A Method and Device for Collecting Urban Management Problems Based on Artificial Intelligence Large Language Model

Abstract

This invention belongs to the field of intelligent city management technology, and specifically discloses a method and device for collecting urban management issues based on an artificial intelligence large language model. The method includes: using an artificial intelligence large language model to identify the types of urban management sub-zones within the urban management area, identifying commercial and residential urban management sub-zones, performing spatial proximity analysis, extracting violation analysis points in adjacent commercial and residential areas, analyzing data collected on violation issues from these violation analysis points over multiple historical management cycles, filtering out adjacent violation points, and summarizing adjacent violation sequences. This allows the artificial intelligence large language model to prioritize these high-frequency and stable violation points when collecting urban management issues, enabling more targeted coverage of key areas and locations, improving collection and inspection efficiency and targeting.

Description

Technical Field

[0001] This invention belongs to the field of intelligent city management technology, specifically a method and device for collecting urban management problems based on an artificial intelligence large language model. Background Technology

[0002] With the acceleration of urbanization, the scale of cities is constantly expanding, and the population is becoming increasingly dense, urban management is facing an unprecedentedly complex situation. Urban management covers many areas, including but not limited to the appearance of the city, environmental sanitation, traffic order, and business operation regulations, involving different functional areas such as commercial areas and residential areas.

[0003] In existing technologies, urban management points at the boundary between commercial and residential areas do not consider their respective impacts on the commercial and residential areas. They simply count the number of violations within a certain range around the violation point without taking into account the functional differences between the two areas and their radiating influence. In commercial and residential areas, the same number of violations may have drastically different impacts on urban management due to differences in area function. For example, there are temporal and spatial differences between commercial and residential areas. Therefore, it is necessary to adjust the collection strategy according to the temporal patterns and spatial distribution of violations in commercial and residential areas to enable the collection and inspection work to cover key areas and key locations in a targeted manner, thereby improving the efficiency and relevance of collection and inspection. Summary of the Invention

[0004] To address the shortcomings of existing technologies and solve at least one of the technical problems mentioned in the background, this invention provides a method and apparatus for collecting urban management problems based on an artificial intelligence large language model.

[0005] The technical solution adopted by the present invention to solve the above-mentioned technical problems is as follows:

[0006] The first aspect is a method for collecting urban management issues based on artificial intelligence large language models, including:

[0007] Using an artificial intelligence large language model, the types of urban management sub-zones within the urban management area are identified, distinguishing between commercial urban management sub-zones and residential urban management sub-zones;

[0008] Spatial proximity analysis was performed on the identified commercial and residential urban management sub-zones to extract violation analysis points in adjacent commercial and residential areas. Violation analysis points from multiple historical management cycles were screened and analyzed to identify adjacent violation points and summarize the adjacent violation sequence.

[0009] Violation radiation analysis is performed on neighboring violation points within the adjacent violation sequence to identify violation radiation points, and radiation impact analysis is performed on violation radiation points to assess whether the impact of radiation is significant.

[0010] If the radiation impact is significant, then the time patterns of the illegal radiation points will be explored, urban management patrol periods will be set, and patrol routes will be developed. If the radiation impact is minor, patrols will be conducted according to the original patrol periods.

[0011] Preferably, the identification process for commercial urban management sub-zones and residential urban management sub-zones is as follows:

[0012] The regulations governing each urban management sub-district are input into an AI-powered large language model. The AI-powered large language model is then used to identify and extract keywords from the content of the regulations. Based on these keywords, the corresponding urban management sub-districts are categorized into commercial urban management sub-districts and residential urban management sub-districts.

[0013] Preferably, within the urban management area, any commercial urban management sub-zone is selected as the base sub-zone for merging;

[0014] Extract the urban management sub-areas adjacent to the base sub-area. If the urban management sub-area is a residential urban management sub-area, then merge them to obtain the adjacent commercial and residential first-level combined area.

[0015] Within the adjacent commercial and residential primary integration zone, arbitrarily select a commercial urban management sub-zone and extract the urban management sub-zone adjacent to the merged basic sub-zone. If the urban management sub-zone is a residential urban management sub-zone, then merge them to obtain the adjacent commercial and residential secondary integration zone.

[0016] Traverse all urban management sub-areas within the urban management area, and continuously iterate and merge according to the merging method of adjacent commercial and residential first-level combined area and adjacent commercial and residential second-level combined area until the urban management sub-area adjacent to the merged base sub-area is of the same type as the merged base sub-area, then do not perform the merging operation, and obtain the adjacent commercial and residential combined area.

[0017] Within the adjacent commercial and residential urban management area, arbitrarily extract the boundary edge points between the commercial urban management sub-area and the residential urban management sub-area as the basis for violation analysis.

[0018] Preferably, neighboring violation points are selected, and a neighboring violation sequence is obtained. The process is as follows:

[0019] The historical management cycle is divided into several equal historical management periods. The number of violations by each violation analysis point in each historical management period is obtained, and the ratio of the number of violations to the duration of the historical management period is calculated to obtain the violation frequency of each period.

[0020] The average frequency of violations at each historical management period is calculated by averaging the frequency of violations at each violation analysis point, and the average frequency of violations during each period is output.

[0021] The standard deviation of the violation frequency of each violation analysis point within each historical management period is calculated, and the standard deviation of the violation frequency within each period is output.

[0022] The ratio of the average violation frequency to the standard deviation of the violation frequency during a time period is calculated to obtain the violation stability value. If the violation stability value is less than or equal to the violation stability threshold, the analyzed violation points are taken as neighboring violation points, and each neighboring violation point is sorted in ascending order according to its corresponding violation stability value to construct a neighboring violation sequence.

[0023] Preferably, the process for determining illegal radiation points is as follows:

[0024] Arbitrarily select a neighboring violation point within the adjacent violation sequence as a suspected radiation point, and take the historical management period in which the suspected radiation point violated the violation as the suspected radiation period. Within the suspected radiation period, extract the violation points in the commercial urban management sub-zone and the violation points in the residential urban management sub-zone as commercial management violation points and residential management violation points, respectively.

[0025] During the suspected radiation period, the distance between the commercial management violation point and the suspected radiation point is obtained, and the ratio is calculated with the perimeter of the commercial city management sub-zone as the suspected radiation distance ratio.

[0026] The time interval between the occurrence of violations at commercial / residential violation points and the occurrence of violations at suspected radiation points is obtained as a proportion of the suspected radiation period, thus yielding the suspected radiation time interval ratio.

[0027] The suspected radiation interval ratio and suspected radiation distance ratio are summed, and the reciprocal is taken to output the suspected radiation value. If the suspected radiation value is greater than or equal to the suspected radiation threshold, the analyzed suspected radiation point is marked as an illegal radiation point.

[0028] Preferably, the process of analyzing the radiation impact of illegal radiation points is as follows:

[0029] Extract the historical management periods in which violations occurred at the violation radiation points as violation radiation periods. Calculate the ratio of the number of commercial violation points to the total number of commercial violation points in the commercial urban management sub-district, and the ratio of the number of private violation points to the total number of private violation points in the residential urban management sub-district, to obtain the ratio of commercial violation radiation to private violation radiation. Then sum them up to obtain the total number of violations radiation.

[0030] The duration of violations at each commercial and residential violation point within the period of violation radiation is obtained. The average duration of each violation point is calculated to obtain the ratio of duration of commercial violations and residential violations. The sum of these values ​​is then used to obtain the ratio of duration of violations radiating outwards.

[0031] Preferably, the assessment of the impact of radiation is carried out as follows:

[0032] The radiation impact value is calculated by multiplying the ratio of the number of radiation violations by the ratio of the duration of radiation violations.

[0033] If the radiation impact value is greater than or equal to the radiation impact threshold, it will be displayed as a large radiation impact signal.

[0034] Preferably, the process for setting the urban management patrol time period is as follows:

[0035] Arbitrarily select a historical management period as the target analysis period, extract the violation radiation time period corresponding to the violation radiation point within the target analysis period as the target analysis time period, and obtain the time series within the target analysis period where the target analysis time period is located as the benchmark violation radiation time series;

[0036] Extract the illegal radiation points corresponding to the illegal radiation periods in other historical management cycles to obtain multiple comparison and analysis periods, and obtain the time series within the corresponding historical management cycle of the comparison and analysis periods as the comparison radiation time series;

[0037] The baseline violation radiation time series is compared with the comparison violation radiation time series. If the baseline violation radiation time series and the comparison violation radiation time series overlap, the violation radiation period corresponding to the comparison violation radiation time series is marked as the overlapping violation radiation period.

[0038] The ratio of commercial violations to residential violations within each overlapping violation radiation period is obtained, and the coefficient of variation is calculated to obtain the stability coefficient of commercial violations. If the stability coefficient of commercial violations is less than or equal to the threshold of the stability coefficient of commercial violations, the overlapping violation radiation period is marked as the urban management patrol period.

[0039] Preferably, the process for determining the data collection and inspection route is as follows:

[0040] During the urban management patrol period in the historical management cycle, the time point of each commercial management violation point is obtained, and a cycle violation time point sequence is constructed according to the corresponding time sequence.

[0041] Get the time points of each business management violation point within the violation time point sequence of each cycle, sort them, calculate the standard deviation, and output the standard deviation of the violation time point;

[0042] Obtain the distance between each violation point and the violation radiation point, and calculate the ratio with the perimeter of the corresponding urban management sub-zone, outputting the violation distance ratio;

[0043] The violation distance ratio is summed with the standard deviation of the violation time point, and the violation priority inspection value is output.

[0044] Starting from each violation point, a data collection and inspection path is constructed according to the violation priority inspection value corresponding to each violation point, in ascending order.

[0045] Secondly, the present invention provides an urban management problem collection device based on an artificial intelligence large language model, comprising:

[0046] Sub-zone identification module: Utilizes an artificial intelligence large language model to identify the types of urban management sub-zones within the urban management area, distinguishing between commercial urban management sub-zones and residential urban management sub-zones;

[0047] Sequence construction module: Performs spatial proximity analysis on the identified commercial and residential urban management sub-zones, extracts violation analysis points in adjacent commercial and residential areas, and filters and analyzes violation analysis points in multiple historical management cycles to select adjacent violation points and summarize the adjacent violation sequence.

[0048] Radiation assessment module: Performs radiation analysis on neighboring violation points within the adjacent violation sequence to identify the violation radiation points, and conducts radiation impact analysis on the violation radiation points to assess the magnitude of the radiation impact.

[0049] Data collection and planning module: If the radiation impact is large, the time pattern of the illegal radiation points will be explored, the urban management patrol time period will be set, and the data collection and patrol route will be planned.

[0050] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0051] 1. This invention utilizes an artificial intelligence large language model to identify the types of urban management sub-zones within the urban management area, distinguishing between commercial and residential urban management sub-zones. It then performs spatial proximity analysis to extract violation analysis points within adjacent commercial and residential areas. Furthermore, it analyzes data collected on violation issues from these analysis points over multiple historical management cycles, filtering out neighboring violation points and summarizing a neighboring violation sequence. This not only helps the artificial intelligence large language model prioritize high-frequency and stable violation points when collecting urban management information, avoiding wasting effort on points where violations occur only occasionally or are unstable, but also allows for more targeted coverage of key areas and locations, improving collection efficiency and focus.

[0052] 2. This invention performs violation radiation analysis on neighboring violation points within a nearby violation sequence to identify violation radiation points. It then analyzes the radiation impact of these points and assesses the magnitude of the impact. This helps clarify the degree of influence of violation radiation points on commercial and residential urban management sub-zones, aiding in the identification of areas with more serious urban management problems and enabling the development of targeted management strategies for these areas. If the radiation impact is significant, the invention explores the temporal patterns of violation radiation points, setting urban management patrol periods and residential violation patrol periods. This ensures that urban management personnel can appear promptly during peak violation periods, guaranteeing that patrol work always matches the temporal characteristics of violations. The invention also develops data collection and patrol routes to ensure that patrol routes cover key areas and locations where violations occur, avoiding blind patrols and facilitating targeted optimization of the regional layout of urban management. Attached Figure Description

[0053] Figure 1 This is a flowchart illustrating the steps of the urban management problem collection method based on an artificial intelligence large language model according to the present invention.

[0054] Figure 2 This is a flowchart illustrating the judgment process of the urban management problem collection method based on an artificial intelligence large language model, as described in this invention.

[0055] Figure 3 This is a schematic diagram of the urban management problem collection system based on the artificial intelligence large language model of the present invention;

[0056] Figure 4 This is a schematic diagram of the urban management problem collection device based on the artificial intelligence large language model of the present invention; wherein: 1. Collection box body; 2. Collection system module; 3. Collection camera. Detailed Implementation

[0057] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0058] Example 1

[0059] like Figures 1-2 As shown in the embodiment of the present invention, the urban management problem collection method based on an artificial intelligence large language model includes:

[0060] Step 1: Use an artificial intelligence large language model to identify the types of urban management sub-zones within the urban management area, and identify commercial urban management sub-zones and residential urban management sub-zones;

[0061] In some embodiments, the urban management area is divided according to the regional boundary lines on the city map to obtain multiple urban management sub-areas;

[0062] The regulations governing each urban management sub-district are input into an AI-powered large-scale language model. The model then extracts keywords from the regulations and classifies the corresponding urban management sub-district based on these keywords. The process is as follows:

[0063] If the keyword "commercial" is identified and extracted from the content of the district management regulations, it indicates that the corresponding urban management sub-district is a commercial urban management sub-district.

[0064] If the keyword "resident" is identified and extracted from the content of the district management regulations, it indicates that the corresponding urban management sub-district is a resident-urban management sub-district.

[0065] It should be noted that the purpose of identifying commercial and residential urban management sub-zones is to clarify the characteristics and priorities of urban management issues in different areas by dividing the urban management area into these sub-zones. For example, commercial urban management sub-zones typically have high pedestrian and freight traffic and frequent commercial activities, and may face more issues such as street vending, illegal billboards, and difficulties in maintaining environmental sanitation during peak business hours. Residential urban management sub-zones, on the other hand, focus on issues related to residents' daily lives, such as unauthorized construction within residential areas, improper garbage sorting, and damage to public facilities affecting residents' lives. This allows for targeted problem collection, avoiding indiscriminate collection and improving the efficiency and accuracy of problem collection.

[0066] Secondly, identifying sub-area types helps to rationally allocate urban management resources based on the characteristics of different areas. For example, commercial areas have higher requirements for the creation of a city image and business atmosphere, requiring more resources to be invested in environmental improvement and order maintenance; residential areas, on the other hand, are more concerned with living comfort and safety, and resource allocation will focus on infrastructure construction and security. Therefore, developing appropriate management strategies helps to improve the precision of urban management.

[0067] Step 2: Perform spatial proximity analysis on the identified commercial and residential urban management sub-zones, extract violation analysis points in adjacent commercial and residential areas, analyze the data collected on violation issues of violation analysis points in multiple historical management cycles, screen out adjacent violation points, and summarize the adjacent violation sequence.

[0068] In some embodiments, within an urban management area, adjacent commercial urban management sub-areas and residential urban management sub-areas are merged to obtain adjacent commercial-residential combined areas;

[0069] For example, within the urban management area, any commercial urban management sub-zone can be selected as the base sub-zone for merging;

[0070] Extract the urban management sub-areas adjacent to the base sub-area. If the urban management sub-area is a residential urban management sub-area, then merge them to obtain the adjacent commercial and residential first-level combined area.

[0071] If the urban management sub-district is a commercial urban management sub-district, then no merger operation will be performed;

[0072] Within the adjacent commercial and residential primary integration zone, arbitrarily select a commercial urban management sub-zone and extract the urban management sub-zone adjacent to the merged basic sub-zone. If the urban management sub-zone is a residential urban management sub-zone, then merge them to obtain the adjacent commercial and residential secondary integration zone.

[0073] Traverse all urban management sub-districts within the urban management area, and continuously iterate through the merging process according to the merging method of adjacent commercial and residential primary merging areas and adjacent commercial and residential secondary merging areas, until the urban management sub-districts adjacent to the merging base sub-district are of the same type as the merging base sub-district, then do not perform the merging operation, and obtain the adjacent commercial and residential merging area.

[0074] For example, if the basic sub-area of ​​A is a commercial urban management sub-area, then extract the urban management sub-area B that is adjacent to the basic sub-area of ​​A. If the urban management sub-area B is a residential urban management sub-area, then merge them to obtain the adjacent commercial and residential first-level combined area.

[0075] If City B is a commercial city management sub-district, then no merger operation will be performed;

[0076] Within the adjacent commercial and residential primary integration zone, arbitrarily select a C commercial urban management sub-zone and extract the D urban management sub-zone adjacent to the merged basic sub-zone. If the D urban management sub-zone is a residential urban management sub-zone, then merge them to obtain the adjacent commercial and residential secondary integration zone.

[0077] If City Management Sub-district D is a Commercial City Management Sub-district, then no merger operation will be performed;

[0078] Within the adjacent commercial and residential mixed-use area, arbitrarily extract the boundary edge points between the commercial urban management sub-area and the residential urban management sub-area as violation analysis points.

[0079] It should be noted that the boundary point refers to the specific physical location where two management sub-zones with different functional attributes, namely a commercial area and a residential area, are spatially adjacent and intersecting. Because these points are the junctions of the functional boundaries between the two types of areas, and due to differences in the attributes of commercial activities and residential life, such as pedestrian density, activity types, and different management emphases, they are prone to becoming high-incidence areas for urban management problems and are also key monitoring targets for urban management problem collection. For example, the boundary point is the junction between the end of the pedestrian walkway of a commercial street and the wall of a residential area. Common management problems collected at this junction include: merchants piling up garbage cans and debris at this junction; and residents damaging wall facilities for convenience. Another example is the intersection of the exit of a shopping mall's unloading area and the back alley of a residential building: shopping mall trucks obstructing residents' passage when unloading goods; and residents dumping household waste at this junction.

[0080] The historical management cycle is divided into several equal historical management periods, each of which has an equal duration.

[0081] Furthermore, the duration of each historical management period within each historical management cycle is equal, and the number of historical management periods is also equal.

[0082] For example, the number of violations by a violation analysis point within each historical management period is obtained, and the ratio is calculated with the duration of the historical management period to obtain the violation frequency for that period.

[0083] The average frequency of violations at each historical management period is calculated by averaging the frequency of violations at each violation analysis point, and the average frequency of violations during each period is output.

[0084] The standard deviation of the violation frequency of each violation analysis point within each historical management period is calculated, and the standard deviation of the violation frequency within each period is output.

[0085] The ratio of the average violation rate to the standard deviation of the violation rate during a given time period is calculated to output the stable violation value.

[0086] Understandably, the meaning of the violation stability value is that it reflects the overall stability of violations at a violation analysis point over multiple historical management cycles. On the one hand, this helps AI large language models to focus on these high-frequency, stable violation points when collecting urban management information, avoiding wasting effort on points where violations occur only occasionally or are unstable, thereby improving the accuracy of problem collection. On the other hand, urban management departments can rationally allocate urban management resources based on this information. For neighboring violation points with lower violation stability values, it indicates that there are relatively stable violation problems at these points over a long period of time, requiring more manpower and resources for key supervision and rectification. For points with lower violation stability values, resource investment can be appropriately reduced.

[0087] If the violation stability value is greater than the violation stability threshold, it indicates that the violation frequency of the analyzed violation points is relatively unstable over multiple historical periods.

[0088] If the violation stability value is less than or equal to the violation stability threshold, it indicates that the violation frequency of the analyzed violation point is relatively stable over multiple historical periods, and the analyzed violation point is taken as the neighboring violation point.

[0089] Each neighboring violation point is sorted in ascending order according to its corresponding violation stability value to construct a neighboring violation sequence.

[0090] The specific solution in this embodiment is as follows: An artificial intelligence large language model is used to identify the types of urban management sub-zones within the urban management area, distinguishing between commercial and residential urban management sub-zones. Spatial proximity analysis is then performed to extract violation analysis points within adjacent commercial and residential areas. Data collected on violation issues at these violation analysis points over multiple historical management cycles is analyzed to filter out neighboring violation points and summarize neighboring violation sequences. This not only helps the artificial intelligence large language model prioritize these high-frequency and stable violation points when collecting urban management issues, but also avoids wasting resources on unstable violation analysis points.

[0091] Example 2

[0092] like Figures 1-2As shown, based on Example 1, the urban management problem collection method based on artificial intelligence large language model described in this embodiment of the invention includes:

[0093] Step 3: Perform violation radiation analysis on neighboring violation points within the adjacent violation sequence to identify violation radiation points, and conduct radiation impact analysis on violation radiation points to assess the magnitude of radiation impact.

[0094] In some embodiments, an adjacent violation point within an adjacent violation sequence is arbitrarily selected as a suspected radiation point, and the historical management period during which the suspected radiation point violated regulations is taken as the suspected radiation period.

[0095] During the suspected radiation period, violations in the commercial urban management sub-zone and violations in the residential urban management sub-zone were extracted separately as commercial management violation points and residential management violation points, respectively.

[0096] It should be noted that "commercial management violation points" refer to locations within the commercial urban management sub-zone that violate urban management regulations.

[0097] "Residential management violation points" refers to locations within residential urban management sub-areas that violate urban management regulations.

[0098] For example, during the suspected radiation period, the distance between the commercial management violation point / residential management violation point and the suspected radiation point is obtained, and the ratio is calculated with the perimeter of the commercial urban management sub-zone / residential urban management sub-zone as the suspected radiation distance ratio;

[0099] The time interval between the occurrence of violations at commercial / residential violation points and the occurrence of violations at suspected radiation points is obtained as a proportion of the suspected radiation period, thus yielding the suspected radiation time interval ratio.

[0100] The suspected radiation time interval ratio and suspected radiation distance ratio are summed, and the reciprocal is taken to output the suspected radiation value.

[0101] It is understandable that the suspected radiation value means that it is obtained by superimposing standardized calculations of spatial proximity and temporal correlation. It is used to measure the potential impact of a certain violation point (suspected radiation point) on the surrounding area, namely the commercial urban management sub-area and residential urban management sub-area in the adjacent commercial and residential urban management area, in terms of spatial and temporal dimensions.

[0102] If the suspected radiation value is greater than or equal to the suspected radiation threshold, it indicates that the suspected radiation point and the commercial management violation point are similar in terms of occurrence time and spatial dimension, and the probability of mutual influence is relatively high. The analyzed suspected radiation point is marked as a violation radiation point.

[0103] If the suspected radiation distance is less than the suspected radiation distance threshold, it means that the suspected radiation point and the commercial management violation point are far apart in terms of occurrence time and spatial dimension, and the probability of mutual influence is small. The analyzed suspected radiation point is marked as a non-violation radiation point.

[0104] Extract the historical management periods in which violations occurred at the violation radiation points as the violation radiation periods. Count the number of commercial management violation points within the violation radiation period and calculate the ratio of the number of commercial management violation points in the commercial city management sub-zone. Output the ratio of the number of commercial violation points in the radiation period.

[0105] Similarly, the number of illegal residential areas within the period of violation radiation is counted, and the ratio is calculated to the total number of illegal residential areas within the urban management sub-district. The output is the ratio of the number of illegal residential areas radiating outwards.

[0106] The ratio of radiation violations is obtained by summing the ratio of commercial violations to residential violations.

[0107] Obtain the duration of each violation at each commercial management violation point within the violation radiation period, perform average calculation, and output the ratio of commercial violation duration;

[0108] Obtain the duration of violations at each community-managed violation point within the period of violation radiation, perform mean calculation, and output the ratio of community violation duration;

[0109] The ratio of duration of commercial violations to duration of duration of residential violations is summed to obtain the ratio of duration of radiation violations.

[0110] The radiation impact value is calculated by multiplying the ratio of the number of radiation violations by the ratio of the duration of radiation violations.

[0111] It is understandable that the radiation impact value represents a core quantitative indicator for measuring the comprehensive impact of a violation's radiation point on the surrounding area. It is a multi-dimensional assessment framework for the radiation effect of violations, formed by integrating the scale of the violation (quantitative dimension) with the persistence of the violation (temporal dimension). Specifically:

[0112] In terms of spatial dimension, it is necessary to clarify the degree of impact of the violation radiation points on the commercial urban management sub-areas and the residential urban management sub-areas. Since the difference in radiation impact values ​​in different areas can reflect the spread and impact range of violations in different spaces, it is helpful to screen out areas with more serious urban management problems and formulate targeted management strategies for these areas.

[0113] In terms of time dimension, the calculation of radiation impact value involves factors such as the time interval and duration of violations, which can reveal the distribution pattern of violations over time. Therefore, it can not only predict the timing of future violations based on the changes in radiation impact value in different time periods, but also improve the efficiency of collecting information on urban management issues by reasonably arranging patrol times.

[0114] If the radiation impact value is greater than or equal to the radiation impact threshold, it indicates that the analyzed illegal radiation point has a significant radiation impact on the commercial urban management sub-area and the residential urban management sub-area, which is a signal of large radiation impact.

[0115] If the radiation impact value is less than the radiation impact threshold, it indicates that the radiation impact of the analyzed illegal radiation point on the commercial urban management sub-area and the residential urban management sub-area is small, and it is displayed as a small radiation impact signal. In this case, the patrol will be carried out according to the original patrol time period.

[0116] Step 4: If the impact of the radiation is significant, explore the temporal patterns of the violation radiation points, set urban management patrol periods, formulate patrol routes for data collection, and complete the data collection work on urban management issues.

[0117] In some embodiments, any historical management period is arbitrarily selected as the target analysis period;

[0118] Extract the illegal radiation period corresponding to the illegal radiation point within the target analysis period, and use it as the target analysis period. Also, obtain the time series within the target analysis period where the target analysis period is located, and use it as the baseline illegal radiation time series.

[0119] Extract the illegal radiation points corresponding to the illegal radiation periods in other historical management cycles to obtain multiple comparison and analysis periods, and obtain the time series within the corresponding historical management cycle of the comparison and analysis periods as the comparison radiation time series;

[0120] The baseline violation radiation time series is compared with the comparison violation radiation time series. If the baseline violation radiation time series and the comparison violation radiation time series overlap, the violation radiation period corresponding to the comparison violation radiation time series is marked as the overlapping violation radiation period.

[0121] If there is no overlap between the baseline violation timing and the comparison violation timing, no operation will be performed;

[0122] Obtain the ratio of commercial violations to residential violations during each overlapping period of illegal radiation exposure.

[0123] The coefficient of variation is calculated for the ratio of radiation quotient violations during all overlapping violation radiation periods, and the stability coefficient of radiation quotient violations is output.

[0124] If the radiation violation stability coefficient is greater than the radiation violation stability coefficient threshold, it means that the number of commercial violation points affected by radiation is relatively unstable during the overlapping violation radiation period, and no action is taken.

[0125] If the radiation commercial violation stability coefficient is less than or equal to the radiation commercial violation stability coefficient threshold, it indicates that the number of commercial violation points affected by the radiation is relatively stable during the overlapping violation radiation period. The overlapping violation radiation period is marked as the urban management patrol period.

[0126] For example, during the urban management patrol period (urban management patrol period for commercial violations or urban management patrol period for residential violations) in the historical management cycle, the time point when each commercial violation point (commercial violation point or residential violation point) occurs is obtained, and the violation time point is constructed by comparing the corresponding time sequence before and after.

[0127] It should be noted that the periodic violation time sequence (either the periodic commercial violation time sequence or the periodic residential violation time sequence) only corresponds to the urban management patrol period (either the commercial violation urban management patrol period or the residential violation urban management patrol period) within a historical management cycle.

[0128] Get the time point ranking of each commercial violation point in each cycle violation time point sequence (cycle commercial violation time point sequence or cycle civil violation time point sequence), calculate the standard deviation, and output the standard deviation of the violation time point (standard deviation of commercial violation time point or standard deviation of civil violation time point).

[0129] Obtain the distance between each violation point (commercial or residential) and the violation radiation point, and calculate the ratio with the perimeter of the corresponding urban management sub-zone (commercial or residential). Output the violation distance ratio (commercial or residential).

[0130] The violation distance ratio (either commercial or residential) is summed with the violation time standard deviation (either commercial or residential) to output the violation priority inspection value (either commercial or residential).

[0131] Starting from each violation point, a data collection and inspection path is constructed according to the violation priority inspection value (standard deviation of commercial violation time point or standard deviation of residential violation time point) corresponding to each violation point (commercial violation point or residential violation point), in ascending order.

[0132] The specific scheme of this embodiment is as follows: Violation radiation analysis is performed on adjacent violation points within the adjacent violation sequence to determine the violation radiation points. Radiation impact analysis is then conducted on these violation radiation points to assess the magnitude of the radiation impact. This helps to clarify the degree of impact of the violation radiation points on commercial and residential urban management sub-areas, and helps to screen areas with more serious urban management problems. Targeted management strategies can then be developed for these areas. If the radiation impact is large, the temporal patterns of the violation radiation points are explored, and urban management patrol periods and residential violation urban management patrol periods are set. This ensures that urban management personnel can appear promptly during periods of high violation incidence, ensuring that patrol work always matches the temporal characteristics of violations. A data collection and patrol route is developed to ensure that the patrol route covers key areas and important locations where violations occur, avoiding blind patrols and helping to optimize the regional layout of urban management in a targeted manner.

[0133] Example 3

[0134] like Figure 3 As shown in the embodiment of the present invention, the urban management problem collection system based on an artificial intelligence large language model includes:

[0135] Sub-zone identification module: Utilizes an artificial intelligence large language model to identify the types of urban management sub-zones within the urban management area, distinguishing between commercial urban management sub-zones and residential urban management sub-zones;

[0136] Sequence construction module: Performs spatial proximity analysis on the identified commercial and residential urban management sub-zones, extracts violation analysis points in adjacent commercial and residential areas, and filters and analyzes violation analysis points in multiple historical management cycles to select adjacent violation points and summarize the adjacent violation sequence.

[0137] Radiation assessment module: Performs radiation analysis on neighboring violation points within the adjacent violation sequence to identify the violation radiation points, and conducts radiation impact analysis on the violation radiation points to assess the magnitude of the radiation impact.

[0138] Data collection and planning module: If the radiation impact is large, the time pattern of the illegal radiation points will be explored, the urban management patrol time period will be set, and the data collection and patrol route will be planned.

[0139] Example 4

[0140] like Figure 4 As shown in the figure, the urban management problem collection device based on artificial intelligence large language model provided in this embodiment of the invention includes a collection box body 1, a collection system module 2, and a collection camera 3.

[0141] like Figure 4 As shown, the urban management problem collection device based on the artificial intelligence large language model can be configured as the trunk of electric bicycles, motorcycles and other vehicle types, which facilitates the collection of data around the patrol route during the patrol process.

[0142] Specifically, the main body 1 of the acquisition box is equipped with an acquisition system module 2, which includes: a sub-region identification module, a sequence construction module, a radiation assessment module, and an acquisition planning module.

[0143] Four acquisition cameras 3 are evenly embedded on the outside of the acquisition box body 1, and each acquisition camera 3 is electrically connected to the acquisition system module 2.

[0144] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art can make various changes and modifications to the present invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for collecting urban management issues based on an artificial intelligence large language model, characterized by: include: Using an artificial intelligence large language model, the types of urban management sub-zones within the urban management area are identified, distinguishing between commercial urban management sub-zones and residential urban management sub-zones; Spatial proximity analysis was performed on the identified commercial and residential urban management sub-zones to extract violation analysis points in adjacent commercial and residential areas. Violation analysis points from multiple historical management cycles were screened and analyzed to identify adjacent violation points and summarize the adjacent violation sequence. Violation radiation analysis is performed on neighboring violation points within the adjacent violation sequence to identify violation radiation points, and radiation impact analysis is performed on violation radiation points to assess whether the impact of radiation is significant. The process for identifying illegal radiation points is as follows: Arbitrarily select a neighboring violation point within the adjacent violation sequence as a suspected radiation point, and take the historical management period in which the suspected radiation point violated the violation as the suspected radiation period. Within the suspected radiation period, extract the violation points in the commercial urban management sub-zone and the violation points in the residential urban management sub-zone as commercial management violation points and residential management violation points, respectively. During the suspected radiation period, the distance between the commercial management violation point and the suspected radiation point is obtained, and the ratio is calculated with the perimeter of the commercial city management sub-zone as the suspected radiation distance ratio. The time interval between the occurrence of violations at commercial / residential violation points and the occurrence of violations at suspected radiation points is obtained as a proportion of the suspected radiation period, thus yielding the suspected radiation time interval ratio. The suspected radiation interval ratio and suspected radiation distance ratio are summed, and the reciprocal is taken to obtain the suspected radiation value. If the suspected radiation value is greater than or equal to the suspected radiation threshold, the analyzed suspected radiation point is marked as an illegal radiation point. If the radiation impact is significant, then the time patterns of the illegal radiation points will be explored, urban management patrol periods will be set, and patrol routes will be developed. If the radiation impact is minor, patrols will be conducted according to the original patrol periods.

2. The urban management problem collection method based on an artificial intelligence large language model according to claim 1, characterized in that: The process for identifying commercial urban management zones and residential urban management zones is as follows: The regulations governing each urban management sub-district are input into an AI-powered large language model. The AI-powered large language model is then used to identify and extract keywords from the content of the regulations. Based on these keywords, the corresponding urban management sub-districts are categorized into commercial urban management sub-districts and residential urban management sub-districts.

3. The urban management problem collection method based on artificial intelligence large language model according to claim 1, characterized in that: The method for extracting violation analysis points within adjacent commercial and residential areas is as follows: Within the urban management area, arbitrarily select a commercial urban management sub-zone as the base sub-zone for merging; Extract the urban management sub-areas adjacent to the base sub-area. If the urban management sub-area is a residential urban management sub-area, then merge them to obtain the adjacent commercial and residential first-level combined area. Within the adjacent commercial and residential primary integration zone, arbitrarily select a commercial urban management sub-zone and extract the urban management sub-zone adjacent to the merged basic sub-zone. If the urban management sub-zone is a residential urban management sub-zone, then merge them to obtain the adjacent commercial and residential secondary integration zone. Traverse all urban management sub-areas within the urban management area, and continuously iterate and merge according to the merging method of adjacent commercial and residential first-level combined area and adjacent commercial and residential second-level combined area until the urban management sub-area adjacent to the merged base sub-area is of the same type as the merged base sub-area, then do not perform the merging operation, and obtain the adjacent commercial and residential combined area. Within the adjacent commercial and residential urban management area, arbitrarily extract the boundary edge points between the commercial urban management sub-area and the residential urban management sub-area as violation analysis points.

4. The urban management problem collection method based on artificial intelligence large language model according to claim 1, characterized in that: The process of filtering out neighboring violation points and summarizing the neighboring violation sequence is as follows: The historical management cycle is divided into several equal historical management periods. The number of violations by each violation analysis point in each historical management period is obtained, and the ratio of the number of violations to the duration of the historical management period is calculated to obtain the violation frequency of each period. The average frequency of violations at each historical management period is calculated by averaging the frequency of violations at each violation analysis point, and the average frequency of violations during each period is output. The standard deviation of the violation frequency of each violation analysis point within each historical management period is calculated, and the standard deviation of the violation frequency within each period is output. The ratio of the average violation frequency to the standard deviation of the violation frequency during a time period is calculated to obtain the violation stability value. If the violation stability value is less than or equal to the violation stability threshold, the analyzed violation points are taken as neighboring violation points, and each neighboring violation point is sorted in ascending order according to its corresponding violation stability value to construct a neighboring violation sequence.

5. The method for collecting urban management problems based on an artificial intelligence large language model according to claim 1, characterized in that: The process of analyzing the radiation impact of the illegal radiation points is as follows: Extract the historical management periods in which violations occurred at the violation radiation points as violation radiation periods. Calculate the ratio of the number of commercial violation points to the total number of commercial violation points in the commercial urban management sub-district, and the ratio of the number of private violation points to the total number of private violation points in the residential urban management sub-district, to obtain the ratio of commercial violation radiation to private violation radiation. Then sum them up to obtain the total number of violations radiation. The duration of violations at each commercial and residential violation point within the period of violation radiation is obtained. The average duration of each violation point is calculated to obtain the ratio of duration of commercial violations and residential violations. The sum of these values ​​is then used to obtain the ratio of duration of violations radiating outwards.

6. The method for collecting urban management problems based on an artificial intelligence large language model according to claim 5, characterized in that: The process for assessing the impact of radiation is as follows: The radiation impact value is calculated by multiplying the ratio of the number of radiation violations by the ratio of the duration of radiation violations. If the radiation impact value is greater than or equal to the radiation impact threshold, it will be displayed as a large radiation impact signal.

7. The method for collecting urban management problems based on an artificial intelligence large language model according to claim 1, characterized in that: The process for setting urban management patrol times is as follows: Arbitrarily select a historical management period as the target analysis period, extract the violation radiation time period corresponding to the violation radiation point within the target analysis period as the target analysis time period, and obtain the time series within the target analysis period where the target analysis time period is located as the benchmark violation radiation time series; Extract the illegal radiation points corresponding to the illegal radiation periods in other historical management cycles to obtain multiple comparison and analysis periods, and obtain the time series within the corresponding historical management cycle of the comparison and analysis periods as the comparison radiation time series; The baseline violation radiation time series is compared with the comparison violation radiation time series. If the baseline violation radiation time series and the comparison violation radiation time series overlap, the violation radiation period corresponding to the comparison violation radiation time series is marked as the overlapping violation radiation period. The ratio of commercial violations to residential violations within each overlapping violation radiation period is obtained, and the coefficient of variation is calculated to obtain the stability coefficient of commercial violations. If the stability coefficient of commercial violations is less than or equal to the threshold of the stability coefficient of commercial violations, the overlapping violation radiation period is marked as the urban management patrol period.

8. The method for collecting urban management problems based on an artificial intelligence large language model according to claim 1, characterized in that: The process for determining the data collection and inspection route is as follows: During the urban management patrol period in the historical management cycle, the time point of each commercial management violation point is obtained, and a cycle violation time point sequence is constructed according to the corresponding time sequence. Get the time points of each business management violation point within the violation time point sequence of each cycle, sort them, calculate the standard deviation, and output the standard deviation of the violation time point; Obtain the distance between each violation point and the violation radiation point, and calculate the ratio with the perimeter of the corresponding urban management sub-zone, outputting the violation distance ratio; The violation distance ratio is summed with the standard deviation of the violation time point, and the violation priority inspection value is output. Starting from each violation point, a data collection and inspection path is constructed according to the violation priority inspection value corresponding to each violation point, in ascending order.

9. A device for collecting urban management problems based on an artificial intelligence large language model, characterized in that: For implementing the urban management problem collection method based on an artificial intelligence large language model as described in any one of claims 1-8, the collection device comprises: Sub-zone identification module: Utilizes an artificial intelligence large language model to identify the types of urban management sub-zones within the urban management area, distinguishing between commercial urban management sub-zones and residential urban management sub-zones; Sequence construction module: Performs spatial proximity analysis on the identified commercial and residential urban management sub-zones, extracts violation analysis points in adjacent commercial and residential areas, and filters and analyzes violation analysis points in multiple historical management cycles to select adjacent violation points and summarize the adjacent violation sequence. Radiation assessment module: Performs radiation analysis on neighboring violation points within the adjacent violation sequence to identify the violation radiation points, and conducts radiation impact analysis on the violation radiation points to assess the magnitude of the radiation impact. Data collection and planning module: If the radiation impact is large, the time pattern of the violation radiation points will be explored, the urban management patrol time period will be set, and the data collection and patrol route will be planned. If the radiation impact is small, the patrol will be carried out according to the original patrol time period.

Images