Campus network security emergency management method and system based on public opinion monitoring

Abstract

The invention discloses a campus network security emergency management method and system based on public opinion monitoring, relates to the technical field of campus network security, and solves the technical problem that in the prior art, after network public opinions occur, network security measures with high exclusiveness are collected, and related data of the network public opinions are difficult to reasonably protect. According to the method, whether data in the campus network will be impacted or not is judged by combining the abnormal traffic of the campus network and the public opinion popularity of the network public opinions, and the access of the campus network is limited in advance according to the judgment result; according to the method, after the network public opinions occur, the access of the campus network is limited by comparing the relationship between the network public opinions and the abnormal traffic, so that the impact on data assets in the campus network can be avoided; according to the method, the time sequence relationship between the network public opinion and the campus network traffic abnormality is considered, and the emergency assessment coefficient is calculated in combination with the trend correlation degree, the abnormal traffic and the public opinion score, so that the accuracy and rationality of emergency measures collected according to the emergency assessment coefficient are ensured.

Description

Technical Field

[0001] The present application belongs to the field of campus network security technology, and specifically is a campus network security emergency management method and system based on public opinion monitoring. Background Art

[0002] The campus network is a network specially built to meet the needs of school teaching, scientific research, management and other educational activities. It is highly specialized and professional. Because the campus network connects teachers and students, teaching resources and the outside world, it stores a large amount of sensitive data such as personal information, academic achievements, research data, etc. Ensuring the security of this data is very important for the school.

[0003] With the development of information technology, any speech posted online by anyone may become a flashpoint. Once a flashpoint is formed, the relevant information of the person concerned may be targeted by criminals. If the person concerned is related to the school, criminals may obtain relevant information and expose it in various ways. Even if the campus network is equipped with certain security measures, it is still difficult to prevent large-scale network attacks. Information leakage will cause irreversible damage to the person concerned and even the school.

[0004] This application provides a campus network security emergency management method and system based on public opinion monitoring to solve the above-mentioned technical problems. Summary of the invention

[0005] The present application aims to solve at least one of the technical problems existing in the prior art; to this end, the present application proposes a campus network security emergency management method and system based on public opinion monitoring, which is used to solve the technical problem that the prior art collects highly exclusive network security measures after the occurrence of online public opinion, making it difficult to reasonably protect the data related to online public opinion.

[0006] To achieve the above objectives, the first aspect of the present application provides a campus network security emergency management method based on public opinion monitoring, including:

[0007] Real-time monitoring of online public opinion related to the school, evaluating the sentiment index and the heat of the online public opinion; calculating the public opinion score of the online public opinion based on the sentiment index and the heat of the public opinion; wherein the sentiment index is a comprehensive index including sentiment polarity and sentiment intensity, and the public opinion score is used to indicate the degree of negative impact of the online public opinion on the campus network;

[0008] Monitor abnormal traffic in the campus network; determine the emergency assessment coefficient of the campus network based on the abnormal traffic and the public opinion score; restrict access to characteristic data in the campus network based on the emergency assessment coefficient; wherein characteristic data refers to data related to network public opinion, and the method of restricting access includes restricting user access or hiding characteristic data.

[0009] Preferably, the emotional index of evaluating the network public opinion includes:

[0010] Extracting original texts related to online public opinion from various online channels, and obtaining target texts after preprocessing the original texts; wherein the preprocessing includes text cleaning, word segmentation, part-of-speech tagging or stem extraction;

[0011] The target text is identified through a sentiment recognition model to obtain sentiment indicators of each network channel; the sentiment indicators of each network channel are weighted and calculated to obtain the sentiment indicators of the network public opinion; wherein the sentiment recognition model is constructed based on artificial intelligence.

[0012] Preferably, evaluating the public opinion heat of the network public opinion includes:

[0013] Obtaining sentiment indicators and user influence indicators of various network channels related to the network public opinion, and determining the time decay factor of the network public opinion; wherein the user influence indicator includes user popularity and user influence;

[0014] The public opinion heat of the corresponding network public opinion is calculated based on the time decay factor and the sentiment index and user influence of each network channel.

[0015] Preferably, calculating the public opinion heat corresponding to the network public opinion includes:

[0016] According to the formula δ(t) = e -λt Calculate the time decay factor δ(t); mark the sentiment index of each network channel as e(i), and mark the user popularity and user influence as c(j,t) and u(j) respectively; where λ is the decay rate, which is obtained by counting the same type of network public opinions, t is time, i represents the network channel, and j represents the user;

[0017] By formula r(t) = ∑ i∈d sra(i)×(∑ j∈u(i) c(j,t)×u(j))×δ(t)×e(i) is used to calculate the public opinion heat r(t) of the corresponding network public opinion; among them, sra(i) is the weight of network channel i, d is the set of network channels, and u(i) is the set of all users in network channel i.

[0018] Preferably, the monitoring of abnormal traffic on the campus network includes:

[0019] Acquire campus network data packets in real time; wherein the data packets include basic features, content features, traffic statistics features and original load;

[0020] Abnormal traffic in data packets is identified through a traffic monitoring model; wherein the traffic monitoring model is built based on the ARMA traffic model.

[0021] Preferably, determining the emergency assessment coefficient of the campus network according to the abnormal traffic and the public opinion score includes:

[0022] Performing time series analysis on the abnormal traffic and the public opinion score to identify their trend characteristics over time; wherein the time series analysis method includes an autoregressive model or a moving average model;

[0023] Introducing the time deviation, aligning the trend characteristics of the abnormal traffic and the public opinion score based on the time deviation, and calculating the trend correlation between the two; wherein the time trend refers to the chronological relationship between the abnormal traffic and the public opinion score;

[0024] The trend correlation, abnormal traffic and public opinion score are combined to determine the emergency assessment coefficient of the campus network; wherein the emergency assessment coefficient refers to the degree of influence of network public opinion on the campus network.

[0025] Preferably, calculating the trend correlation comprises:

[0026] The abnormal traffic and public opinion scores at time t are marked as L t and P t , through the formula The trend correlation C is calculated; where: and The time series means of abnormal traffic and public opinion scores respectively, and n is the number of time points.

[0027] Preferably, restricting access to characteristic data in the campus network according to the emergency assessment coefficient includes:

[0028] When the emergency assessment coefficient is less than the coefficient threshold 1, characteristic data in the campus network is determined according to the network public opinion, and access to the characteristic data is restricted;

[0029] When the emergency assessment coefficient is not less than the coefficient threshold value 1, the characteristic data is encrypted and stored in isolation to achieve hiding of the characteristic data.

[0030] Preferably, the restricted access duration to the feature data is determined according to the changing trend of the public opinion score, including:

[0031] When the public opinion score shows a downward trend and the emergency assessment coefficient is less than coefficient threshold 2, the restricted access to the feature data is lifted; wherein coefficient threshold 2 is less than or equal to coefficient threshold 1.

[0032] The second aspect of the present application provides a campus network security emergency management system based on public opinion monitoring, including a decision analysis module, and a public opinion monitoring module and a security control module connected thereto;

[0033] Public opinion monitoring module: used to monitor online public opinion related to the school in real time, and evaluate and obtain the public opinion score corresponding to the online public opinion;

[0034] Decision analysis module: used to monitor abnormal traffic of the campus network; determine the emergency assessment coefficient of the campus network according to the abnormal traffic and the public opinion score;

[0035] Security control module: restricting access to characteristic data in the campus network according to the emergency assessment coefficient.

[0036] Compared with the prior art, the beneficial effects of this application are:

[0037] This application monitors the generation and development of online public opinion in real time, and determines whether it will have an impact on the data in the campus network based on the abnormal traffic of the campus network and the heat of the online public opinion, and restricts access to the campus network in advance based on the judgment results; after the occurrence of online public opinion, this application limits access to the campus network by comparing the relationship between online public opinion and abnormal traffic, which helps to avoid impacts on data assets in the campus network.

[0038] When calculating and comparing abnormal traffic and public opinion scores, this application introduces time deviation to align the abnormal traffic and public opinion scores, and then calculates the trend correlation between the two; this application considers the time sequence between online public opinion and campus network traffic anomalies, determines the actual impact of online public opinion on the campus network, and then calculates the emergency assessment coefficient based on trend correlation, abnormal traffic and public opinion scores to ensure the accuracy and rationality of emergency measures collected according to the emergency assessment coefficient. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0040] Figure 1 A schematic diagram of the process of campus network security emergency management in Example 1 of the present application;

[0041] Figure 2 This is a flow chart of obtaining network public opinion sentiment indicators in the second embodiment of the present application;

[0042] Figure 3 This is a flow chart of obtaining the popularity of online public opinion in the second embodiment of the present application;

[0043] Figure 4This is a flow chart of obtaining an emergency assessment coefficient based on abnormal traffic and public opinion scores in Example 3 of the present application;

[0044] Figure 5 This is a schematic diagram of the system principle of this application. DETAILED DESCRIPTION

[0045] The technical solution of the present application will be described clearly and completely in conjunction with the embodiments below. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application.

[0046] The campus network is a network specially built to meet the daily activities of teachers and students in the school. The network is not only used for daily activities within the school, but also connected to the external network. The campus network stores a large amount of sensitive information, such as personal information of teachers and students, various research data and academic achievements, etc. These data are either personal sensitive information or important assets of the school. The impact of data leakage is unbearable.

[0047] The campus network is also used to communicate with the outside world for external inquiries and understanding. With the development of network technology, any speech may become an exposure point at the speed of network transmission, and these exposure points are good and bad. Once the parties involved in these exposure points are related to the school, they will certainly attract the attention of criminals. These criminals will invade the campus network in various ways and steal data related to the parties involved, which will have a bad impact on both the parties involved and the school.

[0048] At present, after a school-related online public opinion event occurs, the security measures taken by the campus network are generally highly exclusive, that is, most people are excluded from access, and only a few authorized people have access rights. If the online public opinion event does not cause relevant personnel to attack the campus network, this highly exclusive measure will affect the access of people who really need it. Therefore, the network security measures of the existing solutions are not reasonable enough and it is difficult to limit access according to actual conditions.

[0049] Embodiment 1: This embodiment monitors school-related online public opinion in real time. If the online public opinion is likely to affect relevant data assets in the campus network, network measures are used to control it in advance to avoid data leakage.

[0050] See also Figure 1 , the campus network security emergency management method based on public opinion monitoring includes:

[0051] Real-time monitoring of online public opinion related to the school, evaluating the sentiment index and sentiment heat of online public opinion; calculating the public opinion score of online public opinion based on the sentiment index and sentiment heat; where the sentiment index is a comprehensive index including sentiment polarity and sentiment intensity, and the public opinion score is used to indicate the degree of negative impact of online public opinion on the campus network;

[0052] Monitor abnormal traffic in the campus network; determine the emergency assessment coefficient of the campus network based on abnormal traffic and public opinion scores; restrict access to characteristic data in the campus network based on the emergency assessment coefficient; characteristic data refers to data related to network public opinion, and methods of restricting access include restricting user access or hiding characteristic data.

[0053] In order to determine the impact of online public opinion on the campus network in advance, it is necessary to keep an eye on the school-related content on the Internet, and it is necessary to monitor the online public opinion related to the school in real time. The monitoring of online public opinion can be achieved through existing solutions, such as building a public opinion software monitoring platform, social media monitoring, search engine monitoring, news monitoring, etc., and combined with natural language processing technology for identification and tracking to extract online public opinion related to the school.

[0054] After determining that public opinion related to the school has appeared on the Internet, it is necessary to analyze the online public opinion and determine the emotional index and public opinion heat of the online public opinion. The emotional index is a comprehensive index calculated mainly based on the emotional polarity and emotional intensity. Then the emotional index and public opinion heat are converted into a public opinion score to indicate the degree of negative impact of online public opinion on the small campus network.

[0055] It should be noted that online public opinion can be positive, negative or neutral. For positive online public opinion, such as a teacher at a school publishing a high-level paper or a student winning a competition award, it is generally believed that no special security protection is required, and only redundant network resources are provided for the surge in traffic. For neutral or negative online public opinion, there is a high possibility of impacting the campus network. Most importantly, these two situations are also very likely to cause criminals to steal the relevant data of the parties involved.

[0056] As online public opinion ferments, it is bound to be accompanied by an increase in abnormal traffic in the campus network. Real-time monitoring of abnormal traffic in the campus network is carried out, and the abnormal traffic and public opinion scores are correlated to determine whether it is necessary to adjust the campus network to deal with the adverse effects of online public opinion. The larger the abnormal traffic and the higher the public opinion score, the higher the corresponding emergency assessment coefficient, and the stricter the access restriction measures that need to be taken to achieve the purpose of ensuring data security in the campus network.

[0057] Abnormal traffic on the campus network can be obtained in the following ways:

[0058] Acquire data packets of the campus network in real time; wherein the data packets include basic features, content features, traffic statistics features and original load; identify abnormal traffic in the data packets through a traffic monitoring model; wherein the traffic monitoring model is built based on the ARMA traffic model.

[0059] The above basic features mainly include source address and destination address, protocol type, port number, header information, etc. If the abnormal traffic of the campus network is closely related to network public opinion, it is necessary to determine what countermeasures the campus network needs to take based on the abnormal traffic and public opinion scores.

[0060] This embodiment monitors the online public opinion related to the school in real time. Once the public opinion score of the online public opinion is correlated with the abnormal traffic of the campus network and has a greater impact on the campus network, an emergency assessment coefficient is set based on the two, and the access mode of the campus network is adjusted by the emergency assessment coefficient. This embodiment adjusts the access mode in advance by predicting the impact of online public opinion on relevant data assets in the campus network, avoiding the leakage of data assets during network attacks.

[0061] Example 2: This example provides a method for calculating the public opinion score based on Example 1. Figures 2 to 3 , the specific steps are as follows:

[0062] 1. For evaluating sentiment indicators of online public opinion, please refer to Figure 2 ,include:

[0063] Extract the original text related to online public opinion from various online channels, and obtain the target text after preprocessing the original text; the preprocessing includes text cleaning, word segmentation, part-of-speech tagging or stemming;

[0064] The target text is identified through the sentiment recognition model to obtain the sentiment index of each network channel; the sentiment index of each network channel is weighted and calculated to obtain the sentiment index of online public opinion; among them, the sentiment recognition model is built based on artificial intelligence.

[0065] The original texts related to online public opinion can be extracted from various online channels through crawler algorithms. These original texts include both the text content of the original posts and the converted text of the audio in the video. Online channels mainly refer to different information publishing platforms.

[0066] After the target text is obtained through preprocessing, the target text is subjected to feature extraction, and the extracted content mainly includes lexical features, semantic features and context features, and then input into the emotion recognition model for recognition to obtain the emotion index corresponding to each network channel. It should be noted that the emotion index in this embodiment is a comprehensive expression of emotion polarity and emotion intensity, which can be represented by the probability value of positive emotion, the probability value of neutral emotion or the probability value of negative emotion, and of course it can be the weighted result of emotion polarity and emotion intensity.

[0067] The emotion recognition model can be built based on the support vector machine model and the random forest model. Of course, it can also be implemented by calling the emotion analysis API interface provided by the existing cloud platform. As long as it is possible to obtain the emotion polarity and emotion intensity corresponding to each network channel by identifying the text data. For the acquisition of emotion indicators, other existing technical solutions can also be referred to, and the relevant content will not be repeated here.

[0068] 2. To evaluate the heat of online public opinion, please refer to Figure 3 ,include:

[0069] Obtain sentiment indicators and user influence indicators of various network channels related to online public opinion, and determine the time decay factor of online public opinion; among them, user influence indicators include user popularity and user influence;

[0070] The public opinion heat of the corresponding online public opinion is calculated based on the time decay factor, the sentiment indicators of each network channel, and the user influence.

[0071] The influence range of each network channel is different, and the development of public opinion in each network channel will be affected by various factors. Therefore, the sentiment index, user influence index and time decay factor of each network channel are included in the influencing factors of public opinion heat.

[0072] The specific calculation method is as follows:

[0073] According to the topic matching of network public opinion, the decay rate λ of the corresponding topic public opinion is obtained, and the time decay factor is calculated according to the decay rate λ. The calculation formula is δ(t) = e -λt , t is time;

[0074] The sentiment index calculated in the above scheme is marked as e(i), and the user popularity and user influence are marked as c(j,t) and u(j) respectively;

[0075] By formula r(t) = ∑ i∈d sra(i)×(∑ j∈u(i)c(j,t)×u(j))×δ(t)×e(i) is used to calculate the public opinion heat r(t) of the corresponding network public opinion; among them, sra(i) is the weight of network channel i, d is the set of network channels, and u(i) is the set of all users in network channel i.

[0076] This example fully considers the influencing factors when calculating the heat of public opinion. For example, the heat of public opinion will change over time, emotional color will affect the heat of public opinion, and users who make relevant comments will also affect the heat of public opinion. Therefore, time attenuation factor, emotional index and user influence index are introduced to improve the accuracy of public opinion heat.

[0077] In some other preferred embodiments, multi-dimensional weights can be introduced to calculate the public opinion heat, and the calculation formula is referenced r(t)=∑ i∈d sra(i)×(∑ j∈u(i) ∑ k∈K c(j,k,t)×w(k)×u(j))×δ(t)×e(i); where c(j,k,t) is the popularity index of user j at dimension k and time t, w(k) is the weight of dimension k, and K is the set of all dimensions. The dimension here can be the user source dimension, such as ordinary users, authenticated users, VIP users, etc., or it can be the traffic dimension, such as the total amount or speed of network traffic.

[0078] Next, the public opinion score of the network public opinion is calculated based on the sentiment index and the heat of public opinion. Considering that the impact of positive network public opinion on the campus network is mainly on the network load, corresponding redundant measures can be set in combination with the heat of public opinion. For negative or neutral network public opinion, it is necessary to prevent it from stealing data assets in the campus network. Therefore, when the sentiment index is negative or neutral, the public opinion score of the network public opinion is determined according to the heat of public opinion.

[0079] This embodiment divides online public opinion into positive and non-positive. For positive online public opinion, the existing solution can be used to deal with the surge in visits, but it is considered that the emotional polarity may change or even reverse during the spread of online public opinion. Therefore, it is still necessary to monitor positive online public opinion in real time. Once its emotional polarity changes, targeted measures need to be taken. For non-positive online public opinion, it is necessary to evaluate its public opinion heat. If the public opinion heat is high, relevant measures need to be taken to ensure the security of the campus network. This embodiment converts emotional indicators and public opinion heat into public opinion scores. The higher the public opinion score, the greater the potential impact of non-positive online public opinion on the campus network, and relevant measures need to be taken in advance.

[0080] It should be noted that the public opinion score can be converted only based on the public opinion heat. For example, the public opinion score and the public opinion heat are linearly related, and of course they can be equal. The public opinion score can also be obtained by combining the sentiment index and the public opinion score, as follows:

[0081] If the sentiment index corresponds to negative sentiment, set the proportional coefficient a; if the sentiment index corresponds to neutral sentiment, set the proportional coefficient to 1; mark the public opinion heat as r(t), then the public opinion score = a×r(t). It should be considered that negative sentiment poses a greater potential security threat to the campus network than neutral sentiment, so a≥1.

[0082] Example 3: Based on Example 1 or Example 2, the emergency assessment coefficient of the campus network is determined according to the abnormal traffic and public opinion score of the campus network, and the response measures to be taken are determined according to the emergency assessment coefficient. Figure 4 .

[0083] There is a time difference between online public opinion and abnormal traffic on campus networks. For example, after online public opinion occurs, it stimulates the motivation of lawbreakers to dig up relevant information about the parties involved, which leads to abnormal traffic on campus networks. Therefore, time alignment is needed to avoid misjudging the impact of online public opinion.

[0084] This embodiment first performs a time series analysis on the abnormal traffic and public opinion scores to identify their trend characteristics over time; wherein the time series analysis method includes an autoregressive model or a moving average model. After alignment, the trend correlation between the abnormal traffic and the public opinion score is calculated. If the trend correlation is large, it means that the network public opinion has largely caused the abnormal traffic on the campus network.

[0085] On this basis, the emergency assessment coefficient of the campus network can be calculated by combining trend relevance, abnormal traffic and public opinion score. The greater the trend relevance, the greater the abnormal traffic and the higher the public opinion score, the higher the corresponding emergency assessment coefficient. Weights can be set for the three dependent variables for calculation. For example, a simple calculation formula can be referred to: E = αC + βT + γP, α, β and γ are the set weights, C, T and P are trend relevance, abnormal traffic value and public opinion score respectively. Of course, other reasonable calculation models can also be constructed.

[0086] The above time deviation indicates the chronological relationship between abnormal traffic and public opinion score (public opinion change trend), and is mainly used to analyze the trend characteristics of abnormal traffic and public opinion score. The calculation formula can be referred to as: D is the time deviation, L t and P tis the abnormal traffic and public opinion score corresponding to time t. In other preferred embodiments, the time deviation degree can also be obtained by statistical means, that is, statistically analyzing the time deviation degree between the public opinion change trend and the abnormal traffic.

[0087] The above trend correlation indicates the correlation between abnormal traffic and public opinion scores, which can be calculated by Pearson correlation coefficient. The specific calculation formula is: and The time series means of abnormal traffic and public opinion scores respectively, and n is the number of time points.

[0088] This embodiment restricts access to characteristic data in the campus network according to the emergency assessment coefficient. The emergency assessment coefficient is set according to trend relevance, abnormal traffic and public opinion score. When the emergency assessment coefficient is less than the set coefficient threshold of one, the characteristic data in the campus network is set to restrict access.

[0089] It should be noted that characteristic data refers to data related to online public opinion, such as personal information of the parties involved, information about participation in activities, or information about awards and honors, etc.; restricted access refers to restricting access to specific personnel, such as restricting external IP access, etc., or restricting by setting up a blacklist. The coefficient threshold 1 can be set based on experience.

[0090] When the emergency assessment coefficient is not less than the set coefficient threshold of one, it is considered that the online public opinion will have a significant impact on the campus network and the data assets stored therein, and the characteristic data will be hidden, such as temporarily stored in a safe location after encryption.

[0091] Of the two methods mentioned above, restricting access and hiding feature data, the former is used to restrict external IPs and external personnel from accessing campus network data, while the latter further expands the scope of restriction compared to the former, and also restricts access to unauthorized IPs or personnel within the school. It can be understood that hiding feature data is to prevent outsiders from invading using the accounts or IPs of internal school personnel. It should be noted that after the feature data is encrypted, it is temporarily stored, so it is sufficient to use highly secure encryption methods, such as quantum key encryption, to ensure high data security during the temporary storage stage.

[0092] Example 4: Based on Example 3, the duration of restricted access is determined, that is, when the public opinion score shows a downward trend and the emergency assessment coefficient is less than coefficient threshold 2, the restricted access to the feature data is lifted; wherein coefficient threshold 2 is less than or equal to coefficient threshold 1. The restricted access here includes restrictions on internal and external personnel.

[0093] In some other preferred embodiments, the lower limit of the emergency assessment coefficient may also be specified when setting the restricted access mode, such as:

[0094] When the emergency assessment coefficient is greater than a certain set value and less than the coefficient threshold 1, the characteristic data in the campus network is determined according to the network public opinion, and access to the characteristic data is restricted; the set value is less than the coefficient threshold 2;

[0095] When the emergency assessment coefficient is not less than the coefficient threshold of one, the characteristic data is encrypted and stored in isolation to achieve hiding of the characteristic data.

[0096] See also Figure 5 , the second aspect of the present application provides a campus network security emergency management system based on public opinion monitoring, including: a decision analysis module, and a public opinion monitoring module and a security control module connected thereto;

[0097] Public opinion monitoring module: used to monitor online public opinion related to the school in real time, and evaluate and obtain the public opinion score corresponding to the online public opinion; decision analysis module: used to monitor abnormal traffic in the campus network; determine the emergency assessment coefficient of the campus network based on abnormal traffic and public opinion scores; security control module: restrict access to characteristic data in the campus network based on the emergency assessment coefficient.

[0098] The public opinion monitoring module uses built-in programs to capture information from the internal campus network and the external Internet, and determines whether online public opinion has been generated. The decision analysis module combines the public opinion analysis results and the abnormal traffic of the campus network to determine the corresponding emergency assessment coefficient, and collects corresponding measures based on the emergency assessment coefficient. The security control module controls the campus network according to the determined measures.

[0099] The above embodiments are only used to illustrate the technical method of the present application and are not intended to limit it. Although the present application has been described in detail with reference to the preferred embodiments, a person of ordinary skill in the art should understand that the technical method of the present application may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present application.

Claims

1. A campus network security emergency management method based on public opinion monitoring, characterized in that: include: Monitor online public opinions related to the school in real time and evaluate the sentiment indicators and popularity of the online public opinions; Calculating the public opinion score of the network public opinion based on the sentiment index and the public opinion heat; wherein the sentiment index is a comprehensive index including sentiment polarity and sentiment intensity, and the public opinion score is used to indicate the degree of negative impact of the network public opinion on the campus network; Monitor abnormal traffic in the campus network; determine the emergency assessment coefficient of the campus network based on the abnormal traffic and the public opinion score; restrict access to characteristic data in the campus network based on the emergency assessment coefficient; wherein characteristic data refers to data related to network public opinion, and the method of restricting access includes restricting user access or hiding characteristic data.

2. The campus network security emergency management method based on public opinion monitoring according to claim 1 is characterized in that: The emotional index for evaluating the online public opinion includes: Extracting original texts related to online public opinion from various online channels, and obtaining target texts after preprocessing the original texts; wherein the preprocessing includes text cleaning, word segmentation, part-of-speech tagging or stem extraction; The target text is identified through a sentiment recognition model to obtain sentiment indicators of each network channel; the sentiment indicators of each network channel are weighted and calculated to obtain the sentiment indicators of the network public opinion; wherein the sentiment recognition model is constructed based on artificial intelligence.

3. The campus network security emergency management method based on public opinion monitoring according to claim 2 is characterized in that: Evaluating the heat of the online public opinion includes: Obtaining sentiment indicators and user influence indicators of various network channels related to the network public opinion, and determining the time decay factor of the network public opinion; wherein the user influence indicator includes user popularity and user influence; The public opinion heat of the corresponding network public opinion is calculated based on the time decay factor and the sentiment index and user influence of each network channel.

4. The campus network security emergency management method based on public opinion monitoring according to claim 3 is characterized in that: Calculating the public opinion heat of the corresponding network public opinion, including: According to the formula δ(t) = e -λt Calculate the time decay factor δ(t); mark the sentiment index of each network channel as e(i), and mark the user popularity and user influence as c(j,t) and u(j) respectively; where λ is the decay rate, which is obtained by counting the same type of network public opinions, t is time, i represents the network channel, and j represents the user; By formula r(t) = ∑ i∈d sra(i)×(∑ j∈u(i) c(j,t)×u(j))×δ(t)×e(i) is used to calculate the public opinion heat r(t) of the corresponding network public opinion; among them, sra(i) is the weight of network channel i, d is the set of network channels, and u(i) is the set of all users in network channel i.

5. The campus network security emergency management method based on public opinion monitoring according to claim 4 is characterized in that: The monitoring of abnormal traffic on the campus network includes: Acquire campus network data packets in real time; wherein the data packets include basic features, content features, traffic statistics features and original load; Abnormal traffic in data packets is identified through a traffic monitoring model; wherein the traffic monitoring model is built based on the ARMA traffic model.

6. The campus network security emergency management method based on public opinion monitoring according to claim 5 is characterized in that: Determining the emergency assessment coefficient of the campus network according to the abnormal traffic and the public opinion score includes: Performing time series analysis on the abnormal traffic and the public opinion score to identify their trend characteristics over time; wherein the time series analysis method includes an autoregressive model or a moving average model; Introducing the time deviation, aligning the trend characteristics of the abnormal traffic and the public opinion score based on the time deviation, and calculating the trend correlation between the two; wherein the time trend refers to the chronological relationship between the abnormal traffic and the public opinion score; The trend correlation, abnormal traffic and public opinion score are combined to determine the emergency assessment coefficient of the campus network; wherein the emergency assessment coefficient refers to the degree of influence of network public opinion on the campus network.

7. The campus network security emergency management method based on public opinion monitoring according to claim 6 is characterized in that: Calculating the trend correlation includes: The abnormal traffic and public opinion scores at time t are marked as L t and P t , through the formula The trend correlation C is calculated; where: and The time series means of abnormal traffic and public opinion scores respectively, and n is the number of time points.

8. The campus network security emergency management method based on public opinion monitoring according to claim 1 is characterized in that: Access to characteristic data in the campus network is restricted according to the emergency assessment coefficient, including: When the emergency assessment coefficient is less than the coefficient threshold 1, characteristic data in the campus network is determined according to the network public opinion, and access to the characteristic data is restricted; When the emergency assessment coefficient is not less than the coefficient threshold value 1, the characteristic data is encrypted and stored in isolation to achieve hiding of the characteristic data.

9. The campus network security emergency management method based on public opinion monitoring according to claim 8 is characterized in that: The duration of restricted access to the characteristic data is determined based on the trend of changes in the public opinion score, including: When the public opinion score shows a downward trend and the emergency assessment coefficient is less than coefficient threshold 2, the restricted access to the feature data is lifted; wherein coefficient threshold 2 is less than or equal to coefficient threshold 1.

10. A campus network security emergency management system based on public opinion monitoring, used to implement the campus network security emergency management method based on public opinion monitoring according to any one of claims 1 to 9, characterized in that: It includes a decision analysis module, as well as a public opinion monitoring module and a security control module connected to it; Public opinion monitoring module: used to monitor online public opinion related to the school in real time, and evaluate and obtain the public opinion score corresponding to the online public opinion; Decision analysis module: used to monitor abnormal traffic of the campus network; determine the emergency assessment coefficient of the campus network according to the abnormal traffic and the public opinion score; Security control module: restricting access to characteristic data in the campus network according to the emergency assessment coefficient.

Images