Regional risk assessment method based on alarm information
By constructing a multi-level spatial resolution regional association network and conducting risk assessment based on police information, the problem of one-sidedness and lag in risk assessment in existing technologies is solved. This enables accurate identification and diversity analysis of risk propagation paths, thereby improving the effectiveness of emergency response and decision support.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-03
- Publication Date
- 2026-03-13
AI Technical Summary
Existing technologies lack the ability to conduct dynamic risk assessments with multiple dimensions and spatial resolutions during emergencies, resulting in one-sided and delayed early warning results. They also make it difficult to integrate heterogeneous data from multiple sources and accurately depict the risk propagation path and scope of impact.
By collecting police incident information, a multi-level spatial resolution regional correlation network is constructed. By combining the geographical information of the location where the incident occurred, the place of residence, and the workplace, the regional risk index is quantitatively assessed, risk hotspots and the diversity of their sources are identified, and the assessment threshold is dynamically adjusted.
It has achieved the accuracy and coverage of multi-dimensional risk assessment, improved the adaptability and decision-making efficiency of emergency response, supported differentiated prevention and control strategies, and enhanced the response speed and resource allocation efficiency of public safety incidents.
Smart Images
Figure CN121660451A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of public safety and emergency management information technology, specifically a method for regional risk assessment based on police information. Background Technology
[0002] Quantitative assessment of regional risk is a crucial requirement for improving the emergency management system for sudden incidents. Emergency information generated by a sudden incident in its affected area is disseminated by affected individuals, subsequently impacting other areas. Emergency information generated by direct participants or witnesses in the face of a crisis is one of the direct pieces of evidence of the impact of a sudden incident on its affected entities. The incident can be a natural disaster, an accident, a public health emergency, or a social security incident, etc., and the emergency information can be the stress response or dissemination behavior of affected individuals under the influence of the incident. Identifying the spatiotemporal patterns of individual behavior and predicting abnormal group situations and trends can help assess the initial impact of a sudden incident, assist in the allocation of emergency resources, and thus improve the overall effectiveness of the emergency response system.
[0003] However, existing methods have not fully utilized the data resources generated by relevant systems for comprehensive regional risk analysis and assessment in practice. Current analysis and early warning methods mostly rely on single data sources or static indicators, lacking the ability to dynamically assess risks across multiple dimensions and spatial resolutions, resulting in some bias and lag in early warning results. Especially when facing complex and ever-changing emergency scenarios, existing technologies struggle to effectively integrate multi-source heterogeneous data such as "people, places, things, and events," failing to accurately depict the risk propagation paths and impact ranges between regions, thus limiting the scientific rigor and foresight of emergency management decisions.
[0004] Therefore, there is an urgent need for a regional risk analysis method that can make full use of existing information system data resources, has multi-scale risk assessment capabilities, and can dynamically optimize early warning thresholds, in order to improve the accuracy of early warning and the efficiency of emergency response to public safety incidents. Summary of the Invention
[0005] To address the problems of existing technologies, this invention provides a method for regional risk assessment based on police information.
[0006] To solve the above-mentioned technical problems, the present invention is achieved through the following technical solution: a regional risk assessment method based on police information, characterized by comprising the following steps: S1: Basic Data Acquisition and Mapping Collect police report data from affected individuals in the incident, and record it as... ,in As a unique alarm identifier, This is the timestamp of the incident. The geographical coordinates of the location where the incident occurred; The geographical information of the individual's permanent residence and workplace is collected and denoted as follows: ,in The geographical coordinates of an individual's permanent residence. The geographical coordinates of the individual's workplace; The coordinates of the location where the incident occurred Coordinates of permanent residence and work location coordinates By mapping these data to predefined administrative region identifiers at multiple spatial resolution levels, two structured datasets are obtained: Dataset of the location of police incidents Its data structure is ,in , , , These are unique codes representing the provincial, municipal, district / county, and township / street-level administrative regions where the incident occurred; Dataset of Geographical Attribution of Police Incidents Its data structure is ,in to These represent the provincial to township / street level administrative codes of the individual's place of residence. to These represent the provincial to township / street level administrative regions where the individual's workplace is located; S2: Construction of Regional Association Networks Based on the aforementioned datasets AER and AHO, at the specified spatial resolution level Next, construct a weighted directed graph, called the regional association network. ,in: This is a set of nodes containing all the geographic codes that appear at this resolution; For a set of directed edges, for each police incident record, create an edge starting from the geographical code of the incident occurrence (denoted as node). ) points to the corresponding individual's place of residence or place of work geographical code (denoted as node) The directed edges of ) ; Let the set of edge weights be the weights of each directed edge. weight All alarm records in the data are in this direction (from) arrive The sum of the number of connections; S3: Regional Security Risk Assessment Based on the aforementioned regional association network Perform the following quantitative assessment: Each geographical node in the computational network Comprehensive risk index ; b. Based on the aforementioned risk index A set of risk hotspot areas was identified. ; c calculates each regional node Risk source diversity index .
[0007] In one specific implementation, the spatial resolution level r includes provincial level r=1, prefecture-level level r=2, district / county level r=3 and township / street level r=4, and the regional codes AID(r) of each level are hierarchically encoded according to administrative affiliation, with the higher-level regional code forming the prefix of the lower-level regional code.
[0008] In one specific implementation, the process of mapping geographic coordinates to administrative region identifiers in step S1 includes: Use a spatial point-to-surface inclusion judgment algorithm to determine the township / street-level geographical polygon where the coordinate point is located. Based on the township / street-level regional code, the pre-stored administrative hierarchy table is used to automatically associate and obtain the district / county-level, prefecture-level, and provincial-level regional codes to which it belongs.
[0009] In one specific implementation, step S2 involves constructing a regional association network. The specific algorithms include: initialization It is an empty set. It is an empty set. All weights in the middle are initialized to 0; Traversal Each police report in the dataset; exist Find individual records with the same ID in the dataset; For each pair of matching records found, extract the data at resolution. The code i for the location where the incident occurred and the code j for the selected affected location (place of residence or place of work); like There is no such thing as from point to If there is a directed edge, then create that edge and add it. ; Side weight The value increases by 1.
[0010] In one specific implementation, the risk index The calculation formula is: in, Spatial distance similarity, used to attenuate the impact of geographical distance on risk propagation, is calculated using the following formula: For region and The Euclidean distance between the geometric centers; , , These are configurable influence control parameters, used to weight output risk, input risk, and internal risk, respectively, to meet the following requirements. and .
[0011] In one specific implementation, influence control parameters , , The harmonic average method is used to set the values as follows: in, Score the level of attention given to risk tracing. Score the level of concern regarding the spread of risk. The internal risk level is scored, with a rating of 1 to 4, corresponding to key concern, concern, moderate concern, and general concern, respectively.
[0012] In one specific implementation, a set of risk hotspot regions The recognition formula is: in, This is a dynamically adjustable proportional threshold. .
[0013] In one specific implementation, the risk source diversity indicator in step S3 The calculation process is as follows: First, construct the region Risk source feature vector: Each component represents the normalized proportion of output-type, input-type, and self-type risks in the total risk index, and the calculation formula is as follows: , , ; Subsequently, the diversity of the above distributions is calculated based on information entropy: Entropy The higher the value, the more complex and diverse the sources of risk in the region.
[0014] In one specific implementation, when constructing the network in step S2, the area affected by the alarm can be configured as the individual's place of residence or place of work to adapt to different application scenarios of assessing "impact on residents' lives" or "impact on economic activities".
[0015] In one specific implementation, the method can independently perform risk assessments at four spatial resolutions: provincial, municipal, district / county, and township / street, and output risk indices for each region. Hotspot region list and the diversity of risk sources The assessment report is used to assist emergency management departments in resource pre-positioning and tiered early warning decision-making.
[0016] The beneficial effects of this invention are as follows: 1. This invention integrates multi-dimensional geographic location data, including the location of the incident, the home and workplace of the affected individuals, to construct a four-level spatial resolution assessment system from the provincial to the township / street level, overcoming the limitations of traditional methods that rely on a single data source. By dynamically characterizing risk propagation paths through regional correlation networks, it organically combines micro-level individual behavior with macro-level regional risk trends, achieving comprehensive quantification of output-type, input-type, and self-type risks. This significantly improves the coverage dimensions and accuracy of risk assessment, providing more comprehensive data support for emergency warnings. 2. Based on the mapping relationship between police incident data and geographic information, this invention constructs a regional association network with clear directionality at different spatial resolutions. It quantifies the risk propagation intensity through edge weights and dynamically calculates the regional risk index by combining spatial distance similarity. This method can accurately identify risk hotspots and propagation paths. Furthermore, through parametric design, it supports management departments in adjusting the assessment focus as needed, significantly improving the system's adaptability to complex risk scenarios and the efficiency of source tracing analysis. 3. By defining risk hotspot geographic sets and risk source diversity indicators, this invention not only achieves the classification of regional risks but also reveals the diverse characteristics of risk composition. The risk source diversity indicator calculated by information entropy can intuitively reflect the complexity of risk structure, assisting management departments in formulating differentiated prevention and control strategies. In addition, the dynamic threshold adjustment mechanism and multi-resolution output results directly serve emergency resource pre-positioning and early warning decision-making, effectively improving the response speed and handling efficiency of public safety incidents. Attached Figure Description
[0017] Figure 1 This is a schematic diagram of the risk assessment system framework based on police incident data of the present invention.
[0018] Figure 2 This is a schematic diagram of the algorithm for constructing a regional association network according to the present invention. Detailed Implementation
[0019] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.
[0020] like Figures 1 to 2 This illustrates a regional risk assessment method based on police information.
[0021] This invention focuses on the geographical areas where emergencies occur and their disaster-bearing entities. It constructs a regional correlation network based on emergency information, defines a regional risk index, establishes the correlation between micro-level individual behavioral patterns and macro-level regional risks, and achieves multi-spatial resolution regional risk assessment. This provides a reference for decision-making related to emergency prevention and preparedness. The main steps of this invention are as follows: Basic data collection The regional security assessment and analysis conducted in this invention uses a spatial resolution of four granularities: provincial, municipal, district / county, and township / street. For ease of description, the regional identifiers are defined as follows: Definition 1 (Region): For regions at four resolutions: provincial, prefecture-level, district / county, and township / street, It is coded and identified, where, At the provincial level, At the prefecture-level city level, At the district / county level, For township / street level areas, the four-level regional coding format is shown in Table 1, where... Representing a single digit, each level of identifier is coded according to administrative affiliation. The highest digit represents the higher-level region to which it belongs, and the last two digits represent the internal code of this level.
[0022] Table 1. Regional Coding Format Geographical data of incidents: utilizing individual incident data related to the incidents. The geographical location of the incident These are mapped to four levels of geographic identifiers. Each location where an incident occurred has a unique attribution to each level of geographic polygon, forming a dataset of incident location attribution information. The specific method is as follows: Step 1: Data Preparation Collect and organize data on individual police reports related to the incident. ,in For police alerts, For time, Provide the geographical coordinates of the location where the incident occurred. Simultaneously, prepare four levels of regional polygon data, clearly defining the boundary coordinates of each region to form polygonal areas.
[0023] Step 2: Determine the relationship between the coordinate points and the polygon for Each coordinate point in the dataset The binary search method is used to determine whether a point lies within a township / street-level geographical polygon. Specifically, for each township / street-level geographical polygon, an arbitrary vertex is selected, and its coordinates are calculated relative to the given point. The distances are ranked from smallest to largest (those closest are judged first), and the binary search method is used to check the coordinate points in sequence. Check if it is inside the polygon. If it is outside the polygon, continue to check the next polygon. If it is inside the polygon, go to step three.
[0024] Step 3: Determining Attribution If coordinates If the location is within a certain township / street-level geographical polygon, then record the identifier of that region. Next, by querying a pre-established hierarchical relationship table or spatial database, the hierarchical relationship between the township / street level region and the district / county level region, the prefecture / city level region, and the provincial level region is queried, and the corresponding three-level regional identifiers are found and recorded in sequence. Secondary regional identifiers and primary regional identifiers .
[0025] Step 4: Dataset Construction Each police incident Time t and defined regional identifiers at various levels This information is combined into a single record and stored in a dataset containing information on the geographical location of the incident. middle.
[0026] Step 5: Result Validation and Optimization For the construction The dataset is validated to ensure that the geographical location of each incident is correctly mapped to the corresponding geographic polygon, and that the geographic affiliation at each level is accurate. If any errors or omissions are found, return to step two or three for correction.
[0027] Data on the geographical impact of police incidents: utilizing the geographic information of the homes and workplaces of individuals affected by the incident. Their location information is then mapped to four levels of geographic regions (using the same method as above) to form a dataset of geographic affiliation information for each individual's home and workplace. .
[0028] Regional network construction The regional association network constructed in this invention can be implemented at four different regional resolutions. A region is a node in the association network. Directed edges are established from the region where an individual's incident occurred to their home region. The number of directed connections established between regions for incident data at the group level is the weight of that directed edge. Here, a directed edge pointing from the incident location to the home can be used to discuss the impact of emergencies on residents; if it points from the incident location to the workplace, it can be used to discuss the impact of emergencies on workers. The former will be used as an example in subsequent explanations.
[0029] Definition 2 (Regional Association Network) It is a weighted directed network, in which It refers to spatial resolution, within a specified range. Down, It is a collection of regions. It is a set of directed edges between regions. It is a set of weighted directed edges. Please refer to the section on areas affected by police incidents, using the home as the reference. Figure 2 The construction algorithm is as follows: Algorithm 1: Regional Association Network Construction Algorithm Input: region collection The dataset of regions where the incident occurred (AER) and the dataset of regions affected by the incident (AHO) Output: Regional Association Network Initialize ; foreach do ; while do k←k+1; ; if then ; ; return ; The construction process of the algorithm is as follows: Figure 2 As shown: Regional security risk assessment Based on the aforementioned regional network, this invention defines regional risk index, risk hotspot regions, and the diversity of risk sources to assess the regional security risk situation.
[0030] Definition 3 (Regional Risk Index) at a specified spatial resolution down, region Regional risk index for: in, For resolution Spatial distance similarity, For region and The distance between the geometric centers, For influence control parameters, and satisfying... This is set according to the analytical needs of emergency management departments. A higher regional risk index indicates a greater overall impact of the event on the region, requiring close attention from emergency management departments. Here, the parameters... These represent the attention weights for output-type, input-type, and self-type risks, respectively, and are set using a harmonic average method: According to the "Emergency Response Law of the People's Republic of China," emergency warning levels are divided into four levels from high to low: Level I (Extremely Serious), Level II (Serious), Level III (Relatively Serious), and Level IV (General). This section assigns a risk level score to each of these four risk categories. Divided into four levels: 1 (Key Focus), 2 (Focus), 3 (Relatively Focus), and 4 (General Focus), then the parameters... The setup method is as follows: in Score the level of attention given to risk tracing. Score the level of concern regarding the spread of risk. Score the level of concern regarding internal risks.
[0031] Definition 4 (Risk Hotspot Regions): Based on the regional risk index, the set of hotspot regions is defined as follows: in, This method sets a threshold for management departments to assess hotspot areas. When the relative value of the risk index exceeds this threshold, the area is added to the hotspot area set. This method performs a simple binary classification of areas at different resolutions: hotspot areas and non-hotspot areas. This parameter can be customized by emergency management departments according to application needs. The basic steps for dynamically adjusting this parameter are as follows: Step 1: The management department (or special task force) takes the lead in clarifying the assessment objectives.
[0032] (1) Determine the types of hotspot areas to be monitored (such as urban core areas, transportation hubs, large event venues, densely populated communities, etc.); (2) Referring to industry standards such as the “Guidelines for the Construction of Urban Safety Risk Comprehensive Monitoring and Early Warning Platform”, the regional risk index is incorporated into the evaluation system and confirmed by the department head.
[0033] Step 2: Set the threshold adjustment range.
[0034] Based on historical data, the initial range of the undetermined threshold is set to [0.5, 1.0), with an initial value of 0.5 and an adjustment step size of 0.1. Hotspot classification modeling is performed for each threshold, and the classification results are output.
[0035] Step 3: Optimize the threshold through data validation.
[0036] (1) Data preparation: The actual data is the original data of the regional risk index for the past year; (2) Performance evaluation: Accuracy (the percentage of correctly classified regions), recall (the percentage of actual hotspot regions that are correctly identified), and F1 score are used as indicators to compare the classification performance under different thresholds; (3) Threshold selection: Prioritize the threshold with the highest F1.
[0037] Step 4: Threshold confirmation and dynamic adjustment.
[0038] The organization and management departments, along with frontline implementation teams (street offices, security departments, etc.), will hold a review meeting to determine the final threshold based on actual management needs (such as emergency response time ≤ 2 hours, maximum manpower for resource allocation, etc.). Subsequent quarterly threshold reviews will be conducted using newly collected regional risk index data and frontline feedback (such as false alarms and omissions). If the accuracy rate drops by more than 5%, or in the event of major policy adjustments or emergencies, the dynamic adjustment process will be initiated immediately.
[0039] Definition 5 (Sources of Regional Risk) The risk source characteristics at spatial resolution r can be represented by the following triplet: Among them, the first item Used to quantify the proportion of the safety impact of region i on residents of other regions in the total risk index; the second item Used to quantify the proportion of security impacts on residents of region i in other regions in the total risk index; the third item This is used to quantify the proportion of the safety impact of region i on residents within the region in the total risk index.
[0040] Based on the above three indicators, the diversity of risk sources in region i can be defined using information entropy: This indicator quantifies the diversity of outward, inward, and internal risk sources in a region, providing emergency management departments with recommendations for response strategies. A lower indicator suggests a single risk source, requiring focused attention on that specific type of risk; a higher indicator indicates diverse risk sources, necessitating comprehensive management.
[0041] Example 1. Basic Data Preparation (1) Data set of crime incident locations Police incident dataset ER sample (individual) Anonymization processing, in order to , ...Example): Example of AER dataset for crime scene location: (2) Dataset of regions affected by police incidents Geographic data set of homes and workplaces of individuals affected by the event Example: Data set of geographic affiliation information for individuals' homes and workplaces. Example: 2. Construction of regional interconnected networks (at spatial resolution) ,by (As an example of township / street-level regional identifiers) enter: ; node : (geographical collection).
[0042] Directed edge and weight : Traverse events in AER and match Chinese correspondence Home region, statistical edge weights: event Police incident area , home region →Add edge 1; event Police incident area Home region j=B→Add edge , ; event Police incident area Home region j=C → Add edge , .
[0043] Network results: W(4)={ =1, =1, =1}.
[0044] Output: Network .
[0045] 3. Calculation of Regional Risk Index Parameter settings: Influence control parameters: , , (Calculated from the harmonic mean, assuming) ); Spatial distance similarity: (Assuming) ).
[0046] Risk indices of various regions : Region A: =0.545×(1×0.7+1×0.6)+0.273×0+0.182×0=0.7085 Region B: =0.545×0+0.273×(1×0.7)+0.182×1=0.3731 Region C: =0.545×0+0.273×(1×0.6)+0.182×0=0.1638 area There are no related edges, so 4. Identification of risk hotspot areas (Risk index of region A); threshold (Assuming it is determined through the threshold adjustment step); Judgment conditions: : A: →Hot areas; B →Not a hot topic.
[0047] Result: HotA(4)={A}.
[0048] 5. Analysis of Regional Risk Sources Taking regions A and B as examples, calculate the characteristics and diversity of risk sources: Region A: Information entropy (The risk source is singular and output-oriented.)
[0049] Region B: Information entropy (The sources of risk are diverse, including both imported and self-related risks).
[0050] Final output: Regional Relationship Network : , , ; Risk Index: ; Hotspot areas: ; Risk sources are diverse: A entropy = 0 (single), B entropy = 0.691 (relatively diverse).
[0051] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.
Claims
1. A regional risk assessment method based on police information, characterized in that: Includes the following steps: S1: Basic Data Acquisition and Mapping Collect police report data from affected individuals in the incident, and record it as... ,in As a unique alarm identifier, The timestamp of the incident. The geographical coordinates of the location where the incident occurred; The geographical information of the individual's permanent residence and workplace is collected and denoted as follows: ,in The geographical coordinates of an individual's permanent residence. The geographical coordinates of the individual's workplace; The coordinates of the location where the incident occurred Coordinates of permanent residence and work location coordinates By mapping these data to predefined administrative region identifiers at multiple spatial resolution levels, two structured datasets are obtained: Dataset of the location of police incidents Its data structure is ,in , , , These are unique codes representing the provincial, municipal, district / county, and township / street-level administrative regions where the incident occurred; Dataset of Geographical Attribution of Police Incidents Its data structure is ,in to These represent the provincial to township / street level administrative codes of the individual's place of residence. to These represent the provincial to township / street level administrative regions where the individual's workplace is located; S2: Construction of Regional Association Networks Based on the aforementioned datasets AER and AHO, at the specified spatial resolution level Next, construct a weighted directed graph, called the regional association network. ,in: This is a set of nodes containing all the geographic codes that appear at this resolution; For a set of directed edges, for each police incident record, create an edge starting from the geographical code of the incident occurrence (denoted as node). ) points to the corresponding individual's place of residence or place of work geographical code (denoted as node) The directed edges of ) ; Let the set of edge weights be the weights of each directed edge. weight All alarm records in the data are in this direction (from) arrive The sum of the number of connections; S3: Regional Security Risk Assessment Based on the aforementioned regional association network Perform the following quantitative assessment: Each geographical node in the computational network Comprehensive risk index ; b. Based on the aforementioned risk index A set of risk hotspot areas was identified. ; c calculates each regional node Risk source diversity index .
2. The regional risk assessment method based on police information according to claim 1, characterized in that: The spatial resolution level Including provincial level prefecture-level District and county level and township / street level And regional codes at all levels The region is coded hierarchically according to its administrative affiliation, with the higher-level region code forming the prefix of the lower-level region code.
3. The regional risk assessment method based on police information according to claim 1, characterized in that: The process of mapping geographic coordinates to administrative region identifiers in S1 includes: Use a spatial point-to-surface inclusion judgment algorithm to determine the township / street-level geographical polygon where the coordinate point is located. Based on the township / street-level regional code, the pre-stored administrative hierarchy table is used to automatically associate and obtain the district / county-level, prefecture-level, and provincial-level regional codes to which it belongs.
4. The regional risk assessment method based on police information according to claim 1, characterized in that: In S2, a regional association network is constructed. The specific algorithms include: initialization It is an empty set. It is an empty set. All weights in the middle are initialized to 0; Traversal Each police report in the dataset; exist Find individual records with the same ID in the dataset; For each pair of matching records found, extract the data at resolution. The code i for the location where the incident occurred and the code j for the selected affected location (place of residence or place of work); like There is no such thing as from point to If there is a directed edge, then create that edge and add it. ; Side weight The value increases by 1.
5. The regional risk assessment method based on police information according to claim 1, characterized in that... The risk index The calculation formula is: in, Spatial distance similarity, used to attenuate the impact of geographical distance on risk propagation, is calculated using the following formula: For region and The Euclidean distance between the geometric centers; , , These are configurable influence control parameters, used to weight output risk, input risk, and internal risk, respectively, to meet the following requirements. and .
6. The regional risk assessment method based on police information according to claim 5, characterized in that: The influence control parameters , , The harmonic average method is used to set the values as follows: in, Score the level of concern regarding risk source tracing. Score the level of concern regarding the spread of risk. The internal risk level is scored, with a rating of 1 to 4, corresponding to key concern, concern, moderate concern, and general concern, respectively.
7. The regional risk assessment method based on police information according to claim 1, characterized in that, The set of risk hotspot regions The recognition formula is: in, This is a dynamically adjustable proportional threshold. .
8. The regional risk assessment method based on police information according to claim 1, characterized in that, The risk source diversity indicator in S3 The calculation process is as follows: First, construct the region Risk source feature vector: Each component represents the normalized proportion of output-type, input-type, and self-type risks in the total risk index, and the calculation formula is as follows: ， ， ； Subsequently, the diversity of the above distributions is calculated based on information entropy: Entropy The higher the value, the more complex and diverse the sources of risk in the region.
9. The regional risk assessment method based on police information according to claim 1, characterized in that, When constructing the network in step S2, the affected area of the alarm can be configured as the area of an individual's permanent residence or workplace to adapt to different application scenarios of assessing "impact on residents' lives" or "impact on economic activities".
10. The regional risk assessment method based on police information according to claim 1, characterized in that, The method can independently perform risk assessments at four spatial resolutions: provincial, municipal, district / county, and township / street, and output risk indices for each region. Hotspot region list and the diversity of risk sources The assessment report is used to assist emergency management departments in resource pre-positioning and tiered early warning decision-making.