshold; if the first correlation is greater than the preset correlation threshold, a third correlation greater than a preset second correlation threshold is obtained between a certain target threat intelligence data and other target threat intelligence data, and between another target threat intelligence data and other target threat intelligence data, and it is determined whether the number of third correlations greater than the preset second correlation threshold corresponding to a certain target threat intelligence data is greater than the number of third correlations greater than the preset second correlation threshold corresponding to another target threat intelligence data; if the number of third correlations greater than the preset second correlation threshold corresponding to a certain target threat intelligence data is greater than the number of third correlations greater than the preset second correlation threshold corresponding to another target threat intelligence data, then a certain target threat intelligence data is clustered as a clustering center to obtain a target threat intelligence data set; if the number of third correlations greater than the preset second correlation threshold corresponding to a certain target threat intelligence data is not greater than the number of third correlations greater than the preset second correlation threshold corresponding to another target threat intelligence data, then another target threat intelligence data is clustered as a clustering center to obtain a target threat intelligence data set.
[0038] Furthermore, if the first correlation between two target threat intelligence data is not greater than a preset first correlation threshold, neither of the two target threat intelligence data is used as a clustering center, but other clustering centers are sought.
[0039] It should be noted that determining the first correlation between two target threat intelligence data based on semantic similarity is specifically as follows: vectorizing two target threat intelligence data based on a semantic pre-training model to obtain two threat intelligence semantic vectors, and using a similarity algorithm to calculate the similarity of two threat intelligence semantic vectors, that is, obtaining the first correlation between the two target threat intelligence data. Among them, the semantic pre-training model includes at least two of a first semantic pre-training model based on word vectors, a second semantic pre-training model based on characters, and a third semantic training model based on word embedding. Similarity algorithms include but are not limited to cosine distance, Hamming distance, Manhattan distance, Euclidean distance or Jaccard similarity coefficient.
[0040] Specifically, the first semantic training model based on word vectors can be, for example, a TF-IDF (termfrequency-inverse document frequency) algorithm that converts each word in the target threat intelligence data into a threat intelligence semantic vector. The second semantic pre-training model based on characters can be, for example, a threat intelligence semantic vector obtained by a SimHash algorithm. The third semantic training model based on word embedding can be, for example, a word2vec algorithm that converts each word in the target threat intelligence data into a word vector to obtain a threat intelligence semantic vector.
[0041] Step S103, sorting each target threat intelligence data in the at least one target threat intelligence data set according to the size of the first correlation degree, and obtaining at least one target threat intelligence data sequence corresponding to the cluster center of the at least one target threat intelligence data set, wherein the cluster center is a target threat intelligence data in the target threat intelligence data set that has a first correlation degree with all target threat intelligence data other than itself.
[0042] For example, if the cluster center of a target threat intelligence data set is target threat intelligence data A, then other target threat intelligence data in the target threat intelligence data set have first correlations of different sizes with the target threat intelligence data A, and the minimum value of these first correlations of different sizes is greater than the preset first correlation threshold.
[0043] Step S104, aligning the target threat intelligence data in two target threat intelligence data sequences based on the position relationship, and calculating the second correlation degree between each group of aligned target threat intelligence data, wherein a group of target threat intelligence data includes first target threat intelligence data and second target threat intelligence data, the first target threat intelligence data is the target threat intelligence data at a certain position in one target threat intelligence data sequence, and the second target threat intelligence data is the target threat intelligence data at a certain position in another target threat intelligence data sequence.
[0044] In this step, the first target threat intelligence data in a target threat intelligence data sequence at the same position point is aligned with the second target threat intelligence data in another target threat intelligence data sequence; and the second correlation between the first target threat intelligence data and the second target threat intelligence data is calculated based on semantic similarity.
[0045] For example, in the target threat intelligence data sequence a, there are target threat intelligence data A1 located at the first position, target threat intelligence data A2 located at the second position, and target threat intelligence data A3 located at the third position; in the target threat intelligence data sequence b, there are target threat intelligence data B1 located at the first position, target threat intelligence data B2 located at the second position, target threat intelligence data B3 located at the third position, and target threat intelligence data B4 located at the fourth position, wherein the first position, the second position, the third position, and the fourth position all refer to the sorting position of the target threat intelligence data in the target threat intelligence data sequence.
[0046] After alignment according to the method of this embodiment, the target threat intelligence data A1 corresponds to the target threat intelligence data B1, the target threat intelligence data A2 corresponds to the target threat intelligence data B2, and the target threat intelligence data A3 corresponds to the target threat intelligence data B3. Then, based on the semantic similarity, the second correlation between the target threat intelligence data A1 and the target threat intelligence data B1, the second correlation between the target threat intelligence data A2 and the target threat intelligence data B2, and the second correlation between the target threat intelligence data A3 and the target threat intelligence data B3 are calculated respectively.
[0047] Step S105: determining the target correlation between the two target threat intelligence data sequences according to the second correlation between the groups of target threat intelligence data and the target weights corresponding to the groups of target threat intelligence data.
[0048] In this step, the number of groups of target threat intelligence data in two target threat intelligence data sequences that are aligned based on the position relationship is obtained;
[0049] If the number of groups is an even number, each group of target threat intelligence data is divided into a first part and a second part of the same number of groups, and the target weight corresponding to each group of target threat intelligence data in the first part is determined according to a preset first calculation rule, and the target weight corresponding to each group of target threat intelligence data in the second part is determined according to a preset second calculation rule, wherein the expression of the first calculation rule is:
[0050]
[0051] Where W i is the target weight corresponding to the i-th group of target threat intelligence data in the first part, and N is the number of groups of target threat intelligence data in two target threat intelligence data sequences after alignment based on the position relationship;
[0052] The expression of the second calculation rule is:
[0053]
[0054] Where N′ is the number of groups of target threat intelligence data after alignment in the first part, and W j is the target weight corresponding to the j-th group of target threat intelligence data in the second part;
[0055] If the number of groups is an odd number, each group of target threat intelligence data is divided into a first part, a second part, and a third part located in the middle with the same number of groups, and the target weight corresponding to each group of target threat intelligence data in the first part is determined according to a preset first calculation rule, the target weight corresponding to each group of target threat intelligence data in the second part is determined according to a preset second calculation rule, and the target weight corresponding to a group of target threat intelligence data in the third part is determined according to a preset third calculation rule, wherein the expression of the third calculation rule is:
[0056]
[0057] Where W k is the target weight corresponding to the kth group of target threat intelligence data in the third part.
[0058] For example, after the target threat intelligence data sequence a corresponding to the target threat intelligence data A with the cluster center and the target threat intelligence data sequence b corresponding to the target threat intelligence data B with the cluster center are aligned, four groups of target threat intelligence data are generated: the first group (target threat intelligence data A1 and target threat intelligence data B1), the second group (target threat intelligence data Δ2 and target threat intelligence data B2), the third group (target threat intelligence data A3 and target threat intelligence data B3) and the fourth group (target threat intelligence data A4 and target threat intelligence data B4).
[0059] Among them, the target weight of the first group is: 1 / 4+1 / 10, the target weight of the second group is: 1 / 4+1 / 20, the target weight of the third group is: 1 / 4-1 / 20, and the target weight of the fourth group is: 1 / 4-1 / 10.
[0060] For another example, after the target threat intelligence data sequence a and the target threat intelligence data sequence b are aligned, five groups of target threat intelligence data are generated: the first group (target threat intelligence data A1 and target threat intelligence data B1), the second group (target threat intelligence data A2 and target threat intelligence data B2), the third group (target threat intelligence data A3 and target threat intelligence data B3), the fourth group (target threat intelligence data A4 and target threat intelligence data B4) and the fifth group (target threat intelligence data A5 and target threat intelligence data B5).
[0061] Among them, the target weight of the first group is: 1 / 4+1 / 10, the target weight of the second group is: 1 / 4+1 / 20, the target weight of the third group is: 1 / 4, the target weight of the fourth group is: 1 / 4-1 / 20, and the target weight of the fourth group is: 1 / 4-1 / 10.
[0062] Afterwards, the target weight index of each group is multiplied by the second correlation degree of the corresponding group to obtain the target correlation degree between the target threat intelligence data sequence a and the target threat intelligence data sequence b. Therefore, the potential target correlation degree between the cluster center being the target threat intelligence data A and the cluster center being the target threat intelligence data B is obtained.
[0063] Step S106, determining whether the target correlation between two target threat intelligence data sequences is greater than a preset threshold.
[0064] In this step, if it is not greater than the preset threshold, the two cluster centers corresponding to the two target threat intelligence data sequences will not be associated, and the target correlation between the other two target threat intelligence data sequences will be determined to be greater than the preset threshold. For example, if the potential target correlation between the cluster center target threat intelligence data A and the cluster center target threat intelligence data B is not greater than the preset threshold, it means that the potential relationship between the target threat intelligence data A and the cluster center target threat intelligence data B is not strong, that is, in the process of fusing the set where the target threat intelligence data A is located, the target threat intelligence data B can be discarded, and the phenomenon of low fusion accuracy is not likely to occur.
[0065] Step S107: If it is greater than a preset threshold, two cluster centers corresponding to the two target threat intelligence data sequences are associated.
[0066] Step S108, based on the preset field fusion rule table, each target threat intelligence data in a target threat intelligence data set and other cluster centers associated with the cluster center of the target threat intelligence data set are fused, and the fused threat intelligence data is written into a preset database.
[0067] In this step, deduplication fields and fusion fields are determined based on the type of each target threat intelligence data, and the field fusion rule table is generated; deduplication operations are performed on the deduplication fields in each target threat intelligence data in the target threat intelligence data set and the deduplication fields in other cluster centers associated with the cluster center of the target threat intelligence data set according to the field fusion rule table; fusion operations are performed on the fusion fields in each target threat intelligence data in the target threat intelligence data set and the fusion fields in other cluster centers associated with the cluster center of the target threat intelligence data set according to the field fusion rule table.
[0068] Specifically, the types of target threat intelligence data may include Threat-actor (attack subject), Campaign (attack activity), Course of action (countermeasures), Report (report), attack indicator (Indicator), attack target (ExploitTarget), attack method (TTP), etc. According to the different types of target threat intelligence data, the deduplication fields and fusion fields are determined by manual methods.
[0069] In summary, the method of the present application sorts each target threat intelligence data in at least one target threat intelligence data set according to the size of the first correlation degree, and obtains at least one target threat intelligence data sequence corresponding to the cluster center of at least one target threat intelligence data set. After that, the target threat intelligence data in two target threat intelligence data sequences are aligned based on the position relationship, and the second correlation degree between each group of aligned target threat intelligence data is calculated. Then, the target correlation degree between two target threat intelligence data sequences is determined according to the second correlation degree between each group of target threat intelligence data and the target weight corresponding to each group of target threat intelligence data, and it is judged whether the target correlation degree between two target threat intelligence data sequences is Greater than a preset threshold, if greater than the preset threshold, then associate two cluster centers corresponding to two target threat intelligence data sequences, and finally, based on the preset field fusion rule table, fuse each target threat intelligence data in a target threat intelligence data set and other cluster centers associated with a cluster center of a target threat intelligence data set, and write the fused threat intelligence data into the preset database, which can fuse target threat intelligence data with potential correlation together, improve the integration efficiency of threat intelligence in the fusion process, and solve the problem that the existing fusion method will lead to the omission of some intelligence data with high potential correlation, which will cause false alarms to a certain extent and affect network security protection capabilities.
[0070] See also Figure 2 , which shows a structural block diagram of a multi-source threat intelligence fusion analysis system for network security in the present application.
[0071] like Figure 2 As shown, the multi-source threat intelligence fusion analysis system 200 includes an acquisition module 210, a clustering module 220, a sorting module 230, a calculation module 240, a determination module 250, a judgment module 260, an association module 270 and a fusion module 280.
[0072] The acquisition module 210 is configured to acquire at least one threat intelligence data within the same preset time period, and perform deduplication processing on the at least one threat intelligence data according to a preset screening rule to obtain at least one target threat intelligence data;
[0073] The clustering module 220 is configured to determine a first correlation between each target threat intelligence data based on semantic similarity, and cluster the at least one target threat intelligence data according to the first correlation to obtain at least one target threat intelligence data set, wherein any two target threat intelligence data sets do not contain the same target threat intelligence data;
[0074] The sorting module 230 is configured to sort each target threat intelligence data in the at least one target threat intelligence data set according to the size of the first correlation degree, and obtain at least one target threat intelligence data sequence corresponding to the cluster center of the at least one target threat intelligence data set, wherein the cluster center is a target threat intelligence data in the target threat intelligence data set that has a first correlation degree with all target threat intelligence data other than itself;
[0075] The calculation module 240 is configured to align the target threat intelligence data in two target threat intelligence data sequences based on the position relationship, and calculate the second correlation degree between each group of aligned target threat intelligence data, wherein a group of target threat intelligence data includes first target threat intelligence data and second target threat intelligence data, the first target threat intelligence data is the target threat intelligence data at a certain position in one target threat intelligence data sequence, and the second target threat intelligence data is the target threat intelligence data at a certain position in another target threat intelligence data sequence;
[0076] A determination module 250 is configured to determine a target correlation between the two target threat intelligence data sequences according to a second correlation between each group of target threat intelligence data and a target weight corresponding to each group of target threat intelligence data;
[0077] A determination module 260 is configured to determine whether a target correlation between two target threat intelligence data sequences is greater than a preset threshold;
[0078] The association module 270 is configured to associate two cluster centers corresponding to the two target threat intelligence data sequences if the value is greater than a preset threshold;
[0079] The fusion module 280 is configured to fuse each target threat intelligence data set in a target threat intelligence data set and other cluster centers associated with the cluster center of the target threat intelligence data set based on a preset field fusion rule table, and write the fused threat intelligence data into a preset database.
[0080] It should be understood that Figure 2 Modules and references documented in Figure 1 Therefore, the operations and features described above for the method and the corresponding technical effects are also applicable to Figure 2 The modules in it will not be described in detail here.
[0081] In some other embodiments, the embodiments of the present invention further provide a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor executes the multi-source threat intelligence fusion analysis method for network security in any of the above method embodiments;
[0082] As an implementation mode, the computer-readable storage medium of the present invention stores computer-executable instructions, and the computer-executable instructions are configured as follows:
[0083] Acquire at least one target threat intelligence data within the same preset time period;
[0084] Determining a first degree of association between each target threat intelligence data based on semantic similarity, and clustering the at least one target threat intelligence data according to the first degree of association to obtain at least one target threat intelligence data set, wherein any two target threat intelligence data sets do not contain the same target threat intelligence data;
[0085] sorting the target threat intelligence data in the at least one target threat intelligence data set according to the magnitude of the first correlation degree, and obtaining at least one target threat intelligence data sequence corresponding to the cluster center of the at least one target threat intelligence data set, wherein the cluster center is a target threat intelligence data in the target threat intelligence data set that has a first correlation degree with all target threat intelligence data other than itself;
[0086] Aligning target threat intelligence data in two target threat intelligence data sequences based on a positional relationship, and calculating a second correlation degree between each group of aligned target threat intelligence data, wherein a group of target threat intelligence data includes first target threat intelligence data and second target threat intelligence data, the first target threat intelligence data is target threat intelligence data at a certain position in one target threat intelligence data sequence, and the second target threat intelligence data is target threat intelligence data at a certain position in another target threat intelligence data sequence;
[0087] Determine the target correlation between the two target threat intelligence data sequences according to the second correlation between the groups of target threat intelligence data and the target weights corresponding to the groups of target threat intelligence data;
[0088] Determine whether the target correlation between two target threat intelligence data sequences is greater than a preset threshold;
[0089] If it is greater than a preset threshold, two cluster centers corresponding to the two target threat intelligence data sequences are associated;
[0090] Based on the preset field fusion rule table, each target threat intelligence data in a target threat intelligence data set and other cluster centers associated with the cluster center of the target threat intelligence data set are fused, and the fused threat intelligence data is written into a preset database.
[0091] The computer-readable storage medium may include a program storage area and a data storage area, wherein the program storage area may store an operating system and application programs required for at least one function; the data storage area may store data created according to the use of a multi-source threat intelligence fusion analysis system for network security, etc. In addition, the computer-readable storage medium may include a high-speed random access memory, and may also include a memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the computer-readable storage medium may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the multi-source threat intelligence fusion analysis system for network security via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.
[0092] Figure 3 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention, such as Figure 3 As shown, the device includes: a processor 310 and a memory 320. The electronic device may also include: an input device 330 and an output device 340. The processor 310, the memory 320, the input device 330 and the output device 340 may be connected via a bus or other means. Figure 3 In the example, the connection via bus is taken as an example. The memory 320 is the above-mentioned computer-readable storage medium. The processor 310 executes various functional applications and data processing of the server by running the non-volatile software programs, instructions and modules stored in the memory 320, that is, the multi-source threat intelligence fusion analysis method for network security of the above-mentioned method embodiment is implemented. The input device 330 can receive input digital or character information, and generate key signal input related to user settings and function control of the multi-source threat intelligence fusion analysis system for network security. The output device 340 may include display devices such as display screens.
[0093] The electronic device can execute the method provided by the embodiment of the present invention, and has the functional modules and beneficial effects corresponding to the execution method. For technical details not described in detail in this embodiment, please refer to the method provided by the embodiment of the present invention.
[0094] As an implementation mode, the electronic device is applied to a multi-source threat intelligence fusion analysis system for network security, and is used for a client, and includes: at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can:
[0095] Acquire at least one target threat intelligence data within the same preset time period;
[0096] Determining a first degree of association between each target threat intelligence data based on semantic similarity, and clustering the at least one target threat intelligence data according to the first degree of association to obtain at least one target threat intelligence data set, wherein any two target threat intelligence data sets do not contain the same target threat intelligence data;
[0097] sorting the target threat intelligence data in the at least one target threat intelligence data set according to the magnitude of the first correlation degree, and obtaining at least one target threat intelligence data sequence corresponding to the cluster center of the at least one target threat intelligence data set, wherein the cluster center is a target threat intelligence data in the target threat intelligence data set that has a first correlation degree with all target threat intelligence data other than itself;
[0098] Aligning target threat intelligence data in two target threat intelligence data sequences based on a positional relationship, and calculating a second correlation degree between each group of aligned target threat intelligence data, wherein a group of target threat intelligence data includes first target threat intelligence data and second target threat intelligence data, the first target threat intelligence data is target threat intelligence data at a certain position in one target threat intelligence data sequence, and the second target threat intelligence data is target threat intelligence data at a certain position in another target threat intelligence data sequence;
[0099] Determine the target correlation between the two target threat intelligence data sequences according to the second correlation between the groups of target threat intelligence data and the target weights corresponding to the groups of target threat intelligence data;
[0100] Determine whether the target correlation between two target threat intelligence data sequences is greater than a preset threshold;
[0101] If it is greater than a preset threshold, two cluster centers corresponding to the two target threat intelligence data sequences are associated;
[0102] Based on the preset field fusion rule table, each target threat intelligence data in a target threat intelligence data set and other cluster centers associated with the cluster center of the target threat intelligence data set are fused, and the fused threat intelligence data is written into a preset database.
[0103] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of each embodiment or some parts of the embodiment.
[0104] 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 aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A multi-source threat intelligence fusion analysis method for network security, characterized in that: The analytical method specifically comprises the following steps: Acquire at least one target threat intelligence data within the same preset time period; Determining a first degree of association between each target threat intelligence data based on semantic similarity, and clustering the at least one target threat intelligence data according to the first degree of association to obtain at least one target threat intelligence data set; sorting the target threat intelligence data in the at least one target threat intelligence data set according to the magnitude of the first correlation degree, and obtaining at least one target threat intelligence data sequence corresponding to the cluster center of the at least one target threat intelligence data set, wherein the cluster center is a target threat intelligence data in the target threat intelligence data set that has a first correlation degree with all target threat intelligence data other than itself; Aligning target threat intelligence data in two target threat intelligence data sequences based on a positional relationship, and calculating a second correlation degree between each group of aligned target threat intelligence data, wherein a group of target threat intelligence data includes first target threat intelligence data and second target threat intelligence data, the first target threat intelligence data is target threat intelligence data at a certain position in one target threat intelligence data sequence, and the second target threat intelligence data is target threat intelligence data at a certain position in another target threat intelligence data sequence; Determine the target correlation between the two target threat intelligence data sequences according to the second correlation between the groups of target threat intelligence data and the target weights corresponding to the groups of target threat intelligence data; Determine whether the target correlation between two target threat intelligence data sequences is greater than a preset threshold; If it is greater than a preset threshold, two cluster centers corresponding to the two target threat intelligence data sequences are associated; Based on the preset field fusion rule table, each target threat intelligence data in a target threat intelligence data set and other cluster centers associated with the cluster center of the target threat intelligence data set are fused, and the fused threat intelligence data is written into a preset database.
2. The analysis method according to claim 1, characterized in that The target threat intelligence data is at least one target threat intelligence data from the same source and / or at least one target threat intelligence data from different sources.
3. The analysis method according to claim 1, characterized in that The determining of a first correlation between each target threat intelligence data based on semantic similarity, and clustering the at least one target threat intelligence data according to the first correlation to obtain at least one target threat intelligence data set includes: Determine a first correlation between two target threat intelligence data based on semantic similarity, and determine whether the first correlation is greater than a preset first correlation threshold; If the first correlation is greater than the preset correlation threshold, then respectively obtain a third correlation greater than the preset second correlation threshold between a certain target threat intelligence data and other target threat intelligence data, and between another target threat intelligence data and other target threat intelligence data, and determine whether the number of third correlations greater than the preset second correlation threshold corresponding to a certain target threat intelligence data is greater than the number of third correlations greater than the preset second correlation threshold corresponding to another target threat intelligence data; If the number of third correlations greater than the preset second correlation threshold corresponding to a certain target threat intelligence data is greater than the number of third correlations greater than the preset second correlation threshold corresponding to another target threat intelligence data, clustering is performed using the certain target threat intelligence data as a clustering center to obtain a target threat intelligence data set; If the number of third correlation degrees greater than the preset second correlation threshold corresponding to a certain target threat intelligence data is not greater than the number of third correlation degrees greater than the preset second correlation threshold corresponding to another target threat intelligence data, then another target threat intelligence data is clustered as the clustering center to obtain a target threat intelligence data set.
4. The analysis method according to claim 1, characterized in that The step of aligning the target threat intelligence data in two target threat intelligence data sequences based on the position relationship and calculating the second correlation degree between each group of aligned target threat intelligence data includes: Aligning first target threat intelligence data in a target threat intelligence data sequence with second target threat intelligence data in another target threat intelligence data sequence at the same position point; A second degree of association between the first target threat intelligence data and the second target threat intelligence data is calculated based on the semantic similarity.
5. The analysis method according to claim 1, characterized in that Before determining the target correlation between the two target threat intelligence data sequences according to the second correlation between the groups of target threat intelligence data and the target weights corresponding to the groups of target threat intelligence data, the method further includes: Obtain the number of groups of target threat intelligence data in two target threat intelligence data sequences that are aligned based on the position relationship; If the number of groups is an even number, each group of target threat intelligence data is divided into a first part and a second part of the same number of groups, and the target weight corresponding to each group of target threat intelligence data in the first part is determined according to a preset first calculation rule, and the target weight corresponding to each group of target threat intelligence data in the second part is determined according to a preset second calculation rule, wherein the expression of the first calculation rule is: Where W i is the target weight corresponding to the i-th group of target threat intelligence data in the first part, and N is the number of groups of target threat intelligence data in two target threat intelligence data sequences aligned based on the position relationship; The expression of the second calculation rule is: Where N′ is the number of groups of target threat intelligence data after alignment in the first part, and W j is the target weight corresponding to the j-th group of target threat intelligence data in the second part; If the number of groups is an odd number, each group of target threat intelligence data is divided into a first part, a second part, and a third part located in the middle with the same number of groups, and the target weight corresponding to each group of target threat intelligence data in the first part is determined according to a preset first calculation rule, the target weight corresponding to each group of target threat intelligence data in the second part is determined according to a preset second calculation rule, and the target weight corresponding to a group of target threat intelligence data in the third part is determined according to a preset third calculation rule, wherein the expression of the third calculation rule is: Where W k is the target weight corresponding to the kth group of target threat intelligence data in the third part.
6. The analysis method according to claim 1, characterized in that After determining whether the target correlation between two target threat intelligence data sequences is greater than a preset threshold, the method further includes: If it is not greater than the preset threshold, the two cluster centers corresponding to the two target threat intelligence data sequences are not associated, and it is continued to be determined whether the target correlation between the other two target threat intelligence data sequences is greater than the preset threshold.
7. The analysis method according to claim 1, characterized in that The fusing operation of each target threat intelligence data in a target threat intelligence data set and other cluster centers associated with the cluster center of the target threat intelligence data set based on the preset field fusion rule table includes: Determine the deduplication fields and fusion fields based on the types of each target threat intelligence data, and generate the field fusion rule table; Deduplication operations are performed on deduplication fields in each target threat intelligence data in the target threat intelligence data set and deduplication fields in other cluster centers associated with the cluster center of the target threat intelligence data set according to the field fusion rule table; According to the field fusion rule table, a fusion operation is performed on the fusion fields in each target threat intelligence data in the certain target threat intelligence data set and the fusion fields in other cluster centers associated with the cluster center of the certain target threat intelligence data set.
8. A multi-source threat intelligence fusion analysis system for network security, characterized in that: include: An acquisition module is configured to acquire at least one threat intelligence data within the same preset time period, and perform deduplication processing on the at least one threat intelligence data according to a preset screening rule to obtain at least one target threat intelligence data; A clustering module is configured to determine a first degree of association between each target threat intelligence data based on semantic similarity, and cluster the at least one target threat intelligence data according to the first degree of association to obtain at least one target threat intelligence data set, wherein any two target threat intelligence data sets do not contain the same target threat intelligence data; A sorting module is configured to sort each target threat intelligence data in the at least one target threat intelligence data set according to the size of the first correlation degree, and obtain at least one target threat intelligence data sequence corresponding to the cluster center of the at least one target threat intelligence data set, wherein the cluster center is a target threat intelligence data in the target threat intelligence data set that has a first correlation degree with all target threat intelligence data other than itself; A calculation module is configured to align the target threat intelligence data in two target threat intelligence data sequences based on a position relationship, and calculate a second correlation degree between each group of aligned target threat intelligence data, wherein a group of target threat intelligence data includes first target threat intelligence data and second target threat intelligence data, the first target threat intelligence data is target threat intelligence data at a certain position in one target threat intelligence data sequence, and the second target threat intelligence data is target threat intelligence data at a certain position in another target threat intelligence data sequence; a determination module configured to determine a target correlation degree between the two target threat intelligence data sequences according to a second correlation degree between each group of target threat intelligence data and a target weight corresponding to each group of target threat intelligence data; A judgment module is configured to judge whether the target correlation between two target threat intelligence data sequences is greater than a preset threshold; an association module, configured to associate two cluster centers corresponding to the two target threat intelligence data sequences if the value is greater than a preset threshold; The fusion module is configured to fuse each target threat intelligence data set in a target threat intelligence data set and other cluster centers associated with the cluster center of the target threat intelligence data set based on a preset field fusion rule table, and write the fused threat intelligence data into a preset database.
9. An electronic device, characterized in that: include: At least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method described in any one of claims 1 to 7.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.
Citation Information
Patent Citations
Multi-source threat intelligence fusion method and device, equipment and storage medium
CN114925757A
Security protection construction method based on big data threat perception and AI protection system
CN115065545A
Systems and methods for assessing software vulnerabilities through a combination of external threat intelligence and internal enterprise information technology data
US20210288991A1
