Big data multi-dimensional safety risk analysis system for students at school based on artificial intelligence

Abstract

The invention discloses a school student big data multi-dimensional safety risk analysis system based on artificial intelligence, and belongs to the technical field of school student safety. The platform end comprises a case library and a model optimization module; the case library is used for storing security cases; the model optimization module is used for optimizing and adjusting the security analysis model in real time and synchronously updating the security analysis model applied by the user side; the user side comprises an acquisition module, a risk identification module and an early warning module; the acquisition module is used for acquiring student monitoring data in real time according to the updating record of the safety analysis model; the risk identification module is used for analyzing the student monitoring data in real time to obtain corresponding risk identification data; the early warning module is used for carrying out risk early warning, receiving the risk identification data in real time, and carrying out display processing on the risk identification data according to a preset statistical display mode to obtain corresponding early warning display information; and carrying out early warning processing on corresponding workers according to the early warning display information.

Description

Technical Field

[0001] The present invention belongs to the field of school student safety technology, and specifically is an artificial intelligence-based multi-dimensional safety risk analysis system for school students based on big data. Background Art

[0002] With the rapid development and widespread adoption of information technology, schools, as crucial hubs for knowledge dissemination and talent cultivation, face unprecedented challenges in security management. With a large student population and a wide range of activities, coupled with the widespread application of technologies such as big data, the Internet of Things, and artificial intelligence, students generate massive amounts of multi-dimensional data during their time at school. This data contains a wealth of information, but also harbors various security risks.

[0003] Traditional campus security management methods rely primarily on manual inspections and video surveillance, which have numerous shortcomings. These methods require significant manpower, material resources, and time, and are difficult to implement around-the-clock, comprehensive monitoring. Lack of data sharing and integration between departments leads to severe information silos, hindering the formation of effective collaborative management mechanisms. Traditional security management methods often identify and address security incidents only after they occur, lacking early warning and predictive capabilities, making it difficult to effectively prevent them.

[0004] Based on this, the present invention provides an artificial intelligence-based multi-dimensional security risk analysis system for big data of students in school. Summary of the Invention

[0005] In order to solve the problems existing in the above solutions, the present invention provides an artificial intelligence-based multi-dimensional security risk analysis system for student big data.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] An AI-based multi-dimensional student big data security risk analysis system, including a platform side and a user side;

[0008] The platform includes a case library and a model optimization module;

[0009] The case library is used to store safety cases;

[0010] Furthermore, the safety cases stored in the case library are stored and managed.

[0011] The model optimization module is used to optimize and adjust the security analysis model in real time and to synchronously update the security analysis model of the user-side application.

[0012] Furthermore, the security analysis model is optimized and adjusted in real time, including:

[0013] Identify optimization records of the safety analysis model in real time, and determine simulation cases of the safety analysis model in real time based on the optimization records and a case library;

[0014] Performing simulation analysis on the safety analysis model according to the simulation case to obtain a simulation analysis result of the simulation case, wherein the simulation analysis result is a qualified simulation or a failed simulation;

[0015] Determine the risks to be optimized for the simulation cases that failed the simulation according to the simulation analysis results; and optimize and adjust the safety analysis model according to the risks to be optimized.

[0016] The user terminal includes a collection module, a risk identification module and an early warning module;

[0017] The collection module is used to collect student monitoring data in real time according to the update records of the security analysis model.

[0018] Furthermore, the collection of student monitoring data includes:

[0019] Real-time identification of update records, including the update time and content of the security analysis model, the issues resolved, and the supplementary collected data required to resolve the issues;

[0020] New collection items are determined based on the update records, and supplementary collection is performed based on the new collection items to obtain student monitoring data.

[0021] The risk identification module is used to perform real-time analysis on student monitoring data, obtain corresponding risk identification data, and send the risk identification data to the early warning module.

[0022] The early warning module is used to provide risk early warning, receive risk identification data in real time, display and process the risk identification data according to a preset statistical display method, and obtain corresponding early warning display information; and provide early warning processing to corresponding staff based on the early warning display information.

[0023] Furthermore, the user terminal also includes a personalized warning analysis module, which is used to perform personalized warning analysis on the warning display information, obtain corresponding predicted warning methods and predicted warning results, and display the predicted warning methods and predicted warning results to corresponding staff.

[0024] Furthermore, personalized warning analysis is performed on the warning display information, including:

[0025] Obtain the warning processing results of the warning display information in real time, and integrate the warning display information and warning processing results into personalized warning data.

[0026] Identify the staff information corresponding to the personalized warning data, classify the personalized warning data according to the staff information, and obtain the personalized warning data of the corresponding staff; establish a personalized warning model based on the personalized warning data of each staff member;

[0027] The information of the staff on duty is identified in real time, and the staff information and warning display data are analyzed according to the personalized warning model to obtain the corresponding prediction warning method and prediction warning result.

[0028] Furthermore, the user end also includes a work evaluation module, which is used to evaluate the risk handling ability of the staff and obtain the risk assessment data of the staff, wherein the risk assessment data includes a comprehensive assessment value and a pass rate corresponding to the corresponding risk type; and the risk assessment data is sent to the corresponding management personnel.

[0029] Furthermore, the risk management capabilities of staff members are assessed, including:

[0030] Identify historical warning display information in real time and generate corresponding dynamic simulation risk data based on the historical warning display information;

[0031] Analyze dynamic simulated risk data and staff information based on the personalized early warning model to obtain the simulated early warning method and risk simulation results of the corresponding staff;

[0032] Calculate the qualification rate of corresponding staff for each risk type based on the simulation warning method and risk simulation results;

[0033] A corresponding comprehensive assessment value is calculated based on the pass rate of each risk type, and the comprehensive assessment value and the pass rate of the corresponding risk type are integrated into the risk assessment data of the staff member.

[0034] Furthermore, the calculation of the pass rate includes:

[0035] Establish a result judgment model, the expression of the result judgment model is:

[0036]

[0037] Where: si is the input data, which is the simulation warning method and risk simulation result corresponding to the corresponding risk type, i = 1, 2, ..., n, n is a positive integer; the output data is the result judgment value GP(si), which is 1 or 0;

[0038] Analyze the simulation warning mode and risk simulation results through the result judgment model to obtain the corresponding result judgment value;

[0039] Calculate the pass rate of the corresponding risk type according to the pass rate calculation formula. The pass rate calculation formula is:

[0040]

[0041] Where: g is the pass rate.

[0042] Furthermore, the corresponding comprehensive assessment value is calculated based on the pass rate of each risk type, including:

[0043] The risk type is marked as j, j = 1, 2, ..., m, m is a positive integer; the pass rate is marked as gj;

[0044] Set the proportional coefficient of each risk type and mark the proportional coefficient as μj;

[0045] Calculate the corresponding comprehensive evaluation value according to the comprehensive evaluation formula. The comprehensive evaluation formula is:

[0046]

[0047] Where: ZP is the comprehensive evaluation value.

[0048] Compared with the prior art, the present invention has the following beneficial effects:

[0049] Through the mutual cooperation between various modules, intelligent risk identification of students in school can be achieved, and the accuracy and comprehensiveness of risk identification can be improved; by setting up a personalized early warning analysis module, predictive analysis can be carried out according to the early warning display data and staff information, and a predictive early warning method can be obtained, which is convenient for reducing the workload of staff and realizing predictive processing according to the staff's handling method of the corresponding early warning risk, solving the problem that the safety analysis model is difficult to personalize according to the handling method of the corresponding staff and the differences between students during actual application. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in 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 only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0051] Figure 1 This is a principle block diagram of the present invention. DETAILED DESCRIPTION

[0052] The technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0053] like Figure 1 As shown, a multi-dimensional security risk analysis system for student big data based on artificial intelligence includes a platform end and a user end;

[0054] The platform end is used by the platform party to communicate with the user end used by each school.

[0055] The platform includes a case library and model optimization modules;

[0056] The case library is used to store various safety cases of students in school;

[0057] In one embodiment, safety cases can be collected in real time based on existing big data and other technologies to obtain safety cases of various students in school.

[0058] In one embodiment, the safety cases stored in the case library are stored and managed.

[0059] In one embodiment, storage management may be performed according to an existing storage management method.

[0060] In one embodiment, performing storage management includes:

[0061] In order to distinguish the collected safety cases from the stored safety cases, the collected safety cases are marked as collected cases, the collected cases are identified in real time, the collected cases are screened according to the safety cases stored in the case library, and the remaining collected cases are stored in the case library.

[0062] In one embodiment, the collected cases are screened according to the safety cases stored in the case library. The screening can be performed based on existing screening methods, such as eliminating the collected cases that are identical or too similar.

[0063] The model optimization module is used to optimize and adjust the security analysis model established by the platform in real time. The security analysis model is used to perform real-time analysis on student monitoring data at the user's location and identify the security risks of the corresponding students. The security analysis model uses artificial intelligence algorithms (such as machine learning, deep learning, etc.) to deeply mine and analyze the stored data, identify potential security risks, and predict possible security risks in the future based on historical data and real-time data. The security analysis model is specifically established according to existing methods; the security analysis model is simulated and analyzed in real time based on the case library to determine the risks to be optimized, and the security analysis model is optimized and adjusted based on the risks to be optimized. The platform staff specifically optimizes and adjusts the security analysis model, such as adjusting the corresponding learning data according to the risks to be optimized; the security analysis model at the user end is synchronously updated based on the optimized security analysis model.

[0064] In one embodiment, a real-time simulation analysis of a security analysis model is performed based on a case library, including:

[0065] Identify safety cases in the case library that have not been simulated and analyzed for the safety analysis model, and mark the corresponding safety cases as simulation cases. This is because when the safety analysis model has not been updated or adjusted, some stored safety cases have already been simulated and analyzed as simulation cases, so they do not need to be analyzed again. They will only be re-identified as simulation cases after the safety analysis model is updated or adjusted. Simulation cases can be identified based on the optimization record of the safety analysis model, that is, the optimization record includes the simulation cases applied in each optimization analysis.

[0066] The safety analysis model is simulated and analyzed according to the simulation case to obtain the corresponding simulation analysis results, that is, the corresponding student monitoring data is generated according to the simulation case, and the identification analysis is performed through the safety analysis model to obtain the corresponding safety identification results. The safety identification results are compared with the actual results corresponding to the simulation case to determine the simulation analysis results. The simulation analysis results include simulation pass and simulation fail. The risk to be optimized is determined according to the simulation case corresponding to the simulation fail, that is, the safety risk corresponding to the simulation case.

[0067] The user terminal includes a collection module, a risk identification module and an early warning module;

[0068] The collection module is used to collect student monitoring data from multiple dimensions in real time, and obtain the update record of the security analysis model in real time. The update record includes the update time, update content, problem solving, identification of supplementary collected data required to solve the corresponding problem, and other related information of the security analysis model. New collection items are determined based on the update record, that is, after the security analysis model is updated, in order to solve a certain security risk, whether it is necessary to collect additional data to identify the security risk, that is, whether the existing collection data range can meet the analysis requirements of the updated security analysis model. If not, what data needs to be collected additionally, and then the new collection items are determined; supplementary collection is performed according to the new collection items to obtain student monitoring data, that is, data collection is performed according to the original collection items while data collection is also performed according to the new collection items, and the data collected by both are correspondingly integrated to form complete student monitoring data.

[0069] The risk identification module is used to perform real-time analysis on student monitoring data to obtain corresponding risk identification data. The risk identification data includes risk type, student information, location and other related data. That is, the risk identification model is used to identify the risks in real time, and risk-related data such as student information and location are collected based on the risks; the risk identification data is sent to the early warning module.

[0070] The early warning module is used to carry out risk early warning, receive risk identification data sent by the risk identification module in real time, and display the received risk identification data according to a preset statistical display method, such as extracting corresponding features according to preset standards, and supplementing corresponding videos, images and other information, so that relevant staff can intuitively understand the risk situation, set the corresponding statistical display method according to the display requirements, and then implement the corresponding display processing; obtain the corresponding early warning display information; and carry out early warning processing to the corresponding staff according to the early warning display information. For example, when there is new early warning display information, the staff will be warned and the corresponding processing measures can also be prompted. At the same time, risk events that have already been warned can be supervised and processed.

[0071] In one embodiment, the user terminal further includes a personalized early warning analysis module, which is used to analyze the early warning processing results of the early warning display information, and subsequently analyze the early warning display information according to the actual situation of the corresponding staff, predict the staff's processing method, and perform corresponding processing after the staff determines it. It can also directly process it according to the predicted processing method after authorization, thereby reducing the workload of the staff and realizing personalized risk processing. The process is as follows:

[0072] The warning processing results of the warning display information are obtained in real time. The warning processing results include the corresponding processing process and relevant data such as actual results, and the warning display information and warning processing results are integrated into personalized warning data.

[0073] Identify the staff information corresponding to the personalized warning data, that is, the staff information that performs the corresponding warning processing. Because in order to achieve round-the-clock risk duty, there are at least multiple staff members, so differentiated analysis is required; classify the personalized warning data according to the corresponding staff information to obtain the personalized warning data corresponding to the corresponding staff member.

[0074] A personalized warning model is established based on the personalized warning data of each staff member; specifically, it is established based on intelligent technologies such as deep learning algorithms and machine learning, and the personalized warning data is used to set corresponding training sets, verification sets, etc. for training and verification. The input data is staff information and warning display information, and the output data is predicted warning methods and predicted warning results; the intelligent model after successful training is marked as a personalized warning model.

[0075] Identify staff information in real time, analyze staff information and warning display data according to the personalized warning model, obtain corresponding prediction warning methods and prediction warning results, and display the prediction warning methods and prediction warning results to the corresponding staff.

[0076] In one embodiment, other methods may also be used to perform personalized warning analysis on the warning display information to obtain corresponding prediction warning methods.

[0077] By setting up a personalized early warning analysis module, predictive analysis can be carried out based on the early warning display data and staff information, and a predictive early warning method can be obtained, which is convenient for reducing the workload of staff and realizing predictive processing according to the staff's handling method of the corresponding early warning risks. It solves the problem that the safety analysis model is difficult to personalize according to the handling method of the corresponding staff and the differences between students during actual application.

[0078] In one embodiment, the user terminal further includes a work evaluation module, which is used to evaluate the risk handling ability of the staff and determine their ability to handle student risks. The process includes:

[0079] Real-time identification of the school's historical warning display information, and generation of corresponding dynamic simulation risk data based on the historical warning display information. Dynamic simulation risk data is generated based on the historical warning display information. It is representative warning display information for the school. For example, based on the historical warning display information, the various risk types and proportions of the school are determined. Then, based on the various student information of the school, the warning display information under the corresponding risk type is simulated. The corresponding data volume is generated according to the proportion of the corresponding risk type to form dynamic simulation risk data.

[0080] The dynamic simulated risk data and staff information are analyzed according to the personalized early warning model in the above embodiment to obtain the predicted early warning method and predicted early warning result of the corresponding staff member, and the corresponding predicted early warning method and predicted early warning result are marked as the simulated early warning method and risk simulation result. Based on the simulated early warning method and risk simulation result, it is determined whether the risk treatment of the corresponding dynamic simulated risk data by the staff member meets the school requirements. The school requirements are set by the corresponding school management personnel; based on the judgment result, the qualified rate of the staff member's early warning treatment of the corresponding risk type is calculated, that is, the proportion that meets the school requirements;

[0081] The corresponding comprehensive assessment value is calculated based on the pass rate corresponding to each risk type, the comprehensive assessment value and the pass rate of the corresponding risk type are integrated into the risk assessment data of the staff member, and the risk assessment data is sent to the corresponding management personnel.

[0082] In one embodiment, corresponding dynamic simulation risk data is generated based on historical warning display information, and can also be generated based on other existing methods, such as establishing an intelligent model based on a deep learning algorithm and performing intelligent generation through the intelligent model.

[0083] In one embodiment, the calculation of the pass rate includes:

[0084] Establish a result judgment model, the expression of the result judgment model is:

[0085]

[0086] Where: si is the input data, which is the simulated warning method and risk simulation result corresponding to the corresponding risk type, i = 1, 2, ..., n, n is a positive integer, which is the number of simulated warning methods and risk simulation results corresponding to the risk type, and can also be regarded as the number of dynamic simulation risk data; the output data is the result judgment value GP(si), which is 1 or 0; the training set is set using historical simulated warning methods and historical risk simulation results to realize judgment analysis of the input data.

[0087] Analyze the simulation warning mode and risk simulation results through the result judgment model to obtain the corresponding result judgment value;

[0088] Calculate the pass rate of the corresponding risk type according to the pass rate calculation formula. The pass rate calculation formula is:

[0089]

[0090] Where: g is the pass rate.

[0091] In one embodiment, the corresponding comprehensive evaluation value is calculated according to the qualification rate of each risk type, and a comprehensive evaluation can be performed based on an existing comprehensive evaluation method to obtain the corresponding comprehensive evaluation value.

[0092] In one embodiment, the corresponding comprehensive evaluation value is calculated based on the pass rate of each risk type, including:

[0093] The risk type is marked as j, j = 1, 2, ..., m, m is a positive integer, i.e., the number of risk types; the pass rate is marked as gj;

[0094] Set the proportional coefficient of each risk type. You can set it according to the proportion of the risk type, risk impact, risk loss, etc. For example, the proportional coefficient of risk impact can be set according to the historical impact of the corresponding risk type. You can also set the proportional coefficient based on multiple factors. It is set according to user needs. The proportional coefficient is marked as μj. Calculate the corresponding comprehensive evaluation value according to the comprehensive evaluation formula. The comprehensive evaluation formula is:

[0095]

[0096] Where: ZP is the comprehensive evaluation value.

[0097] The above formulas are all calculated by removing dimensions and taking their numerical values. The formula is a formula that is closest to the actual situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and preset thresholds in the formula are set by technicians in this field according to actual conditions or obtained by simulating a large amount of data.

[0098] The above embodiments are only used to illustrate the technical method of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. An artificial intelligence-based multi-dimensional student big data security risk analysis system, characterized by: Including platform side and user side; The platform includes a case library and a model optimization module; The case library is used to store safety cases; The model optimization module is used to optimize and adjust the security analysis model in real time and synchronously update the security analysis model of the user-side application; The user terminal includes a collection module, a risk identification module and an early warning module; The acquisition module is used to collect student monitoring data in real time according to the update record of the security analysis model; The risk identification module is used to perform real-time analysis on student monitoring data, obtain corresponding risk identification data, and send the risk identification data to the early warning module; The early warning module is used to provide risk early warning, receive risk identification data in real time, display and process the risk identification data according to a preset statistical display method, and obtain corresponding early warning display information; and provide early warning processing to corresponding staff based on the early warning display information.

2. The multi-dimensional security risk analysis system for student big data based on artificial intelligence according to claim 1 is characterized in that: Store and manage safety cases stored in the case library.

3. The multi-dimensional security risk analysis system for student big data based on artificial intelligence according to claim 1 is characterized in that: Real-time optimization and adjustment of security analysis models, including: Identify optimization records of the safety analysis model in real time, and determine simulation cases of the safety analysis model in real time based on the optimization records and a case library; Performing simulation analysis on the safety analysis model according to the simulation case to obtain a simulation analysis result of the simulation case, wherein the simulation analysis result is a qualified simulation or a failed simulation; Determine the risks to be optimized for the simulation cases that failed the simulation according to the simulation analysis results; and optimize and adjust the safety analysis model according to the risks to be optimized.

4. The multi-dimensional security risk analysis system for student big data based on artificial intelligence according to claim 1 is characterized in that: The collection of student monitoring data includes: Real-time identification of update records, including the update time and content of the security analysis model, the issues resolved, and the supplementary collected data required to resolve the issues; New collection items are determined based on the update records, and supplementary collection is performed based on the new collection items to obtain student monitoring data.

5. The multi-dimensional security risk analysis system for student big data based on artificial intelligence according to claim 1 is characterized in that: The user end also includes a personalized warning analysis module, which is used to perform personalized warning analysis on the warning display information, obtain corresponding prediction warning methods and prediction warning results, and display the prediction warning methods and prediction warning results to corresponding staff.

6. The multi-dimensional security risk analysis system for student big data based on artificial intelligence according to claim 5 is characterized in that: Assess staff's risk management capabilities, including: Identify historical warning display information in real time and generate corresponding dynamic simulation risk data based on the historical warning display information; Analyze dynamic simulated risk data and staff information based on the personalized early warning model to obtain the simulated early warning method and risk simulation results of the corresponding staff; Calculate the qualification rate of corresponding staff for each risk type based on the simulation warning method and risk simulation results; A corresponding comprehensive assessment value is calculated based on the pass rate of each risk type, and the comprehensive assessment value and the pass rate of the corresponding risk type are integrated into the risk assessment data of the staff member.

7. The multi-dimensional security risk analysis system for student big data based on artificial intelligence according to claim 5 is characterized in that: The user end also includes a work evaluation module, which is used to evaluate the risk handling ability of the staff and obtain the risk assessment data of the staff. The risk assessment data includes a comprehensive assessment value and a pass rate corresponding to the corresponding risk type; and the risk assessment data is sent to the corresponding management personnel.

8. The multi-dimensional security risk analysis system for student big data based on artificial intelligence according to claim 7 is characterized in that: Assess staff's risk management capabilities, including: Identify historical warning display information in real time and generate corresponding dynamic simulation risk data based on the historical warning display information; Analyze dynamic simulated risk data and staff information based on the personalized early warning model to obtain the simulated early warning method and risk simulation results of the corresponding staff; Calculate the qualification rate of corresponding staff for each risk type based on the simulation warning method and risk simulation results; A corresponding comprehensive assessment value is calculated based on the pass rate of each risk type, and the comprehensive assessment value and the pass rate of the corresponding risk type are integrated into the risk assessment data of the staff member.

9. The multi-dimensional security risk analysis system for student big data based on artificial intelligence according to claim 8 is characterized in that: The calculation of the pass rate includes: Establish a result judgment model, the expression of the result judgment model is: Where: si is the input data, which is the simulation warning method and risk simulation result corresponding to the corresponding risk type, i = 1, 2, ..., n, n is a positive integer; the output data is the result judgment value GP(si), which is 1 or 0; Analyze the simulation warning mode and risk simulation results through the result judgment model to obtain the corresponding result judgment value; Calculate the pass rate of the corresponding risk type according to the pass rate calculation formula. The pass rate calculation formula is: Where: g is the pass rate.

10. The multi-dimensional security risk analysis system for student big data based on artificial intelligence according to claim 8 is characterized in that: The corresponding comprehensive assessment value is calculated based on the pass rate of each risk type, including: The risk type is marked as j, j = 1, 2, ..., m, m is a positive integer; the pass rate is marked as gj; Set the proportional coefficient of each risk type and mark the proportional coefficient as μj; Calculate the corresponding comprehensive evaluation value according to the comprehensive evaluation formula. The comprehensive evaluation formula is: Where: ZP is the comprehensive evaluation value.

Images