Information security management system based on big data
Through big data technology, data positioning, fusion and multi-dimensional relationship analysis are carried out in the information security management system, and dynamic security strategies are built, which solves the shortcomings of existing systems in multi-source data processing and dynamic security management, and improves the flexibility of information security and response speed.
Patent Information
- Application Number
- CN202411438968.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-15
- Publication Date
- 2024-11-26
AI Technical Summary
Existing information security management systems are difficult to effectively process multi-source data in a big data environment, and lack data fusion and analysis mechanisms, resulting in inaccurate security policies and slow response speeds, and cannot be dynamically adjusted to cope with complex security needs.
Adopt a big data-based information security management system to build dynamic security strategies through data positioning, fusion and multi-dimensional relationship analysis to achieve accurate encryption and management of target information data, including data positioning modules, data fusion modules and policy determination modules, and use Semantic networks, conflict handling and multi-dimensional correlation analysis generates multi-dimensional security policies.
It greatly improves the flexibility, response speed and overall security of information security protection, solves the problem of lack of data fusion and dynamic security strategies in the existing technology, and realizes effective fusion and precise encryption of multi-source data.
Smart Images

Figure CN119026158A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of information processing security technology, and in particular to an information security management system based on big data. Background Art
[0002] With the rapid development and widespread application of big data technology, information security management has become an important requirement for various industries. Existing security management systems mostly use a single data source or rely on fixed security policies, which makes it difficult to cope with the complex security needs in a multi-dimensional and dynamic big data environment.
[0003] In the prior art, some information security management systems use simple data encryption or firewall strategies to protect data, but they lack the ability to integrate and process multi-source data in a big data environment, making it difficult to monitor and analyze potential security threats from a global perspective, and it is difficult to dynamically adjust security policies. The attached figure shows the basic architecture of the existing system, the main defect of which is that it cannot effectively process large-scale, multi-dimensional information data, nor can it update security policies in a timely manner according to dynamically changing user needs. The main defect of the existing technical solution is that the existing system lacks an effective data fusion and analysis mechanism when processing multi-source data, resulting in inaccurate security policies and slow response speeds. In addition, data encryption and management technologies are relatively simple and cannot meet the needs of dynamic security management in a big data environment.
[0004] Therefore, the present invention provides an information security management system based on big data. Summary of the invention
[0005] The present invention provides an information security management system based on big data, which is used to build dynamic security strategies for user needs through data positioning, fusion and multi-dimensional relationship analysis, realize accurate encryption and management of target information data, effectively integrate multi-source data, and generate multi-dimensional security strategies based on big data analysis, which greatly improves the flexibility, response speed and overall security of information security protection, and solves the problem of lack of data fusion and dynamic security strategies in the prior art.
[0006] The present invention provides an information security management system based on big data, comprising: Data location module: locates target information data from several preset data sources based on user needs; Data fusion module: performs data fusion on the target information data obtained by positioning to obtain fused information data; Strategy determination module: Analyzes the fusion information data based on the preset algorithm and constructs a multi-dimensional data relationship map to determine the security strategy corresponding to the user's needs; Data management module: Encrypt target information data based on security policies corresponding to user needs and manage it based on preset technologies.
[0007] The present invention provides an information security management system based on big data, a data positioning module, comprising: Request processing unit: pre-processes the original request of the user, obtains the preliminary semantic information of key entities, keywords and sentence structure in the original request of the user, and semantically expands the extracted key entities and keywords based on the pre-trained semantic network, and then obtains the expanded user request information, which includes the original request of the user, hyponyms, and synonyms; Semantic construction unit: The system constructs an initial semantic graph based on the user's original request. The core semantic layer of the initial semantic graph contains the keywords and entities of the user's original request. Semantic expansion unit: performs a first expansion on the initial semantic graph based on hyponyms, synonyms, and at the same time, obtains concepts associated with the user's original request based on a preset semantic library and performs a second expansion on the initial semantic graph; The candidate determination unit: performs reasoning on the multi-level semantic graph based on a preset reasoning algorithm, thereby obtaining the correlation between the user's original request and several preset data sources, and obtaining several preset data sources with a correlation greater than the preset correlation as candidate data sources; Data source matching unit: obtains the user's historical request data, and prioritizes the candidate data sources based on the preset intelligent algorithm and the user's historical request data, and then selects several candidate data sources with priorities greater than a preset priority threshold as several target data sources requested by the user; Data acquisition unit: matches the keywords in the target data source based on the user's original request, and then acquires the target data corresponding to each target data source.
[0008] The present invention provides an information security management system based on big data, a data fusion module, comprising: Conflict handling unit: performs initial data source conflict detection on all target data, and performs conflict handling if a data source conflict occurs; Preprocessing unit: obtains the preprocessing method corresponding to each target data based on the preset data source-preprocessing method data table, and preprocesses all target data after conflict processing; Data fusion unit: fuses the preprocessed target data based on a preset method to obtain fused information data.
[0009] The present invention provides an information security management system based on big data, a conflict processing unit, comprising: Data conflict detection subunit: performs initial data source conflict detection on all target data and locates the conflict point of each target data; Parameter acquisition subunit: obtains relevant parameters of the data source of all target data; Priority coverage subunit: Determines the coverage priority of each data source based on the priority of each data source and the error rate of the data source: ; in, is the coverage priority of the i-th data source, is the priority of the i-th data source, is the priority adjustment coefficient of the i-th data source, is the weight coefficient corresponding to the priority of the i-th data source, Score the quality of the i-th data source, and , is the error rate of the ith data source, is the score of the i-th data source on the j-th reliability factor, is the weight of the jth reliable factor, m is the number of reliable factors, is the redundancy coefficient of the ith data source, is the preset time decay factor, is the current time, is the update time of the i-th data source; Conflict handling subunit: performs overwriting processing on the conflicting points of the target data based on the overwriting priorities of all data sources and preset priority rules.
[0010] The present invention provides an information security management system based on big data, a conflict handling subunit, comprising: Prioritization block: sorts the conflict points of each target data based on the coverage priority of the data source, and checks the data items of the conflict points field by field; Data overwriting block: For conflicting fields, the conflicting fields of the target data of the data source with the highest priority are selected to overwrite the conflicting fields of the target data of other data sources that have caused the conflict. If the priorities are the same, the update time of the data sources is compared, and the target data of the latest data source is selected to overwrite.
[0011] The present invention provides an information security management system based on big data, a policy determination module, comprising: Information extraction unit: performs correlation analysis on fused information data based on a preset learning algorithm, and extracts key information from the fused information data; Graph construction unit: constructs a multi-dimensional data relationship graph based on the association analysis results of the fused information data and the key information of the fused information data combined with the preset graph technology; Attribute acquisition unit: analyzes the multi-dimensional data relationship graph to determine the node attributes of each node; Node determination unit: Analyze the node attributes of each node to determine several key nodes in the multi-dimensional data relationship graph; Path determination unit: determining important data paths in a multi-dimensional data relationship graph based on paths between several key nodes in the multi-dimensional data relationship graph; Policy acquisition unit: Perform risk level assessment on key nodes and important data paths, and determine security policies based on the risk levels of key nodes and important data paths determined by the risk level assessment and the preset risk level-security policy database.
[0012] The present invention provides an information security management system based on big data, an information extraction unit, comprising: The first analysis sub-element: performing a first multi-dimensional correlation analysis on the data source to which the data in the fusion information data belongs; The second analysis subunit: performs a second multi-dimensional correlation analysis on the data in the fusion information data; The third analysis subunit: extracts data features of the fusion information data and performs a third multi-dimensional correlation analysis on the data features; The fourth analysis subunit extracts key information from the fused information data, and determines association information between the key information in the fused information data based on the results of the first multi-dimensional association analysis, the second multi-dimensional association analysis, and the third multi-dimensional association analysis.
[0013] The present invention provides an information security management system based on big data, and the security strategy includes: data detection strategy, data response strategy, security access strategy and data encryption strategy.
[0014] Compared with the prior art, the present invention has the following beneficial effects: Through data positioning, fusion and multi-dimensional relationship analysis, dynamic security strategies tailored to user needs are constructed to achieve accurate encryption and management of target information data, effectively integrate multi-source data, and generate multi-dimensional security strategies based on big data analysis, which greatly improves the flexibility, response speed and overall security of information security protection, and solves the problem of lack of data fusion and dynamic security strategies in existing technologies. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. 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 paying creative work.
[0016] Figure 1 It is a structural diagram of an information security management system based on big data provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0017] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings 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. Example
[0018] The embodiment of the present invention provides an information security management system based on big data, such as Figure 1 As shown, including: Data location module: locates target information data from several preset data sources based on user needs; Data fusion module: performs data fusion on the target information data obtained by positioning to obtain fused information data; Strategy determination module: Analyzes the fusion information data based on the preset algorithm and constructs a multi-dimensional data relationship map to determine the security strategy corresponding to the user's needs; Data management module: Encrypt target information data based on security policies corresponding to user needs and manage it based on preset technologies.
[0019] In this embodiment, user requirements are specific requirements or expectations of users for the system, which usually include requirements for data security, access control requirements, data usage permissions, etc. For example, a bank may require that customer data be highly encrypted and only authorized employees be allowed to access; In this embodiment, the preset data source is a plurality of data sources that are pre-configured in the system and can be used to obtain information. They can be different databases, cloud storage, API interfaces, etc. For example, the preset data source can be the company's internal database, a third-party data service platform, or the storage of a cloud computing service provider.
[0020] In this embodiment, the target information data refers to specific information data selected from a preset data source according to user needs. For example, in a bank case, the target information data may be a customer's account information, transaction records, etc.
[0021] In this embodiment, data fusion is to integrate, clean and remove duplicate target information data from different data sources to make it a unified and complete data set. For example, user transaction data collected from multiple systems can be fused to eliminate duplicate records and unify the format; In this embodiment, the fused information data is a data set that has undergone data fusion processing and includes all relevant information from different data sources. For example, on an e-commerce platform, the fused information data may include a unified view of a user's purchase history, browsing history, and evaluation data.
[0022] In this embodiment, the preset algorithm is an algorithm model pre-defined in the system for analyzing fusion information data, involving machine learning, image analysis, relationship network analysis and other methods. For example, a bank may use a machine learning algorithm to analyze customer data and identify possible fraudulent behavior; In this embodiment, the multi-dimensional data relationship map is a multi-dimensional graphical relationship network based on fused information data generated by the system through a preset algorithm, which shows the complex relationship between data. For example, in an e-commerce platform, the map can show the purchase relationship between users and products, the correlation between products, etc. In this embodiment, the security policy is based on user needs and data relationship maps. The system generates protection policies for different data scenarios, covering data encryption, access permission control, etc. For example, in a banking system, the security policy may include high-intensity encryption of sensitive data and require two-factor authentication for access.
[0023] In this embodiment, the preset technology is a series of technical means pre-configured by the system for managing and protecting data, including encryption algorithms, storage management, identity authentication technology, etc. For example, the AES encryption algorithm can be used as a preset technology to encrypt user account information.
[0024] The beneficial effects of the above technical solution are: through data positioning, fusion and multi-dimensional relationship analysis, dynamic security strategies tailored to user needs are constructed, accurate encryption and management of target information data are achieved, multi-source data is effectively integrated, and multi-dimensional security strategies are generated based on big data analysis, which greatly improves the flexibility, response speed and overall security of information security protection, and solves the problem of lack of data fusion and dynamic security strategies in existing technologies. Example
[0025] The embodiment of the present invention provides an information security management system based on big data, a data positioning module, including: Request processing unit: pre-processes the original request of the user, obtains the preliminary semantic information of key entities, keywords and sentence structure in the original request of the user, and semantically expands the extracted key entities and keywords based on the pre-trained semantic network, and then obtains the expanded user request information, which includes the original request of the user, hyponyms, and synonyms; Semantic construction unit: The system constructs an initial semantic graph based on the user's original request. The core semantic layer of the initial semantic graph contains the keywords and entities of the user's original request. Semantic expansion unit: performs a first expansion on the initial semantic graph based on hyponyms, synonyms, and at the same time, obtains concepts associated with the user's original request based on a preset semantic library and performs a second expansion on the initial semantic graph; The candidate determination unit: performs reasoning on the multi-level semantic graph based on a preset reasoning algorithm, thereby obtaining the correlation between the user's original request and several preset data sources, and obtaining several preset data sources with a correlation greater than the preset correlation as candidate data sources; Data source matching unit: obtains the user's historical request data, and prioritizes the candidate data sources based on the preset intelligent algorithm and the user's historical request data, and then selects several candidate data sources with priorities greater than a preset priority threshold as several target data sources requested by the user; Data acquisition unit: matches the keywords in the target data source based on the user's original request, and then acquires the target data corresponding to each target data source.
[0026] In this embodiment, the original request of the user is the initial request issued by the user in the system, which is usually an unprocessed natural language expression. For example, the original request submitted by the user in the bank system may be "query my bank transaction records for the past three months"; In this embodiment, the key entity is a core object or subject extracted from the user's original request, which is a noun or a noun phrase. For example, in "query my bank transaction records for the past three months", the key entity is "bank transaction records"; In this embodiment, the keywords are the core words in the original request of the user, which can reflect the main content of the request. For example, in "query my bank transaction records for the past three months", the keywords may be "query", "three months" and "transaction records"; In this embodiment, the sentence structure refers to the grammatical structure of the user's original request. The system can understand the semantic relationship in the request by analyzing the sentence structure. For example, in "Query my bank transaction records for the past three months", the sentence structure is "action (query) + time (past three months) + object (bank transaction records)"; In this embodiment, the preliminary semantic information is a preliminary understanding based on the user's original request, including key entities, keywords and sentence structures. The preliminary semantic information helps the system understand the basic meaning of the user's request. For example, for "query my bank transaction records for the past three months", the preliminary semantic information can be: action "query", object "bank transaction records", time "past three months"; In this embodiment, the pre-trained semantic network is a semantic association model that is trained by the system in advance through big data, and is used to help the system understand the vocabulary and semantic relationship in the user request. For example, the semantic network can take "query" and "get" as similar semantic associations to understand that the user wants to view data; In this embodiment, semantic expansion is based on the semantic network, and the key words in the user request are expanded into hyponyms and synonyms to obtain more relevant information. For example, for "query bank transaction records", the system can expand related words such as "obtain", "bank account information", and "transaction details"; In this embodiment, the expanded user request information includes the user's original request and its expanded synonyms, hyponyms, and hyponyms. For example, "query bank transaction records" can be expanded to "obtain account details", "view transaction records", etc.; In this embodiment, a hypernym refers to a word with a broader meaning, and a hyponym refers to a more specific word. For example, "vehicle" is a hypernym of "car", and "sedan" is a hyponym of "car"; In this embodiment, synonyms are words with the same or similar meanings. For example, "query" and "retrieval" can be synonyms, and in semantic expansion, the system will treat them as the same request.
[0027] In this embodiment, the initial semantic graph is a graphical semantic representation constructed by the system based on the user request, showing the keywords and key entities in the request and their interrelationships. For example, "query" and "bank transaction records" in "query bank transaction records" will form a basic semantic graph; In this embodiment, the core semantic layer is the core part of the semantic graph, which contains the most important keywords and entities in the user's original request. For example, in a "query bank transaction records" request, the core semantic layer contains "query" and "bank transaction records".
[0028] In this embodiment, the preset reasoning algorithm is an algorithm pre-configured by the system for reasoning and association analysis from a multi-level semantic graph. For example, the system may use a rule-based reasoning algorithm or a machine learning algorithm to determine the correlation between the user's request and certain data sources, thereby recommending a suitable data source. Suppose the user sends an original request on the e-commerce platform: "Find the electronic products I purchased in the past year", the following steps will be taken: the key entity is "electronic products", the keywords are "find", "past year", and "purchase", and semantic expansion is performed based on the pre-trained semantic network to expand "electronic products" into related words such as "mobile phone", "computer", and "tablet". The core semantic layer of the semantic graph contains "find" and "electronic products", and then the system matches the data sources related to the user request, such as order records and purchase history, through the preset reasoning algorithm.
[0029] The beneficial effects of the above technical solution are: by introducing semantic networks and semantic extension technologies, the user's original request is innovatively semantically expanded and a multi-level semantic graph is constructed, which improves the accurate understanding of user requests and the matching effect of data sources. Combined with preset reasoning algorithms and intelligent algorithms, the system can accurately locate and sort data sources, thereby improving the accuracy and efficiency of data acquisition. Example
[0030] The embodiment of the present invention provides an information security management system based on big data, a data fusion module, including: Conflict handling unit: performs initial data source conflict detection on all target data, and performs conflict handling if a data source conflict occurs; Preprocessing unit: obtains the preprocessing method corresponding to each target data based on the preset data source-preprocessing method data table, and preprocesses all target data after conflict processing; Data fusion unit: fuses the preprocessed target data based on a preset method to obtain fused information data.
[0031] In this embodiment, initial conflict detection means that when the system processes target data from different data sources, it first checks whether there are conflicts or inconsistencies between the data. Such conflicts may include issues such as data format and content differences. For example, two data sources record different address information of the same customer, which is a manifestation of data conflict. For example, in a system, the customer information from database A shows that the customer address is "Beijing", while the customer address from database B is "Shanghai". At this time, the system will find that the two addresses conflict through initial conflict detection; In this embodiment, the data source conflict refers to the situation when there are differences, inconsistencies or contradictions between the data provided by two or more data sources. This conflict may be caused by differences in time, data format, content, etc. between different data sources. For example, the date format of one data source is "YYYY-MM-DD", while the date format of another data source is "MM / DD / YYYY", which is a type of data source conflict. For example, in a medical system, the patient's date of birth provided by one data source is "1990-12-01", while the date format of another data source is "01 / 12 / 1990", and the system will detect that these data formats conflict; In this embodiment, conflict handling is when the system detects conflicts between data sources, measures are taken to resolve these conflicts to ensure the final data consistency and accuracy. Conflict handling can be completed by selecting a priority data source, merging data, or applying preset rules. For example, a more reliable data source can be selected based on the credibility of the source of the data. If one system shows that a customer's address is "Beijing" and the other system shows it as "Shanghai" in two systems, the conflict handling unit can select "Beijing" as the final address through historical data analysis or based on a more reliable data source.
[0032] In this embodiment, the preset data source-preprocessing method data table is a mapping table pre-defined in the system, which records the preprocessing methods applicable to different data sources. Each data source may require a different preprocessing method (such as deduplication, format conversion, data cleaning, etc.). This data table provides rules for how to process target data from different data sources. The "preset data source-preprocessing method data table" may record the following information: Data source A: date format conversion, data source B: deduplication processing, data source C: data cleaning.
[0033] In this embodiment, the preset method is a fixed method or algorithm for data fusion defined in the system, which is used to unify and comprehensively process target data from different data sources. The preset method includes specific operation modes such as weighted average, data optimization, and merging strategy. For example, in a weather data management system, there may be temperature data from three different data sources. The preset method can be to take the weighted average of the three data sources, or select the most credible data as the final temperature value according to the weight of certain data sources.
[0034] In this example, assume that a logistics company has order information from different data sources. Through initial conflict detection, the system finds that the shipping addresses of orders in different data sources are inconsistent (for example, one is displayed as "New York" and the other is displayed as "London"). This situation is a data source conflict. Through conflict resolution, based on historical data and the credibility of the data source, it is determined to use "New York" as the final shipping address. According to the preset data source-preprocessing method data table, each data source is preprocessed accordingly, such as format conversion or data cleaning. Finally, these processed data are merged through the preset method to form consistent order information.
[0035] The beneficial effects of the above technical solution are: the conflict and fusion problems of multiple data sources are solved through a multi-step process of data source conflict handling, preprocessing and data fusion. Based on the preset data tables and methods, the standardization and consistency of data processing are ensured, the efficiency and accuracy of data fusion are improved, the high-quality output of information data is guaranteed, and the overall performance and reliability of the information security management system are improved. Example
[0036] The embodiment of the present invention provides an information security management system based on big data, a conflict processing unit, including: Data conflict detection subunit: performs initial data source conflict detection on all target data and locates the conflict point of each target data; Parameter acquisition subunit: obtains relevant parameters of the data source of all target data; Priority coverage subunit: Determines the coverage priority of each data source based on the priority of each data source and the error rate of the data source: ; in, is the coverage priority of the i-th data source, is the priority of the i-th data source, is the priority adjustment coefficient of the i-th data source, is the weight coefficient corresponding to the priority of the i-th data source, Score the quality of the i-th data source, and , is the error rate of the ith data source, is the score of the i-th data source on the j-th reliability factor, is the weight of the jth reliable factor, m is the number of reliable factors, is the redundancy coefficient of the ith data source, is the preset time decay factor, is the current time, is the update time of the i-th data source; Conflict handling subunit: performs overwriting processing on the conflicting points of the target data based on the overwriting priorities of all data sources and preset priority rules.
[0037] In this embodiment, the conflict point of each target data refers to the specific location where the system detects inconsistency or conflict when comparing target data from different data sources. It is because certain fields or attributes in the data are different, and these conflict points need to be located for subsequent processing. Suppose a customer's order data comes from two different data sources, the order time recorded in the first data source is "2024-09-01", and the time recorded in the second data source is "2024-08-31", then the order time field is the conflict point.
[0038] In this embodiment, the error rate of a data source refers to the frequency of generating erroneous data in the past processing and verification process of a data source. The higher the error rate, the lower the reliability of the data source. The error rate can be obtained by analyzing the historical data of the system and is usually expressed as a percentage. Assuming that data source A in a certain inventory management system has 5 errors in the past 100 records, the error rate of data source A is 5%; In this embodiment, the coverage priority of the data source is a comprehensive score determined based on multiple factors of the data source (such as priority, error rate, quality score, etc.), which is used to determine which data source should give priority to overwriting the conflicting data when a data conflict occurs. Considering the reliability and data quality of each data source, if data source A has a higher priority, a lower error rate, and a higher quality score, its coverage priority will also be higher. For example, when dealing with a conflict between two data sources, the system will give priority to using the value of data source A to resolve the conflict; In this embodiment, the score on the reliability factor refers to the system scoring each data source according to certain specific standards or factors (such as data accuracy, timeliness, reliability of the source, etc.). Different factors will be weighted according to weights to evaluate the overall reliability of the data source. Assume that the system evaluates the reliability of data source B by accuracy, timeliness and data integrity. If the accuracy score of data source B is 90 points, the timeliness score is 85 points, and the data integrity score is 80 points, and the weights of these factors are 0.5, 0.3, and 0.2 respectively, then the comprehensive score of data source B is 90 × 0.5 + 85 × 0.3 + 80 × 0.2 = 87.5; In this embodiment, the redundancy coefficient of a data source indicates the proportion of redundant data in a data source. Redundant data refers to repeated or unnecessary data. The higher the redundancy coefficient, the more repeated information there is in the data source, which will affect its quality score. If 20 out of 100 records in data source C are repeated, its redundancy coefficient is 20%. A higher redundancy coefficient usually means that the priority of the data source may be reduced; In this embodiment, the preset time decay factor indicates that as time goes by, the information of some data sources may no longer be accurate or relevant. This time correlation is measured by a decay coefficient. The time decay factor helps the system evaluate the timeliness of data and ensures that old data will not have too much impact on current results when making decisions. Assume that the last update time of a data source D was 1 year ago, and the preset time decay factor of the system is 0.8 (per year). If the quality score of the data source was 90 points 1 year ago, then after 1 year of time decay, the current effective score of the data source is 90 × 0.8 = 72 points; In this embodiment, it is assumed that an e-commerce platform has multiple data sources (such as inventory, order management, customer information, etc.). When synchronizing order data, it is found that there is a conflict in the order status of different data sources. The following steps are taken to handle it: first, locate the conflict point, for example, the order status field is "shipped", while another data source shows "pending shipment", which is the conflict point of the target data, and calculate the error rate of each data source. For example, the error rate of the inventory management system is low (1%), while the error rate of the order management system is high (5%). According to multiple factors, such as error rate, data quality, update time, etc., the coverage priority of each data source is calculated to determine which data source has more reliable information. Each data source is evaluated based on the score on the reliability factor (such as accuracy, timeliness, etc.), and the final comprehensive score is calculated based on the weights of these factors. If a data source has a high redundancy coefficient, such as a large amount of duplicate order data, the system will lower the priority of the data source, and evaluate the timeliness of the data source in combination with the preset time decay factor to ensure that the most recently updated and reliable data source is selected first to resolve the conflict.
[0039] The beneficial effects of the above technical solution are: by introducing multi-dimensional parameters such as coverage priority, error rate, quality score, redundancy coefficient and time decay factor, the priority of the data source is determined, and accurate coverage processing of target data conflict points is achieved, thereby ensuring the accuracy and reliability of data selection, improving the conflict processing efficiency and data quality of the information security management system, and enhancing the intelligence and dynamic adaptability of the system.
[0040] Example 5 The embodiment of the present invention provides an information security management system based on big data, and a conflict handling subunit, including: Prioritization block: sorts the conflict points of each target data based on the coverage priority of the data source, and checks the data items of the conflict points field by field; Data overwriting block: For conflicting fields, the conflicting fields of the target data of the data source with the highest priority are selected to overwrite the conflicting fields of the target data of other data sources that have caused the conflict. If the priorities are the same, the update time of the data sources is compared, and the target data of the latest data source is selected to overwrite.
[0041] In this embodiment, checking the data items of conflicting points field by field means that when the system processes multiple data sources, it checks each field (data item) in detail one by one to find out whether there is a conflict in each field. Checking field by field means that the system will not only check whether the overall data is conflicting, but also go deep into each specific field, such as name, address, date, etc., to check whether the data in these fields are consistent. Suppose there is a customer information from three different data sources, including the fields "name", "address" and "contact number". The system will first check each field one by one, for example: name field: all consistent, no conflict, address field: the first data source shows "Beijing", the second shows "Haidian District, Beijing", and the third shows "Beijing", there is a conflict, contact number field: two data sources are the same, the other is different, there is a conflict, the system will process these conflicting fields (address, contact number) one by one.
[0042] In this embodiment, the update time of the data source refers to the timestamp of the last update of the target data in the data source. The update time reflects the timeliness of the data source. When a data conflict occurs, the system will give priority to the data source with a more recent update time, because the newer data often represents the latest situation. The update time can be used to determine which data source should be trusted more under the same priority. Suppose there are two data sources A and B in the system, providing the order status of the same customer: the order status of data source A is "shipped" and the update time is "2024-09-01 10:00", and the order status of data source B is "pending shipment" and the update time is "2024-09-01 12:00". Although the two data sources have the same priority, the system will choose B's data to overwrite because its update time is more recent and reflects the latest status.
[0043] The beneficial effects of the above technical solution are: through the innovative method of priority sorting and data coverage, the conflict problem between multiple data sources is effectively solved. Based on the coverage priority and data source update time, the optimal data source is intelligently selected to cover the conflicting fields, ensuring data consistency and accuracy, improving data processing efficiency, avoiding data confusion or errors caused by conflicts, and greatly improving the data reliability of the information security management system.
[0044] Example 6 The embodiment of the present invention provides an information security management system based on big data, a policy determination module, including: Information extraction unit: performs correlation analysis on fused information data based on a preset learning algorithm, and extracts key information from the fused information data; Graph construction unit: constructs a multi-dimensional data relationship graph based on the association analysis results of the fused information data and the key information of the fused information data combined with the preset graph technology; Attribute acquisition unit: analyzes the multi-dimensional data relationship graph to determine the node attributes of each node; Node determination unit: Analyze the node attributes of each node to determine several key nodes in the multi-dimensional data relationship graph; Path determination unit: determining important data paths in a multi-dimensional data relationship graph based on paths between several key nodes in the multi-dimensional data relationship graph; Policy acquisition unit: Perform risk level assessment on key nodes and important data paths, and determine security policies based on the risk levels of key nodes and important data paths determined by the risk level assessment and the preset risk level-security policy database.
[0045] In this embodiment, the preset learning algorithm refers to a machine learning algorithm pre-set in the system for analyzing and processing data. The algorithm can learn from a large amount of fused information data and identify the association between data to help the system extract important information. These learning algorithms can include different types such as classification, clustering, and regression, which are selected according to specific needs. For example, in a financial system, a preset learning algorithm (such as a decision tree algorithm) is used to analyze the transaction data of different customers to find out which transaction behaviors may be associated with potential risk operations, thereby extracting the characteristics of high-risk transactions; In this embodiment, key information refers to data that is very important to the business or system and needs to be focused on, which is extracted through data analysis. It is usually screened out from a large amount of data and has important business significance. For example, in a medical and health system, key information may be a patient's key diagnostic results, allergy information, or emergency contact information. This information is very important for subsequent diagnosis and treatment.
[0046] In this embodiment, the preset graph technology refers to a graph construction and analysis method predefined in the system, which is used to represent different data nodes and the relationships between them as a graph structure. This technology can be a graph structure based on a graph database, or it can be a graph constructed based on a specific algorithm. For example, in a social network analysis system, a user relationship network is constructed using graph technology, where nodes represent users and edges represent social interactions between users. Through this graph, the user's social circle and influence can be analyzed.
[0047] In this embodiment, each node represents a specific data entity, and its node attributes are relevant characteristics or descriptive information of the entity. The node attributes include various descriptive information such as the entity's name, type, timestamp, status, etc. For example, in a supply chain system, a node can represent a supplier, and the node attributes may include the supplier's name, geographic location, delivery cycle, credit rating, and other information.
[0048] In this embodiment, the node attributes of each node are analyzed to determine several key nodes in the multi-dimensional data relationship map. By analyzing the attributes of each node, it is determined which nodes play a vital role in the data relationship map, which are called key nodes. Key nodes are usually important points for data circulation, decision-making, or information aggregation in the map. For example, in a network security system, a key node may be a server or gateway in the network. After analyzing the attributes of these nodes (such as traffic size, number of visits, risk level, etc.), the system identifies these servers as key nodes because they centrally process a large number of network requests.
[0049] In this embodiment, the important data paths in the multidimensional data relationship map are determined based on the paths between several key nodes in the multidimensional data relationship map. After the system constructs the multidimensional data relationship map, it determines which paths are most important for the circulation and processing of data by analyzing the connection paths between the key nodes in the map. These paths reflect the transmission routes of data in the system. For example, in a logistics system, an important data path in the data relationship map may be "warehouse→distribution center→customer". By analyzing the node relationship between the warehouse and the distribution center, the system determines that this path is crucial to the efficiency of goods delivery.
[0050] In this embodiment, risk level assessment refers to the system assessing the security risks of key nodes and data paths according to certain preset standards. The assessment content may include security vulnerabilities of nodes, security of data transmission, external threats, etc. The results are used to measure the risk levels of these nodes and paths. For example, in a banking system, key nodes such as payment gateways and customer information databases will be assessed for their security. If the payment gateway has unpatched vulnerabilities, its risk level may be assessed as "high"; In this embodiment, the preset risk level-security policy database is a database containing different risk levels and corresponding security policies. When the system obtains the risk level of a node or path through risk assessment, the corresponding security policy can be found in the database and applied. If the system detects that the risk level of a key server is assessed as "high", the system can find the corresponding policy from the risk level-security policy database, such as immediately enabling data encryption, restricting external access and other measures to reduce risks.
[0051] The beneficial effects of the above technical solution are: through preset learning algorithms and graph technology, a multi-dimensional data relationship map is innovatively constructed, and combined with node attribute analysis and path determination, key nodes and important data paths are effectively identified. Based on the risk assessment of key nodes and paths, the system accurately determines security strategies, improves the automation and intelligence level of information security management, and significantly enhances the warning and response capabilities for potential risks.
[0052] Example 7 The embodiment of the present invention provides an information security management system based on big data, an information extraction unit, including: The first analysis sub-element: performing a first multi-dimensional correlation analysis on the data source to which the data in the fusion information data belongs; The second analysis subunit: performs a second multi-dimensional correlation analysis on the data in the fusion information data; The third analysis subunit: extracts data features of the fusion information data and performs a third multi-dimensional correlation analysis on the data features; The fourth analysis subunit extracts key information from the fused information data, and determines association information between the key information in the fused information data based on the results of the first multi-dimensional association analysis, the second multi-dimensional association analysis, and the third multi-dimensional association analysis.
[0053] In this embodiment, the first multi-dimensional association analysis is performed. The so-called "multi-dimensional" means that multiple data features or attributes are taken into account in the analysis, so as to explore the relationship between data sources from different angles. The purpose of this process is to discover the implicit connections or patterns that may exist between data sources. Suppose there are three different data sources that provide different information about users, such as data from a social platform, data from an e-commerce platform, and transaction data from a bank. The first multi-dimensional association analysis may analyze the relationship between these data sources, for example, the correlation between user behavior on a social platform and shopping behavior on an e-commerce platform, and whether bank transaction data is consistent with shopping records on an e-commerce platform.
[0054] In this embodiment, the second multi-dimensional association analysis is an association analysis of the data itself in the fused information data, rather than an analysis of the data source. This type of analysis is based on the content of the data and its different characteristics, and explores the association between different data items or records. For example, in a medical data analysis system, the medical records, drug use records, and test results of different patients are integrated. The second multi-dimensional association analysis will focus on the association between the content of these data, such as whether the use of a certain drug is related to a certain test result, and the association between the patient's medical history and the current diagnosis result.
[0055] In this embodiment, the third multi-dimensional association analysis is an analysis of the data features in the fused information data. The data features refer to various attributes of the data, such as fields in structured data (such as name, address, timestamp), or word frequency, topic model, etc. in unstructured data. This step aims to analyze the intrinsic attributes and rules of the data through feature extraction. For example, in an email filtering system, the fused information data may include data such as email content, time, sender, and receiver. Based on the email's word frequency, timestamp, sender location, and other features, it is analyzed whether the email is a spam or phishing email.
[0056] In this embodiment, determining the association information between key information in the fusion information data based on the results of the first multi-dimensional association analysis, the second multi-dimensional association analysis, and the third multi-dimensional association analysis is to comprehensively analyze the three association analysis results, extract key association information from them, and help the system better understand the complex relationship between the data. By summarizing the analysis results from the data source, the data itself, and the data characteristics, the system can determine the specific association between key data items, and then provide a basis for risk assessment and security decision-making. For example, in a financial risk monitoring system, the system first finds that there is an association between a user's transaction behavior between multiple financial platforms through the first multi-dimensional association analysis; then, the second multi-dimensional association analysis reveals the user's specific transaction mode, such as high-frequency trading or fund transfer characteristics; the third multi-dimensional association analysis further determines the user's suspicious behavior by analyzing the time, amount, frequency, and other characteristics of the transaction. Finally, by combining these analysis results, the system determines that there is an association between the user's behavior and a certain transaction mode, thereby extracting key information.
[0057] The beneficial effects of the above technical solution are: through the multi-dimensional correlation analysis method, the source, characteristics and key information of the fused information data are innovatively analyzed at multiple levels to form a complete correlation information chain, which improves the accuracy and comprehensiveness of information extraction, and can automatically identify and analyze the complex relationship between key information, greatly improving the information processing efficiency and decision-making support capabilities in the big data environment.
[0058] Example 8 The embodiment of the present invention provides an information security management system based on big data, and the security strategy includes: data detection strategy, data response strategy, security access strategy and data encryption strategy.
[0059] In this embodiment, the data detection strategy refers to the system automatically scanning, monitoring and analyzing data through preset rules, algorithms or technologies to identify potential risks, abnormal behaviors or illegal operations. It is the first line of defense of the security management system and is used to detect possible problems in advance. For example, in an enterprise's network security system, the data detection strategy can be set to regularly scan network traffic and user activity logs to detect abnormal access behaviors. For example, the system detects that a user account accesses confidential data abnormally frequently at night, while the user's usual access pattern is not the case. This abnormal behavior will be marked as a potential risk. The data detection strategy can also be used to identify malware. For example, the system will detect whether an email attachment or downloaded file contains a known malicious code signature; In this embodiment, the data response strategy refers to the automated or manual response measures taken by the system after detecting a potential security threat or abnormal situation. These measures may include blocking suspicious behavior, issuing alarms, locking accounts, isolating infected systems or devices, etc. The goal of the response strategy is to reduce the harm of security incidents to systems or data. If a bank's security system detects that a user attempts to enter an incorrect password multiple times, triggering a data response strategy, the system will automatically lock the user's account and notify the security administrator. For example, when the network security system detects that malicious traffic attempts to penetrate the enterprise network, the data response strategy will automatically enable firewall rules to prevent further access from malicious IP addresses; In this embodiment, the security access policy refers to the system controlling the access rights to data and resources to ensure that only authorized users or systems can access specific data or functions. The policy can flexibly manage access rights based on multiple factors such as user identity, role, time, device type, etc., thereby ensuring the security of sensitive information. For example, in a medical system, the security access policy may stipulate that only doctors and authorized medical staff can access the patient's complete medical record information, while administrative staff can only view limited non-sensitive information, such as the patient's appointment information. The system can set a two-factor authentication policy, such as employees need to log in with a username and password, and also need to use a mobile phone verification code to further verify their identity, thereby improving security; In this embodiment, the data encryption strategy refers to the use of encryption technology to protect data during data storage, transmission or processing to prevent unauthorized users from accessing or interpreting the data content. The encryption strategy ensures that even if the data is stolen or accidentally leaked, the attacker cannot directly interpret the data and must decrypt it to obtain meaningful information. For example, in an e-commerce platform, the data encryption strategy is used to encrypt the user's payment information, such as credit card numbers and personal identity information. Even if hackers steal this data, they cannot directly use this information because the data has been encrypted. For another example, the end-to-end encryption strategy used by the email system ensures that only the sender and the receiver can read the email content, and even the email server cannot decrypt the content.
[0060] The beneficial effects of the above technical solution are: through the organic combination of data detection, data response, secure access and data encryption strategies, all-round information security management is achieved, which can not only monitor and respond to potential security threats in real time, but also effectively protect sensitive information through sophisticated permission control and data encryption, thereby improving data security and system protection capabilities, and significantly reducing the possibility of information leakage and risk events.
[0061] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.
[0062] 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, 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 described in each embodiment or some parts of the embodiments.
[0063] 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. An information security management system based on big data, characterized in that: include: Data location module: locates target information data from several preset data sources based on user needs; Data fusion module: performs data fusion on the target information data obtained by positioning to obtain fused information data; Strategy determination module: Analyzes the fusion information data based on the preset algorithm and constructs a multi-dimensional data relationship map to determine the security strategy corresponding to the user's needs; Data management module: Encrypt target information data based on security policies corresponding to user needs and manage it based on preset technologies.
2. According to the information security management system based on big data in claim 1, it is characterized in that: Data location module, including: Request processing unit: pre-processes the original request of the user, obtains the preliminary semantic information of key entities, keywords and sentence structure in the original request of the user, and semantically expands the extracted key entities and keywords based on the pre-trained semantic network, and then obtains the expanded user request information, which includes the original request of the user, hyponyms, and synonyms; Semantic construction unit: The system constructs an initial semantic graph based on the user's original request. The core semantic layer of the initial semantic graph contains the keywords and entities of the user's original request. Semantic expansion unit: performs a first expansion on the initial semantic graph based on hyponyms, synonyms, and at the same time, obtains concepts associated with the user's original request based on a preset semantic library and performs a second expansion on the initial semantic graph; The candidate determination unit: performs reasoning on the multi-level semantic graph based on a preset reasoning algorithm, thereby obtaining the correlation between the user's original request and several preset data sources, and obtaining several preset data sources with a correlation greater than the preset correlation as candidate data sources; Data source matching unit: obtains the user's historical request data, and prioritizes the candidate data sources based on the preset intelligent algorithm and the user's historical request data, and then selects several candidate data sources with priorities greater than a preset priority threshold as several target data sources requested by the user; Data acquisition unit: matches the keywords in the target data source based on the user's original request, and then acquires the target data corresponding to each target data source.
3. The information security management system based on big data according to claim 1 is characterized in that: Data fusion module, including: Conflict handling unit: performs initial data source conflict detection on all target data, and performs conflict handling if a data source conflict occurs; Preprocessing unit: obtains the preprocessing method corresponding to each target data based on the preset data source-preprocessing method data table, and preprocesses all target data after conflict processing; Data fusion unit: fuses the preprocessed target data based on a preset method to obtain fused information data.
4. The information security management system based on big data according to claim 3 is characterized in that: Conflict handling unit, including: Data conflict detection subunit: performs initial data source conflict detection on all target data and locates the conflict point of each target data; Parameter acquisition subunit: obtains relevant parameters of the data source of all target data; Priority coverage subunit: Determines the coverage priority of each data source based on the priority of each data source and the error rate of the data source: ; in, is the coverage priority of the i-th data source, is the priority of the i-th data source, is the priority adjustment coefficient of the i-th data source, is the weight coefficient corresponding to the priority of the i-th data source, Score the quality of the i-th data source, and , is the error rate of the ith data source, is the score of the i-th data source on the j-th reliability factor, is the weight of the jth reliable factor, m is the number of reliable factors, is the redundancy coefficient of the ith data source, is the preset time decay factor, is the current time, is the update time of the i-th data source; Conflict handling subunit: performs overwriting processing on the conflicting points of the target data based on the overwriting priorities of all data sources and preset priority rules.
5. The information security management system based on big data according to claim 4 is characterized in that: Conflict handling subunit, including: Prioritization block: sorts the conflict points of each target data based on the coverage priority of the data source, and checks the data items of the conflict points field by field; Data overwriting block: For conflicting fields, the conflicting fields of the target data of the data source with the highest priority are selected to overwrite the conflicting fields of the target data of other data sources that have caused the conflict. If the priorities are the same, the update time of the data sources is compared, and the target data of the latest data source is selected to overwrite.
6. The information security management system based on big data according to claim 3 is characterized in that: The strategy determination module includes: Information extraction unit: performs correlation analysis on fused information data based on a preset learning algorithm, and extracts key information from the fused information data; Graph construction unit: constructs a multi-dimensional data relationship graph based on the association analysis results of the fused information data and the key information of the fused information data combined with the preset graph technology; Attribute acquisition unit: analyzes the multi-dimensional data relationship graph to determine the node attributes of each node; Node determination unit: Analyze the node attributes of each node to determine several key nodes in the multi-dimensional data relationship graph; Path determination unit: determining important data paths in a multi-dimensional data relationship graph based on paths between several key nodes in the multi-dimensional data relationship graph; Policy acquisition unit: Perform risk level assessment on key nodes and important data paths, and determine security policies based on the risk levels of key nodes and important data paths determined by the risk level assessment and the preset risk level-security policy database.
7. The information security management system based on big data according to claim 6 is characterized in that: An information extraction unit, comprising: The first analysis sub-element: performing a first multi-dimensional correlation analysis on the data source to which the data in the fusion information data belongs; The second analysis subunit: performs a second multi-dimensional correlation analysis on the data in the fusion information data; The third analysis subunit: extracts data features of the fusion information data and performs a third multi-dimensional correlation analysis on the data features; The fourth analysis subunit extracts key information from the fused information data, and determines association information between the key information in the fused information data based on the results of the first multi-dimensional association analysis, the second multi-dimensional association analysis, and the third multi-dimensional association analysis.
8. The information security management system based on big data according to claim 6 is characterized in that: Security policies, including: data detection policy, data response policy, secure access policy, and data encryption policy.
Citation Information
Cited By
Data fusion system based on AI large model agent
CN120832631A
