A regional safety evaluation method and device based on public opinion heat diffusion trend analysis, equipment, medium and product

Abstract

The application discloses a regional security evaluation method and device based on public opinion heat diffusion trend analysis, equipment, medium and product, and relates to the field of artificial intelligence. The method comprises the following steps: obtaining public opinion information on an Internet platform and performing data cleaning, feature extraction and format processing to extract formatted information features; classifying and scoring each piece of information based on the formatted information features to obtain a security evaluation score of an index to which each piece of information belongs; clustering each piece of information based on the formatted information features and sorting them in chronological order to identify a set of key events; calculating an event trend index for each key event in the set of key events; calculating an update amount of the influence of each key event on a regional security index based on the event trend index and calculating an updated regional security index; and performing regional security evaluation based on the regional security index before and after the update and the event trend index, which can improve the objectivity, timeliness and accuracy of regional security evaluation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of artificial intelligence, in particular to a regional safety evaluation method and device based on public opinion heat diffusion trend analysis, equipment, medium and product. BACKGROUND

[0002] The traditional regional safety evaluation method is limited to field investigation data, which has the following problems: on the one hand, the input information is single, and it is difficult to ensure the objectivity of the field investigation personnel; on the other hand, the influence of social public opinion trend on regional safety is not considered, and the time series of the data collected by the traditional investigation is short, so it is difficult to add the development trend of public opinion to the consideration of decision-making; in addition, the data collection and processing efficiency is low, and most of them are processed manually, which has the problem of timeliness, that is, the decision-making cannot keep up with the changes. SUMMARY

[0003] The purpose of the present application is to provide a regional safety evaluation method, device, equipment, medium and product based on public opinion heat diffusion trend analysis, so as to improve the objectivity, timeliness and accuracy of regional safety evaluation.

[0004] To achieve the above purpose, the present application provides the following solutions.

[0005] In a first aspect, the present application provides a regional safety evaluation method based on public opinion heat diffusion trend analysis, comprising:

[0006] Obtaining public opinion information on an Internet platform; the public opinion information includes text information, video information, audio information and image information;

[0007] Performing data cleaning, feature extraction and format processing on the public opinion information to extract the formatted information features of each information; the formatted information features of each information include number, region, event, personnel, keyword, time and theme;

[0008] Classifying based on the formatted information features of each information, and classifying each information into each index of the regional safety index; each index of the regional safety index includes crime rate, public order, health safety, economic safety, ecological safety, political safety, social safety, traffic safety and information safety;

[0009] Scoring the formatted information features of each information to obtain the safety evaluation score of the index to which each information belongs;

[0010] Clustering each information based on the formatted information features of each information and sorting in time sequence to identify a set of key events;

[0011] Calculating an event trend index for each key event in the set of key events;

[0012] calculating an updated amount of influence of each key event on the regional safety index based on the event trend index of each key event and calculating an updated regional safety index;

[0013] conducting regional safety evaluation based on the regional safety index before and after the update and the event trend index of each key event.

[0014] Optionally, the public opinion information is subjected to data cleaning, feature extraction and format processing, and the formatted information features of each piece of information are extracted, specifically including:

[0015] The data cleaning is performed on each piece of information in the public opinion information, and useless data and burr data are cleaned away, and then the text features, image features or audio features of each piece of information are extracted by using feature engineering;

[0016] Based on the text features, image features or audio features of each piece of information, the region, event, personnel, keyword, time and theme contained in each piece of information are recognized and expressed in the form of a seven-tuple (i, a, e, p, k, t, s) as the formatted information features of each piece of information; wherein i, a, e, p, k, t and s represent the number, region, event, personnel, keyword, time and theme corresponding to each piece of information respectively.

[0017] Optionally, the classification is performed based on the formatted information features of each piece of information, and each piece of information is classified into each index of the regional safety index, specifically including:

[0018] Based on the formatted information features of each piece of information, a classifier constructed by using Transformer is used to obtain the probability of each piece of information belonging to each index in the regional safety index, and the index corresponding to the maximum probability is determined as the index to which the piece of information belongs.

[0019] Optionally, the scoring is performed on the formatted information features of each piece of information to obtain the safety evaluation score of the index to which each piece of information belongs, specifically including:

[0020] The SVR scorer is constructed or BERT is used to score the formatted information features of each piece of information, and the score is valued in the interval [-5, 5], thereby obtaining the safety evaluation score of the index to which each piece of information belongs.

[0021] Optionally, the event trend index is calculated for each key event in the set of key events, specifically including:

[0022] For each key event in the set of key events, the formula is used to calculate the sentiment score S t ; wherein N t is the number of newly added public opinion information related to the key event within the time t.i The security evaluation score of the indicator to which the i-th piece of information belongs within time t;

[0023] Using formula Calculate the event propagation rate R for each critical event. t Where T is the time window;

[0024] Using formula Calculate the public opinion fluctuation index V t ;

[0025] Using formula Calculate event entropy H t Where k1 is the first proportionality coefficient; P(E) i ) represents the different safety evaluation scores s within time t. i The probability distribution;

[0026] Using formula Calculate the event temperature T t Where k2 is the second proportionality coefficient;

[0027] Using the formula STI t =κ*S t +(1-κ)*STI t-1 Calculate the statistical trend index (STI) for time t. t Where κ is the smoothing coefficient; STI t-1 It is the statistical trend index of the previous time point t-1;

[0028] Based on the sentiment score S for each key event t Event propagation rate R t Public opinion fluctuation index V t Event entropy H t Event temperature T t and the Statistical Trend Index (STI) t Using the formula ETI t =α*S t +β*R t +γ*V t +δ*H t +ε*T t +ζ*STI t Calculate the Event Trend Index (ETI) for this key event. t ;where α, β, γ, δ, ε, ζ are S respectively t R t V t H t T t and STI t The weighting coefficients.

[0029] Optionally, the event trend index based on each key event calculates an updated amount of influence of each key event on the regional security index and calculates an updated regional security index, specifically comprising:

[0030] Based on the diffusion-reaction equation simulation of the development of public opinion, the public opinion concentration C(x, t) of a key event n at time t is calculated;

[0031] The formula is used to calculate the updated amount of influence of the key event n on the regional security index

[0032] The formula is used to calculate the updated regional security index Wherein is the regional security index before updating.

[0033] Secondly, the application provides a regional security evaluation device based on public opinion heat diffusion trend analysis, comprising:

[0034] An opinion information acquisition module is configured to acquire public opinion information on an Internet platform; the public opinion information includes text information, video information, audio information, and image information;

[0035] An opinion information processing module is configured to perform data cleaning, feature extraction, and format processing on the public opinion information, and extract format information features of each piece of information; the format information features of each piece of information include a number, a region, an event, a person, a keyword, a time, and a theme;

[0036] An opinion information classification module is configured to classify based on the format information features of each piece of information, and classify each piece of information into each index of a regional security index; the indexes of the regional security index include a crime rate, a public security situation, a health safety, an economic safety, an ecological safety, a political safety, a social safety, a traffic safety, and an information safety;

[0037] An opinion information scoring module is configured to score the format information features of each piece of information, and obtain a safety evaluation score of an index to which each piece of information belongs;

[0038] An opinion information clustering module is configured to cluster each piece of information based on the format information features of each piece of information, and sort the pieces of information in chronological order, and identify a key event set;

[0039] An event trend index calculation module is configured to calculate an event trend index of each key event in the key event set;

[0040] The regional security index update module is used to calculate the update amount of the impact of each key event on the regional security index based on the event trend index of each key event, and to calculate the updated regional security index.

[0041] The regional security assessment module is used to conduct regional security assessments based on the regional security index before and after the update, as well as the event trend index for each key event.

[0042] Thirdly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the regional security assessment method based on the analysis of the spread trend of public opinion heat.

[0043] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the regional security assessment method based on the analysis of the diffusion trend of public opinion heat.

[0044] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the regional security assessment method based on the analysis of the diffusion trend of public opinion heat.

[0045] According to the specific embodiments provided in this application, the following technical effects are disclosed:

[0046] This application provides a method, apparatus, equipment, medium, and product for regional security assessment based on the analysis of public opinion heat diffusion trends. It combines multimodal public opinion information from internet platforms and achieves a comprehensive assessment of regional security through deep learning technology. By introducing an event trend index, it effectively integrates key factors such as sentiment score, event propagation rate, public opinion fluctuation index, event entropy, event temperature, and statistical trend index, reflecting the real-time development trend of events. Simultaneously, it uses a diffusion-reaction equation to simulate the spread of public opinion information, providing a scientific basis for updating the regional security index. The method in this application comprehensively assesses regional security based on indicators such as crime rate, public security, health security, economic security, ecological security, political security, social security, traffic safety, and information security, enabling decision-makers to adjust policies promptly based on public opinion dynamics and address potential security risks. Results show that the method proposed in this application has broad application prospects in regional security assessment, public opinion monitoring, and intelligent decision-making, providing strong support for sustainable social development. Attached Figure Description

[0047] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed in the embodiments. Obviously, the drawings described below only illustrate some of the embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0048] Figure 1 A flowchart of a regional safety evaluation method based on public opinion heat diffusion trend analysis according to an embodiment of the present application. DETAILED DESCRIPTION

[0049] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0050] In view of the defects in the prior art, the present application provides a regional safety evaluation method, device, equipment, medium and product based on public opinion heat diffusion trend analysis, to improve the objectivity, timeliness and accuracy of regional safety evaluation.

[0051] In order to make the above-mentioned purposes, features and advantages of the present application more apparent and easy to understand, the present application will be further described in detail below with reference to the drawings and specific embodiments.

[0052] In an exemplary embodiment, as shown in Figure 1 A regional safety evaluation method based on public opinion heat diffusion trend analysis is provided, including the following steps 1 to 8.

[0053] Step 1: Obtain public opinion information on an Internet platform; the public opinion information includes text information, video information, audio information and image information.

[0054] The Internet platform can be a platform for netizens to express their opinions, such as regional forums, social media, Q&A platforms, vertical communities in various industries, video platforms and online evaluation platforms. The acquisition of public opinion information needs to comply with the data protection regulations and policies of the place of residence, and needs to be authorized by the corresponding device owner. Specifically, panel data (text information, audio information, video information, image information, etc.) of a certain time sequence can be crawled or obtained by applying for API from mainstream social media, regional forums, video platforms and network evaluation platforms, so as to obtain public opinion information on various Internet platforms, also known as public opinion data.

[0055] Step 2: Data cleaning, feature extraction and formatting processing are performed on the public opinion information to extract the formatted information features of each piece of information.

[0056] The obtained public opinion data is cleaned, feature extracted and formatted, and useless data and burr data are cleaned out. Then, the key attribute features (including region, event, personnel, keyword, time, topic, etc.) of the data are extracted using feature engineering. The text information (or data) can use TF-IDF vector or BERT, etc. NLP (Natural Language Processing) model to extract text features, the image information can use ResNet or VGG, etc. CNN (Convolutional Neural Network) model to extract image features, the audio information can use audio processing methods such as Mel-Spectrogram or MFCC, combined with RNN (Recurrent Neural Network) or CNN to extract audio features, and the video data is divided into frame-by-frame images and audio and processed using a dual-stream model to extract image and audio features. Further, deep learning technology is used to extract the key attribute features of each piece of information, such as topic, time, place (region), personnel, keyword and specific event, from the text features, image features or audio features. Finally, the extracted key attribute feature data is formatted into the following computer recognizable form as the formatted information features of each piece of information.

[0057] Let B be a set of data structures, where each element b can be represented as a seven-tuple (i, a, e, p, k, t, s), where i represents the number, which can be an element in the set of integers Z. a represents the region, which can be an element in the set of regions A. e represents the event, which can be an element in the set of events E. p represents the personnel, which can be an element in the set of personnel P. k represents the keyword, which can be an element in the set of keywords K. t represents the time, which can be an element in the set of times T, usually represented in date-time format. s represents the topic, which can be an element in the set of topics S. Therefore, the data structure B of the formatted information features of each piece of information can be represented as:

[0058] B = {(i, a, e, p, k, t, s) | i ∈ Z, a ∈ A, e ∈ E, p ∈ P, k ∈ K, t ∈ T, s ∈ S} (1)

[0059] Here, Z, A, E, P, K, T and S are the sets of numbers, regions, events, personnel, keywords, times and topics, respectively. Each piece of information, regardless of its media format, is formatted into a unified format after feature extraction to facilitate subsequent processing.

[0060] Step 3: Based on the formatted information features of each piece of information, classify each piece of information into each indicator of the regional security index.

[0061] The various indicators of the regional security index (RSI) include crime rate (CR), public security (PS), health security (HS), economic security (ES), ecological security (EC), political security (POL), social security (SS), traffic safety (TS), and information security (IS), etc. The calculation method of the regional security index RSI is as follows:

[0062] RSI = w_1*CR + w_2*PS + w_3*HS + w_4*ES + w_5*EC + w_6*POL + w_7*SS + w_8*TS + w_9*IS (2)

[0063] wherein RSI is the value of the regional security index. CR, PS, HS, ES, EC, POL, SS, TS, and IS are the values of the nine indicators of crime rate, public security, health security, economic security, ecological security, political security, social security, traffic safety, and information security, respectively, and W_m (m = 1, 2, 3, 4, 5, 6, 7, 8, 9) represents the corresponding weight coefficient. By substituting the initial values of CR, PS, HS, ES, EC, POL, SS, TS, and IS of a certain region and the corresponding weight coefficients into formula (2), the initial regional security index of the region can be calculated. The RSI index of a region is used to evaluate the security of the region.

[0064] Based on the formatted information features of each piece of information, a classifier constructed using a Transformer obtains the probability of each piece of information belonging to each indicator in the regional security index, and determines the indicator corresponding to the maximum probability as the indicator to which the piece of information belongs, i.e., each piece of information is mapped to one independent variable (indicator) in the RSI.

[0065] Step 4: Score the formatted information features of each piece of information to obtain the security evaluation score of the indicator to which each piece of information belongs.

[0066] The SVR scorer or BERT is used to score the formatted information features (person, keyword, event, location, time, topic) of each piece of information, and the score is in the interval [-5, 5]. The score can be normalized on the basis of the score to obtain the safety evaluation score of the index to which each piece of information belongs. The score of the safety evaluation score represents the safety evaluation degree reflected by the information. A negative value represents a negative impact, a positive value represents a positive impact, and the larger the absolute value, the greater the impact.

[0067] Step 5: Cluster each piece of information based on the formatted information features of each piece of information and sort them in chronological order to identify a set of key events.

[0068] According to the formatted information features of each piece of information, the information is clustered and sorted in chronological order, which facilitates subsequent regional safety evaluation. Specifically, the keywords and events of each piece of information extracted previously can be clustered to obtain the events contained in all the obtained information, i.e., what event did the netizen comment on, and so on. Then, according to the clustering results, the events of each category are sorted by time, so that the development trend of the event can be obtained. The clustering set involving a large number of events or some hot events is identified as a set of key events. Subsequently, the trend index of each key event (i.e., the development index of each category) is calculated.

[0069] Step 6: Calculate the event trend index of each key event in the set of key events.

[0070] The calculation of the event trend index (ETI) of the present application involves the following influencing factors.

[0071] Sentiment Score S t : reflects the change of public emotion related to the event.

[0072] Spread Rate R t : measures the spread speed of the event on the Internet.

[0073] Volatility V t : reflects the volatility of public sentiment caused by the event.

[0074] Event Entropy H t : measures the complexity and uncertainty of the event public opinion state.

[0075] Temperature T t : reflects the "intensity" of the event public opinion.

[0076] Statistical Trend Index STI t : Trend change index calculated based on time series data, reflecting the changing trend of public opinion on events.

[0077] Specifically, for each key event n in the key event set, its sentiment score S t is calculated using the following formula:

[0078]

[0079] where N t is the number of newly added public opinion information related to the key event within time t; s i is the safety evaluation score of the i-th information within time t.

[0080] The event propagation rate R t of each key event is calculated using the following formula:

[0081]

[0082] where T is the time window, such as one hour or one day.

[0083] The public opinion fluctuation index V t is calculated using the following formula:

[0084]

[0085] The public opinion fluctuation index V t is the standard deviation of the safety evaluation score s i within time t, measuring the fluctuation of the public opinion sentiment change caused by the event.

[0086] The event entropy H t is calculated using the following formula:

[0087]

[0088] where k1 is the first proportional coefficient; P(E i ) is the probability distribution of different safety evaluation scores s i (such as positive (0, 5], negative [-5, 0), neutral 0) within time t.

[0089] The event temperature T t is calculated using the following formula:

[0090]

[0091] where k2 is the second proportional coefficient, representing the relationship between the intensity of sentiment and the activity of public opinion.

[0092] The statistical trend index STI of time t is calculated using the following formula t :

[0093] STI t = κ * S t + (1-κ) * STI t-1 (8)

[0094] wherein κ is a smoothing coefficient, 0<κ<1; STI t-1 is the statistical trend index of the previous time point t-1.

[0095] According to the sentiment score S t , the event propagation rate R t , the public opinion fluctuation index V t , the event entropy H t , the event temperature T t and the statistical trend index STI t of each key event n, the event trend index ETI t of the key event n is calculated using the following formula t :

[0096] ETI t = α * S t + β * R t + γ * V t + δ * H t + ε * T t + ζ * STI t (9)

[0097] wherein α, β, γ, δ, ε, ζ are weight coefficients of S t , R t , V t , H t , T t and STI t , respectively, indicating the influence degree of each factor on ETI.

[0098] The definition and calculation of the event trend index (ETI) of the present application adopts multiple factors, including basic indicators obtained by statistics and indicator offsets obtained based on the temperature diffusion model in thermodynamics, comprehensively reflecting the public opinion dynamics of the event and its influence on regional security. Specifically, the method of the present application combines the dynamic changes of public opinion data with the temperature concept in the thermodynamic model, quantifies the public opinion heat and diffusion speed of the event through the event temperature T t and the diffusion model, thereby comprehensively reflecting the public opinion dynamics of the event and its influence on regional security. This innovative method combining statistics and thermodynamic model can more comprehensively describe the evolution process of public opinion and its influence on regional security.

[0099] In particular, if the event trend index ETI t > 0, it indicates that the event public opinion is developing positively, the public mood tends to be positive, or the information propagation speed slows down. If the event trend index ETI t < 0, it indicates that the event public opinion is developing negatively, the public mood tends to be negative, or the information propagation speed speeds up, which may trigger a crisis and needs attention.

[0100] Step 7: Calculate the updated amount of influence of each key event on the regional security index based on the event trend index of each key event and calculate the updated regional security index.

[0101] Update the score of each new piece of information into the regional security index. Here, the diffusion-reaction equation summarized in physics is referred to to simulate the development (diffusion) of public opinion. The process of updating the regional security index is introduced as follows.

[0102] Let C(x, t) be the intensity of public opinion information at position x and time t, also known as public opinion concentration or heat; D be the diffusion coefficient, describing the speed of information propagation; R be the reaction rate, representing the decay or reaction rate of information in the propagation process. Then we have:

[0103]

[0104] For the public opinion concentration (i.e. the intensity of public opinion information) C(x, t), define its initial condition as:

[0105] C(x, 0) = S current *w (11)

[0106] where S current is the score of the current event, i.e. the score s i in step 4, and w is a weight factor related to the influence of public opinion, which is usually determined through historical data.

[0107] The boundary condition is:

[0108] C(0, t) = C(L, t) = 0 (12)

[0109] where L is the length of the region.

[0110] For the reaction rate, define:

[0111] R = R0*f network *f activity *f content (13)

[0112] where R0 is the basic propagation speed, which is the initial speed calculated based on the network and time; f networkis the network structure factor, reflecting the influence of network connectivity on the propagation speed, and its value ranges from 0 to 1; f activity is the user activity factor, reflecting the activity of users in the social network, and its value ranges from 0 to 1; f content is the information content factor, reflecting the influence of information type on the propagation speed, and its value ranges from 0 to 1. An upper limit needs to be set for the reaction rate R, and the part exceeding the limit is regarded as a boundary value.

[0113] By solving the solution of the diffusion-reaction equation (10), the public opinion concentration C(x, t) of a key event n at time t can be obtained, and the update amount can be calculated:

[0114]

[0115] which represents the update amount of the influence of key event n on the regional security index.

[0116] Further, the update formula is obtained:

[0117]

[0118] where represents the initial or updated (last period or last year) regional security index. The updated (current period or this year) regional security index is represented as ∑ represents the sum of the update amounts of all key events n. Here, the update is to add the influence of various events on regional security in the current period (this year) to to obtain the regional security evaluation in the current period (this year).

[0119] The present application uses the diffusion-reaction equation (model) to simulate the diffusion process of public opinion propagation, and updates the regional security index dynamically by solving the equation. The innovation of the model lies in that by combining the diffusion and reaction processes of public opinion information heat, real-time dynamic simulation of public opinion propagation and influence is realized. Specifically, the diffusion-reaction equation considers the propagation characteristics of public opinion information in space and time (such as diffusion coefficient D and reaction rate R), and updates the regional security index (RSI) by calculating the solution of the equation. This method can accurately simulate the heat diffusion trend of public opinion, and make the regional security evaluation more timely and accurate.

[0120] Step 8: Perform regional security evaluation based on the updated and updated regional security index and the event trend index of each key event.

[0121] First, the regional security index of the current period and the last period is By comparing, the direction of the current period regional security change can be obtained. If the regional security is getting better, the current decision can be maintained, or the corresponding decision can be made according to the events by checking the influence degree of each event on RSI in the current period. If the regional security is getting worse, the approximate location of the problem can be found by checking the influence degree of each event on RSI in the current period, and further processing is performed on the events that have a negative impact on the regional security. For example, it is found that the regional security is getting worse compared to last year in N years. By comparing the change values of each indicator this year and last year, it is found that the change value of PS (public security) is larger. Then, by checking the events extracted from the PS (public security) part, the corresponding decision is made according to the scores (sentiment scores, ETI, etc.) in the events. For example, it is found that a building in the region has collapsed, causing certain casualties. At this time, corresponding processing can be performed in combination with the data of on-site investigation, such as pacifying the nearby people and resettling the wounded.

[0122] The present application combines multi-modal public opinion information, including data from social media, forums and video platforms, and realizes comprehensive evaluation of regional security through deep learning technology. The event trend index (ETI) is introduced to effectively integrate key indicators such as sentiment scores, transmission rates and fluctuation indexes, and to reflect the development trend of events in real time. At the same time, the diffusion-reaction equation is used to simulate the spread of public opinion information, providing a scientific basis for the update of the regional security index (RSI). The present application method comprehensively evaluates factors such as crime rate, public security, health safety, etc., enabling decision-makers to adjust policies in a timely manner according to public opinion dynamics and to respond to potential security risks. The results show that the proposed method has wide application prospects in the fields of regional security evaluation, public opinion monitoring and intelligent decision-making, and provides strong support for the sustainable development of society.

[0123] A specific embodiment of the present application method is provided below, including the following processes.

[0124] Suppose the regional security situation of F city needs to be evaluated. Recently, a night robbery event that attracted widespread public attention occurred in the city, and the government took relevant measures to strengthen public security management. The present application will use public opinion data obtained from social media, regional forums, Q&A platforms, etc. to dynamically update the regional security index (RSI) of F city based on the event trend index (ETI) and the diffusion-reaction equation.

[0125] S1: Calculate the initial (before updating) regional security index

[0126] The weights of each indicator included in the regional security index (RSI) of F city are as follows:

[0127] Crime rate (CR): 0.3;

[0128] Public security (PS): 0.4;

[0129] Health safety (HS): 0.1;

[0130] Economic safety (ES): 0.05;

[0131] Ecological safety (EC): 0.02;

[0132] Political safety (POL): 0.01;

[0133] Social safety (SS): 0.01;

[0134] Traffic safety (TS): 0.005;

[0135] Information safety (IS): 0.005.

[0136] The initial values of each indicator are as follows:

[0137] Crime rate (CR): 0.8;

[0138] Public security (PS): 0.7;

[0139] Health safety (HS): 0.9;

[0140] Economic safety (ES): 0.85;

[0141] Ecological safety (EC): 0.95;

[0142] Political safety (POL): 0.9;

[0143] Social safety (SS): 0.9;

[0144] Traffic safety (TS): 0.95;

[0145] Information safety (IS): 0.9.

[0146] Substitute the above values into formula (2) to obtain the initial calculation formula of RSI:

[0147]

[0148] S2: Obtain public opinion information on the Internet platform.

[0149] The relevant public opinion information of F city in the past week was obtained from the Internet platform, including:

[0150] Text information on a social media:

[0151] "Last night, a robbery incident occurred in the center of F city, which really made people nervous!"

[0152] "The public security management is strengthened, everyone is discussing this, and feels that it will improve."

[0153] Text information on the regional forum:

[0154] "I encountered a robbery while walking at night, so I suggest everyone to be careful."

[0155] Image information: User-uploaded live photos, such as police patrols at the incident site.

[0156] Video information: Related videos from local news reports, showing the occurrence of the robbery and subsequent police actions.

[0157] S3: Data cleaning, feature extraction, and formatting.

[0158] Perform data cleaning to remove advertisements, spam, and irrelevant comments, ensuring that the information in the data is meaningful, such as only retaining descriptions of real events.

[0159] Extract key attributes such as region, event, personnel, keywords, time, and theme from text information, and use the BERT model to score them. For example, the first comment's safety evaluation score is -0.7 (negative), and the second comment's safety evaluation score is 0.4 (positive).

[0160] For image information, use the ResNet model to extract image features and identify patrolling police and incident sites in the images. For video information, use a dual-stream model to extract image features frame by frame and analyze the event context by combining audio features.

[0161] After extraction, format the public opinion information features into the data structure shown in formula (1): B = {(i, a, e, p, k, t, s) | i ∈ Z, a ∈ A, e ∈ E, p ∈ P, k ∈ K, t ∈ T, s ∈ S}. Format the information features of each piece of information as a seven-tuple, containing event number, region, event, personnel, keywords, time, and theme.

[0162] S4: Information classification and scoring.

[0163] Use the Transformer-based classifier to classify each piece of information into the various indicators of RSI. Then, use BERT to score the formatted information features of each piece of information. Suppose the following results are obtained:

[0164] Information 1 (negative): Classified as Public Security (PS), score -4;

[0165] Information 2 (positive): Classified as Public Security (PS), score 2;

[0166] Information 3...

[0167] The information is mapped to the respective independent variables of the RSI according to the safety evaluation score.

[0168] S5: Clustering and ordering.

[0169] The extracted information is clustered. It is assumed that the following key events are identified:

[0170] Key event 1: a night-time robbery event that caused a large public opinion fluctuation and strong public emotional response.

[0171] Key event 2: the F city government's measures to strengthen public security management received more positive evaluations.

[0172] Then, centering on these two events, similar events are clustered together, and then ordered according to time sequence, facilitating subsequent regional safety evaluation.

[0173] S6: Calculate event trend index.

[0174] Assume the weight coefficients are: a = 0.3, b = 0.25, g = 0.2, d = 0.1, e = 0.1, z = 0.05. For key event 1 and key event 2, the event trend index ETI t .

[0175] Where, the sentiment score S t : the sentiment score of key event 1 is -0.7. The sentiment score of key event 2 is 0.4.

[0176] Event propagation rate R t :

[0177] Set the time window to one day, the number of new public opinion information N t about key event 1 is 200 (social media, forum comments, etc.), so the propagation rate of event 1 is 0.2 after normalization.

[0178] The number of new public opinion information N t about key event 2 is 80, so the propagation rate of event 2 is 0.08 after normalization.

[0179] Public opinion fluctuation index V t According to the standard deviation of the sentiment score, it is assumed that the standard deviation of key event 1 is 0.4 and the standard deviation of key event 2 is 0.2.

[0180] Event entropy H t :

[0181] For key event 1, assume its safety evaluation score s iThe distribution of positive, negative, and neutral is 0.1, 0.8, and 0.1, respectively, and let k1=-1, so the event entropy is:

[0182] H t = -∑P(E i )logP(E i ) = -[0.1log(0.1) + 0.8log(0.8) + 0.1log(0.1)] ≈ 0.32

[0183] For key event 2, assume that the safety evaluation score s i The distribution of positive, negative, and neutral is 0.7, 0.1, and 0.2, respectively, so the event entropy is:

[0184] H t = -∑P(E i )logP(E i ) = -[0.7log(0.7) + 0.1log(0.1) + 0.2log(0.2)] ≈ 0.48

[0185] Event temperature T t : The event temperature of key event 1 is 0.7, and the event temperature of key event 2 is 0.4.

[0186] Calculate ETI:

[0187] For key event 1 (denoted as "robbery"), we have:

[0188] ETI 抢劫 = 0.3×(-0.7) + 0.25×0.2 + 0.2×0.4 + 0.1×0.32 + 0.1×0.7 + 0.05×1 = 0.072

[0189] For key event 2 (denoted as "public security"), we have:

[0190] ETI 治安 = 0.3×0.4 + 0.25×0.08 + 0.2×0.2 + 0.1×0.48 + 0.1×0.4 + 0.05×1 = 0.318

[0191] S7: RSI update process.

[0192] Use the diffusion-reaction equation (10) to simulate the influence of public opinion, assume diffusion coefficient D=0.5, reaction rate R=0.3 (set the limit rate as 1000, f network , f activity , f content Here, all are considered as 1), and the initial public opinion concentration C(0,0)=-0.7. The solution gives the influence of robbery event C 抢劫=-0.5; Impact of enhanced security measures C 治安 =+0.2.

[0193] The impact update of the robbery event:

[0194]

[0195] Impact of security measures (update):

[0196]

[0197] The result is obtained by calculating S1. Substituting the update values ​​of key events 1 and 2 into formula (15), we get:

[0198]

[0199] S8: Conduct a regional security assessment.

[0200] After considering the impact of the two key events, the Regional Safety Index (RSI) for City F was updated from the initial value of 0.69875 to 0.72635. This result indicates that although the robbery had a negative impact on the regional safety index (lowering the RSI by 0.036), the positive impact of security measures (increasing the RSI by 0.0636) was more significant. Therefore, the overall regional safety index improved, indicating that security measures effectively enhanced public safety, offsetting the negative impact of the robbery and ultimately leading to a slight increase in the RSI.

[0201] In one example embodiment, the application also provides a device for regional security evaluation based on public opinion heat diffusion trend analysis, comprising the following functional modules. A public opinion information acquisition module is configured to acquire public opinion information on an Internet platform; the public opinion information comprises text information, video information, audio information and image information; a public opinion information processing module is configured to perform data cleaning, feature extraction and format processing on the public opinion information to extract formatted information features of each piece of information; the formatted information features of each piece of information comprise a number, a region, an event, a person, a keyword, a time and a topic; a public opinion information classification module is configured to classify each piece of information into various indexes of a regional security index based on the formatted information features of each piece of information; the various indexes of the regional security index comprise a crime rate, a public order situation, health safety, economic safety, ecological safety, political safety, social safety, traffic safety and information safety; a public opinion information scoring module is configured to score the formatted information features of each piece of information to obtain a security evaluation score of an index to which each piece of information belongs; a public opinion information clustering module is configured to cluster each piece of information based on the formatted information features of each piece of information and sort the pieces of information in chronological order to identify a set of key events; an event trend index calculation module is configured to calculate an event trend index for each key event in the set of key events; a regional security index updating module is configured to calculate an update amount of an influence of each key event on the regional security index based on the event trend index of each key event and calculate an updated regional security index; and a regional security evaluation module is configured to perform regional security evaluation based on the regional security index before and after the update and the event trend index of each key event.

[0202] In one example embodiment, the application also provides a computer device, which can be a server or a terminal. The computer device comprises a processor, a memory, an input / output interface and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for running the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is configured to exchange information between the processor and external devices. The communication interface of the computer device is configured to communicate with external terminals through a network connection. The computer program is executed by the processor to implement the method for regional security evaluation based on public opinion heat diffusion trend analysis.

[0203] In one example embodiment, the application also provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the method for regional security evaluation based on public opinion heat diffusion trend analysis.

[0204] In one example embodiment, the application also provides a computer program product, which includes a computer program, and the computer program is executed by a processor to implement the method for regional security evaluation based on public opinion heat diffusion trend analysis.

[0205] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by computer program instruction related hardware, and the computer program can be stored in a non-volatile computer readable storage medium. When the computer program is executed, it can include the processes of the above-mentioned embodiment methods. Any reference to memory or other medium in the embodiments provided by the application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read only memory (Read Only Memory, ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive memory (ReRAM), magnetoresistive random access memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric memory (Ferroelectric Random Access Memory, FRAM), phase change memory (Phase Change Memory, PCM), graphene memory, etc. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, etc. As an illustration but not limitation, RAM can be in various forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), etc.

[0206] It should be noted that the information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the application are all information and data authorized by the user or authorized by all parties, and the collection, use and processing of related data need to comply with relevant regulations.

[0207] The application obtains public opinion data from platforms where netizens publish opinions, such as regional forums, social media, question and answer platforms, vertical communities in various industries, video platforms, and online evaluation platforms. After data cleaning, the features of the data are extracted from the data using deep learning technology, including personnel, keywords, events, locations, times, and topics. Then, the processed data is evaluated by region, modality, and event to obtain the safety score of the region and the influence degree of each event. This can help users have a basic understanding of the safety of the region and optimize the safety of the region and support other aspects of decision-making, including but not limited to economy, ecology, etc. It can be used in scenarios such as regional safety evaluation, regional development decision-making, and AI-mediated decision-making.

[0208] The technical features of the above embodiments can be combined in any manner. To make the description concise, not all possible combinations of the technical features in the above embodiments are described, but as long as the combinations of the technical features do not contradict, they should be considered within the scope of the present application.

[0209] The principles and implementation modes of the present application are described by applying specific examples herein. The above descriptions of the embodiments are only used to help understand the method of the present application and its core idea. Meanwhile, for those skilled in the art, the specific implementation modes and application ranges can be changed according to the idea of the present application. In summary, the content of the present application should not be understood as a limitation.

Claims

1. A regional security evaluation method based on public opinion heat diffusion trend analysis, characterized in that, The method comprises the following steps: Obtaining public opinion information on an Internet platform; The public opinion information comprises text information, video information, audio information and image information; Data cleaning, feature extraction and format processing are performed on the public opinion information to extract the format information features of each piece of information; the format information features of each piece of information comprise a number, a region, an event, a person, a keyword, a time and a theme; Classification is performed based on the format information features of each piece of information, and each piece of information is classified into each index of a regional security index; the indexes of the regional security index comprise a crime rate, a public order situation, a health safety, an economic safety, an ecological safety, a political safety, a social safety, a traffic safety and an information safety; Scoring is performed on the format information features of each piece of information to obtain a safety evaluation score of the index to which each piece of information belongs; Clustering and time sequence sorting are performed on each piece of information based on the format information features of each piece of information to identify a set of key events; An event trend index is calculated for each key event in the set of key events; The calculation of the event trend index for each key event in the set of key events comprises: For each key event in the key event set, the formula is used to calculate its sentiment score S t ; wherein N t is the number of public opinion information related to the key event added in time t; s i is the safety evaluation score of the i-th information in time t; The event propagation rate R for each key event is calculated using the formula t ; where T is the time window.​ The public opinion fluctuation index V is calculated by the formula t ;​ The event entropy H is calculated using the formula t ; where k1 is a first proportionality coefficient; P(E i ) is the probability distribution of different safety evaluation scores s i at time t; and k2 is a second proportionality coefficient.​ The event temperature T is calculated using the formula t ; where k2 is a second proportional coefficient;​ Using the formula STI t =κ*S t +(1-κ)*STI t-1 Calculate the statistical trend index (STI) for time t. t Where κ is the smoothing coefficient; STI t-1 It is the statistical trend index of the previous time point t-1; According to the sentiment score S of each key event t , the event propagation rate R t , the public opinion fluctuation index V t , the event entropy H t , the event temperature T t , and the statistical trend index STI t , the event trend index ETI t of the key event is calculated by the formula ETI t = α * S t + β * R t + γ * V t + δ * H t + ε * T t + ζ * STI t ; wherein α, β, γ, δ, ε, ζ are weight coefficients of S t , R t , V t , H t , T t , and STI t , respectively. An update amount of the influence of each key event on the regional security index is calculated based on the event trend index of each key event, and an updated regional security index is calculated; The calculation of the update amount of the influence of each key event on the regional security index based on the event trend index of each key event and the calculation of the updated regional security index comprise: The development of public opinion is simulated based on a diffusion-reaction equation of physics to calculate a public opinion concentration C(x, t) of a key event n at time t; The formula is used to calculate the update of the influence of the key event n on the regional safety index The updated zone safety index is calculated using the formula where is the zone safety index before the update;​ Regional safety evaluation is performed based on the updated regional security index and the event trend index of each key event. 2.The regional security evaluation method based on public opinion heat diffusion trend analysis of claim 1, wherein, The data cleaning, feature extraction and format processing of the public opinion information to extract the format information features of each piece of information comprise: Data cleaning is performed on each piece of information in the public opinion information to clean away useless data and burr data, and then text features, image features or audio features of each piece of information are extracted by using feature engineering; A region, an event, a person, a keyword, a time and a theme contained in each piece of information are identified based on the text features, image features or audio features of each piece of information, and are expressed in the form of a seven-tuple (i, a, e, p, k, t, s) as the format information features of each piece of information; wherein i, a, e, p, k, t and s represent the number, the region, the event, the person, the keyword, the time and the theme of each piece of information, respectively. 3.The regional security evaluation method based on public opinion heat diffusion trend analysis of claim 1, wherein, The classification of each piece of information into each index of the regional security index based on the format information features of each piece of information comprises: A probability that each piece of information belongs to each index of the regional security index is obtained by using a classifier constructed by a Transformer based on the format information features of each piece of information, and the index corresponding to the maximum probability is determined as the index to which the piece of information belongs. 4.The regional security evaluation method based on public opinion heat diffusion trend analysis of claim 1, wherein, The scoring of the format information features of each piece of information to obtain the safety evaluation score of the index to which each piece of information belongs comprises: The SVR scorer or BERT is used to score the formatted information features of each information, and the score is in the interval [-5, 5], to obtain the safety evaluation score of the index to which each information belongs.

5. A regional security evaluation device based on public opinion heat diffusion trend analysis, characterized in that, The method for realizing the regional safety evaluation based on public opinion heat diffusion trend analysis according to any one of claims 1-4; The regional safety evaluation device based on public opinion heat diffusion trend analysis comprises: An opinion information acquisition module is configured to acquire public opinion information on an Internet platform; the public opinion information comprises text information, video information, audio information and image information; An opinion information processing module is configured to perform data cleaning, feature extraction and formatting processing on the public opinion information, and extract formatted information features of each information; the formatted information features of each information comprise a number, a region, an event, a person, a keyword, a time and a theme; An opinion information classification module is configured to classify each information into each index of a regional safety index based on the formatted information features of each information; the indexes of the regional safety index comprise a crime rate, a public order situation, health safety, economic safety, ecological safety, political safety, social safety, traffic safety and information safety; An opinion information scoring module is configured to score the formatted information features of each information, to obtain a safety evaluation score of the index to which each information belongs; An opinion information clustering module is configured to cluster each information based on the formatted information features of each information and sort them in chronological order, to identify a set of key events; An event trend index calculation module is configured to calculate an event trend index for each key event in the set of key events; A regional safety index updating module is configured to calculate an update amount of an influence of each key event on the regional safety index based on the event trend index of each key event, and calculate an updated regional safety index; A regional safety evaluation module is configured to perform regional safety evaluation based on the regional safety index before and after the update and the event trend index of each key event.

6. A computer device comprising: A memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to realize the method for realizing the regional safety evaluation based on public opinion heat diffusion trend analysis according to any one of claims 1-4.

7. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to realize the method for realizing the regional safety evaluation based on public opinion heat diffusion trend analysis according to any one of claims 1-4.

8. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to realize the method for realizing the regional safety evaluation based on public opinion heat diffusion trend analysis according to any one of claims 1-4.

Images