Content data prediction analysis method based on digital twinning

Abstract

The invention discloses a digital twinborn-based content data prediction analysis method, and relates to the technical field of data prediction, and the method comprises the following steps: collecting multi-source heterogeneous data related to a law enforcement case, and carrying out the fusion processing of the multi-source heterogeneous data to construct a digital twinborn model of the law enforcement case; key elements in the law enforcement case are extracted based on a digital twinborn model, an association influence coefficient set between the key elements is determined, and a normal change range value of the association influence coefficient between the key elements is determined; according to the association influence coefficient set, determining the association influence degree of the current law enforcement case and the historical similar case on the key elements, and then obtaining a current association influence degree set, which has the effect that a second risk prediction value is predicted according to the factors; and a comprehensive prediction analysis result is obtained by combining a first risk prediction value obtained based on case key element analysis.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data prediction, more particularly, it relates to a content data prediction analysis method based on digital twinning. BACKGROUND

[0002] Under the background of the continuous promotion of social governance and rule of law construction, law enforcement work is facing increasingly complex and diverse challenges. The traditional law enforcement mode and case analysis method have been difficult to meet the actual demand, and innovative technical solutions are needed to improve the law enforcement efficiency and the scientific nature of decision-making. The law enforcement case content data prediction analysis method based on digital twinning has emerged as the times require.

[0003] The traditional method analyzes the key elements such as case type, spatiotemporal characteristics, behavior pattern of involved personnel, and correlation of evidence chain in isolation, and fails to fully explore the internal relationship and mutual influence among the elements. For example, when analyzing a case, the influence of case type on the behavior pattern of involved personnel and the effect of spatiotemporal characteristics on the integrity of evidence chain are rarely considered, making it difficult to accurately grasp the development trend and potential risks of the case. SUMMARY

[0004] In view of the deficiencies in the prior art, the purpose of the present application is to provide a content data prediction analysis method based on digital twinning.

[0005] To achieve the above-mentioned purpose, the present application provides the following technical solutions: A content data prediction analysis method based on digital twinning, comprising the following steps: Collecting multi-source heterogeneous data related to law enforcement cases, and fusing and processing the multi-source heterogeneous data to construct a digital twinning model of law enforcement cases; Extracting key elements in law enforcement cases based on the digital twinning model, and determining the correlation influence coefficient set between the key elements and the normal variation range value of the correlation influence coefficient between the key elements; According to the correlation influence coefficient set, determine the correlation influence degree of the key elements between the current law enforcement case and the historical similar case, and obtain the current correlation influence degree set. After processing the current correlation influence degree set, compare it with the preset normal influence range value to obtain a comparison result; Combine the deep analysis signal in the comparison result with the dynamic evolution law of the case elements in the digital twinning model of the law enforcement case to predict the change trend of the key elements of the current law enforcement case and generate a first prediction data set; According to the first prediction data set, analyze the development trend of the current law enforcement case and obtain a first risk prediction value; The statistical law enforcement case corresponding associated law enforcement resource distribution is determined, and a comprehensive prediction analysis result is obtained according to the law enforcement resource distribution, social index data and the first risk prediction value.

[0006] Preferably, the multi-source heterogeneous data includes case basic information, on-site investigation data, party information and historical similar case data.

[0007] Preferably, the key elements include case type, space-time characteristics, behavior mode of involved personnel and evidence chain association relationship.

[0008] Preferably, the association influence coefficient set between the key elements is determined, and the normal variation range value of the association influence coefficient between the key elements is determined, specifically including the following steps: The mutual influence relationship between the key elements is determined, and the association influence coefficient set between the key elements is established; The association influence coefficient between the key elements in the historical normal case and the historical abnormal case is obtained respectively, and the historical normal association influence coefficient set and the historical abnormal association influence coefficient set are constructed; The normal variation range value of the association influence coefficient between the key elements is determined according to the historical normal association influence coefficient set and the historical abnormal association influence coefficient set.

[0009] Preferably, the current association influence degree set is processed and compared with the preset normal influence range value to obtain a comparison result, specifically including the following steps: If there is data greater than the preset normal influence range value in the current association influence degree set, the case key elements corresponding to the data greater than the preset normal influence range value in the current association influence degree set are determined as the risk warning signal of the comparison result; If each data in the current association influence degree set is less than the preset normal influence range value, the difference between each data in the current association influence degree set and the preset normal influence range value is calculated to obtain a difference set; The preset difference threshold value, if there is data greater than or equal to the difference threshold value in the difference set, then the key elements of the current law enforcement case are further analyzed and the depth analysis signal of the comparison result is output.

[0010] Preferably, the depth analysis signal is combined with the dynamic evolution law of the case elements in the digital twin model of the law enforcement case to predict the change trend of the key elements of the current law enforcement case to generate a first prediction data set; The dynamic evolution characteristics of the key elements of the current law enforcement case are extracted according to the depth analysis signal, and a dynamic evolution model of the key elements is established by combining the dynamic evolution law of the case elements in the digital twin model of the law enforcement case; The first prediction data set is generated after predicting the change trend of key elements of the current law enforcement case in a preset time period by using a dynamic evolution model, wherein the first prediction data set comprises a case type evolution trend, a spatio-temporal characteristic change trend, an involved person behavior mode change trend, and an evidence chain correlation change trend.

[0011] Preferably, the development trend of the current law enforcement case is analyzed according to the first prediction data set, and a first risk prediction value is obtained, specifically including the following steps: An environmental influence degree set is obtained after judging the influence degree of environmental factors on the key elements of the case based on the first prediction data set and the digital twin model. A second prediction data set of the current law enforcement case affected by factors is predicted according to the environmental influence degree set and the first prediction data set. The development trend of the current law enforcement case is determined based on the second prediction data set, and the first risk prediction value of the current law enforcement case is determined according to the development trend of the current law enforcement case.

[0012] Preferably, the law enforcement resource distribution situation comprises the number of law enforcement personnel, the law enforcement equipment allocation situation, and the law enforcement agency distribution density.

[0013] Preferably, the social index data comprises population density, public security situation, and public opinion index.

[0014] Preferably, a comprehensive prediction analysis result of the current law enforcement case is obtained by comprehensively predicting and analyzing the current law enforcement case according to the law enforcement resource distribution situation, the social index data, and the first risk prediction value, specifically including the following steps: A second risk prediction value of the current law enforcement case is predicted according to the law enforcement resource distribution situation and the social index data. The comprehensive prediction analysis result of the current law enforcement case is determined according to the first risk prediction value and the second risk prediction value.

[0015] Compared with the prior art, the present application has the following beneficial effects: The present application constructs a digital twin model by collecting multi-source heterogeneous data, extracts key elements, and analyzes correlation influence. When the current correlation influence degree set is compared with a preset normal range, a risk early warning signal can be captured in time, such as when the correlation influence degree of key elements exceeds the normal range, the potential risk of the case can be quickly identified, the law enforcement department can intervene in advance, and the case can be prevented from deteriorating. By deeply analyzing the triggering mechanism of the signal, subtle abnormalities can be excavated, key elements can be deeply analyzed, and the deterioration or transformation trend of the case can be predicted in advance, such as predicting the evolution of the case type from public security to criminal, so as to gain valuable time for risk prevention and control.

[0016] Comprehensively consider the distribution of law enforcement resources (the number of law enforcement personnel, equipment allocation, and the density of agency distribution) and social index data (population density, public security situation, and public opinion index). According to these factors, the second risk prediction value is predicted, and the first risk prediction value obtained based on the analysis of the key elements of the case is combined to obtain a comprehensive prediction analysis result. This helps the law enforcement department to clearly grasp the overall picture of the risk of the case, and to reasonably allocate resources. In areas where the risk of the case is high and the law enforcement resources are tight, targeted personnel and equipment investment can be increased, the layout of the agency can be optimized, resource waste can be avoided, the overall law enforcement efficiency can be improved, and the efficient development of law enforcement work can be ensured. BRIEF DESCRIPTION OF DRAWINGS

[0017] Fig. 1 A step schematic diagram of a content data prediction analysis method based on digital twinning is proposed for the present application; Fig. 2 A step schematic diagram of obtaining a normal variation range value in a content data prediction analysis method based on digital twinning is proposed for the present application; Fig. 3 A structural schematic diagram of an electronic device provided by an embodiment of the present application.

[0018] 610, processor; 620, communication interface; 630, memory; 640, communication bus. DETAILED DESCRIPTION

[0019] Referring to Figs. 1 to 3 as shown.

[0020] Embodiments further illustrate a content data prediction analysis method based on digital twinning proposed by the present application.

[0021] A content data prediction analysis method based on digital twinning, the method comprising the following steps: Collecting multi-source heterogeneous data related to law enforcement cases, and performing fusion processing on the multi-source heterogeneous data to construct a digital twinning model of the law enforcement cases; Extracting key elements in the law enforcement cases based on the digital twinning model, determining a set of correlation influence coefficients between the key elements, and determining a normal variation range value of the correlation influence coefficients between the key elements; After determining the correlation influence degree set of the key elements between the current law enforcement case and the historical similar cases according to the set of correlation influence coefficients, processing the current correlation influence degree set, and comparing it with the preset normal influence range value to obtain a comparison result; Combining the deep analysis signal in the comparison result with the dynamic evolution law of the case elements in the digital twinning model of the law enforcement case to predict the change trend of the key elements of the current law enforcement case and generate a first prediction data set; According to the first prediction data set, the development trend of the current law enforcement case is analyzed, and a first risk prediction value is obtained; The distribution of the corresponding associated law enforcement resources of the law enforcement case is counted, and a comprehensive prediction analysis result is obtained by comprehensively predicting and analyzing the current law enforcement case according to the law enforcement resource distribution, the social index data and the first risk prediction value.

[0022] The law enforcement work of the present application involves various types of data, and the multi-source heterogeneous data covers case basic information (such as case nature, occurrence time and place, etc.), on-site investigation data (on-site traces, physical evidence and related information), party information (party identity, past behavior records, etc.), historical similar case data (past similar case processing flow, results, etc.). These data of different sources and different structures are collected. Then the data are fused and processed, mainly including data cleaning (removing errors and duplicate data), data standardization (unifying data format) and other operations. Based on the fused data, a digital twin model of the law enforcement case is constructed. The model is a mapping of the real situation of the law enforcement case in the virtual space, which can dynamically reflect various types of information related to the case.

[0023] On the basis of the constructed digital twin model, the key elements in the case are extracted, including case type (such as public security case, criminal case, etc.), spatio-temporal characteristics (time, place and space environment of the case, etc.), behavior pattern of the involved person (action track and behavior habit of the party, etc.), and evidence chain correlation (mutual confirmation relationship between evidences, etc.). The correlation influence coefficient set between the key elements is determined, i.e. the degree of mutual influence between the elements is quantitatively analyzed, such as the case type may affect the behavior pattern of the involved person, and the quantitative value of this influence is calculated. At the same time, in order to determine whether the correlation influence coefficient is within a reasonable range, the correlation influence coefficients between the key elements in the historical normal case and the historical abnormal case are obtained, and the historical normal correlation influence coefficient set and the historical abnormal correlation influence coefficient set are constructed. The normal variation range value of the correlation influence coefficient between the key elements is determined by analyzing the two sets.

[0024] According to the previously determined set of correlation influence coefficients, the current correlation influence degree set is obtained by comparing the correlation influence degrees of the key elements of the current law enforcement case and the historical similar cases. After processing the current correlation influence degree set, it is compared with the preset normal influence range value: if there is data in the current correlation influence degree set that is greater than the preset normal influence range value, it means that the corresponding case key element is abnormal, and it is determined as the risk warning signal of the comparison result; if each data in the current correlation influence degree set is less than the preset normal influence range value, the difference between each data and the preset normal influence range value is calculated to obtain a difference set, and when there is data in the difference set that is greater than or equal to the preset difference threshold, it indicates that it is necessary to continue to analyze the key elements of the current law enforcement case in depth, and at this time the depth analysis signal of the comparison result is output.

[0025] When the depth analysis signal in the comparison result is obtained, the dynamic evolution characteristics of the key elements of the current law enforcement case are extracted in combination with the dynamic evolution law of the case elements in the law enforcement case digital twin model (such as the change trend of the case type over time, the change of the behavior mode of the involved personnel in different stages, etc.). Based on these characteristics, a dynamic evolution model of the key elements is established, and the change trend of the key elements of the current law enforcement case in a preset time period is predicted by using the model to generate a first prediction data set, which covers the case type evolution trend, the spatio-temporal characteristic change trend, the behavior mode change trend of the involved personnel, and the evidence chain correlation change trend, etc.

[0026] Based on the first prediction data set and the digital twin model, the influence degree of environmental factors (such as population density, public order, public opinion, etc. in the case area) on the key elements of the case is obtained to obtain an environmental influence degree set. According to the environmental influence degree set and the first prediction data set, the development trend of the current law enforcement case affected by these factors is further predicted to obtain a second prediction data set. Based on the second prediction data set, the development trend of the current law enforcement case is determined, such as whether it is developing towards a more complex and dangerous direction, or whether it has a trend of mitigation and solution, and the first risk prediction value of the current law enforcement case is determined according to the development trend, which quantifies the risk degree of the case.

[0027] The distribution of the corresponding law enforcement resources of the law enforcement case is counted, including the number of law enforcement personnel, the equipment allocation situation of law enforcement, and the distribution density of law enforcement agencies, etc. At the same time, combined with social index data (population density, public order, public opinion index, etc.), the second risk prediction value of the current law enforcement case is predicted according to the law enforcement resource distribution and the social index data, and the guarantee ability of law enforcement resources for case handling and the influence of social environment on the case are comprehensively considered. Finally, the comprehensive prediction analysis result of the current law enforcement case is determined according to the first risk prediction value and the second risk prediction value, which provides a comprehensive and scientific decision basis for the law enforcement department, and assists it in reasonably allocating resources and formulating response strategies.

[0028] Multi-source heterogeneous data includes case basic information, scene investigation data, party information and historical similar case data.

[0029] Case basic information includes case number, occurrence time, occurrence place, case type (public security, criminal, etc.).

[0030] Scene investigation data: through professional investigation means to obtain scene traces (such as fingerprints, footprints, bloodstains, etc.), physical evidence (crime tools, left-over items, etc.), scene environment (terrain, building layout, etc.), scene image data (photos, videos), etc. Scene investigation data is the key basis for restoring the case occurrence process and determining the nature of the case.

[0031] Party information: including party identity information (name, age, occupation, address, etc.), contact information, social relations, past illegal records, etc. As the core subject of the case, the information of the party is crucial to understand the case motive and related factors.

[0032] Historical similar case data: sorting out the processing flow, investigation method, evidence collection, and final treatment result of past similar cases. With the help of historical similar case data, common rules can be mined to provide reference for the current case.

[0033] Key elements include case type, spatiotemporal characteristics, behavior patterns of involved personnel, and evidence chain association.

[0034] Case type: clearly define whether the case belongs to public security, criminal, administrative, or other different categories. Different types of cases have differences in processing flow, legal application, and social impact.

[0035] Spatiotemporal characteristics: covering the time (such as daytime or night, workday or holiday, etc.) and space (geographical location, site environment, etc.) information of the case occurrence. Spatiotemporal factors not only affect the probability and characteristics of the case occurrence, but also play an important role in evidence collection, clue tracking, etc.

[0036] Behavior patterns of involved personnel: refers to the action track, language expression, behavior habit, etc. of the involved personnel (including suspects, witnesses, victims, etc.) in the case. Through the behavior patterns, the behavior motive and psychological state of the involved personnel can be mined to provide direction for case investigation and handling.

[0037] Evidence chain association: the relationship between various types of evidence (physical evidence, documentary evidence, witness testimony, etc.) in the case, such as mutual confirmation, complement or contradiction. Clear evidence chain association is the key to restoring the facts of the case and identifying the responsibility of the case.

[0038] And determine the correlation influence coefficient set between each key element and determine the normal fluctuation range value of the correlation influence coefficient between each key element, including the following steps: determining the mutual influence relationship between each key element and establishing a set of correlation influence coefficients between each key element; respectively acquiring the correlation influence coefficients between each key element in the historical normal case and the historical abnormal case, and constructing a set of historical normal correlation influence coefficients and a set of historical abnormal correlation influence coefficients; According to the set of historical normal correlation influence coefficients and the set of historical abnormal correlation influence coefficients, the normal variation range value of the correlation influence coefficients between each key element is determined.

[0039] The key elements in the law enforcement case of the present application include case type, spatiotemporal characteristics, behavior pattern of the person involved in the case, and evidence chain correlation, etc. These elements do not exist in isolation, but interact with each other. For example, the case type will affect the behavior pattern of the person involved in the case. In a theft case, the suspect may choose a hidden time and place for the crime (the spatiotemporal characteristics are affected), and his behavior will tend to be more secret theft. For another example, the evidence chain correlation may change due to the behavior pattern of the person involved in the case. If the person involved in the case deliberately destroys the evidence, the integrity of the evidence chain will be affected.

[0040] Through in-depth analysis of a large amount of law enforcement case data, statistical methods, machine learning methods, etc. are used to mine the internal relationship between elements. For example, regression analysis is used to study the quantitative relationship between case type and behavior pattern of the person involved in the case; or association rule algorithm is used to find out the potential rules between spatiotemporal characteristics and evidence chain correlation. Based on these, the degree of mutual influence between each key element is represented by quantitative numerical values, thereby establishing a set of correlation influence coefficients, so that the influence relationship between each element is clearly presented and measured.

[0041] Construction of the set of historical normal correlation influence coefficients: Collect a large amount of historical law enforcement case data which is recognized as normal processing and the result conforms to the conventional expectation. Extract the correlation influence coefficients between each key element from these cases. The normal case usually refers to the case that occurs in the case that the law enforcement process is standardized, the evidence is clear, and the case processing conforms to the legal provisions and conventional logic. For example, in a normal public security dispute case, according to the established processing flow, the key elements such as case type and behavior pattern of the person involved in the case present stable correlation influence relationship. These coefficients are sorted and summarized to form a set of historical normal correlation influence coefficients.

[0042] Historical abnormal correlation influence coefficient set construction: screening historical abnormal law enforcement cases with complex processing process, deviating from the normal result or special circumstances. These abnormal cases may include controversial cases, cases with increased processing difficulty due to special factors, etc. Similarly, the correlation influence coefficients between the key elements in these cases are extracted to construct the historical abnormal correlation influence coefficient set. For example, in a case involving new criminal means, the correlation between the case type, the behavior pattern of the person involved, and other elements is different from that of a regular case, and the coefficient reflects this particularity.

[0043] After obtaining the historical normal correlation influence coefficient set and the historical abnormal correlation influence coefficient set, statistical analysis is performed on the historical normal correlation influence coefficient set. The mean and standard deviation are calculated, and the mean represents the average level of the correlation influence coefficient under normal circumstances, and the standard deviation reflects the dispersion degree of the coefficient. Generally, within the range of mean ± a certain multiple of standard deviation (such as ± 2 times the standard deviation), it can be considered as the normal fluctuation range. By comparing with the historical abnormal correlation influence coefficient set, this range is further verified and calibrated. If a correlation influence coefficient exceeds this range, it means that the current case may have abnormal circumstances in the corresponding element relationship, which needs further attention and analysis, providing an important reference for the risk warning and in-depth analysis of subsequent law enforcement cases.

[0044] The comparison result is obtained by comparing the current correlation influence degree set with the preset normal influence range value after processing, which includes the following steps: If there is data greater than the preset normal influence range value in the current correlation influence degree set, the case key elements corresponding to the data greater than the preset normal influence range value in the current correlation influence degree set are determined as the risk warning signal of the comparison result; If each data in the current correlation influence degree set is less than the preset normal influence range value, the difference between each data in the current correlation influence degree set and the preset normal influence range value is calculated to obtain a difference set; Preset difference threshold, if there is data greater than or equal to the difference threshold in the difference set, then continue to perform in-depth analysis on the key elements of the current law enforcement case and output the in-depth analysis signal of the comparison result.

[0045] In the law enforcement case analysis system based on digital twinning, the digital twinning model of law enforcement cases has been constructed in the early stage, the key elements have been extracted, and the normal fluctuation range value of the correlation influence coefficient between each key element has been determined. At the same time, the correlation influence degree set of the key elements of the current law enforcement case is obtained by comparing with the historical similar cases.

[0046] When the current correlation impact degree set is analyzed, if there is data greater than the preset normal impact range value, it indicates that the current law enforcement case exceeds the scope of the normal case in the correlation impact degree of the corresponding key elements. For example, in terms of the correlation impact degree of case type and behavior mode of the person involved in the case, if the current value far exceeds the normal range, it means that the case may have an abnormal behavior mode, which is different from the development trend of the conventional case. This situation beyond the normal range is likely to indicate that the case has potential risks, such as a change in the nature of the case, special actions of the suspect, etc., so the case key element corresponding to this data is determined as a risk warning signal, reminding law enforcement personnel to pay attention to it.

[0047] If each data in the current correlation impact degree set is less than the preset normal impact range value, the difference between each data and the preset normal impact range value is calculated to obtain a difference set. This is because although these data are below the normal range, the correlation impact degree of some key elements is too low, which may also indicate that the case has special circumstances. A difference threshold is preset, which is determined according to historical case data and law enforcement experience, and is used to measure whether the difference is significant enough. When there is data greater than or equal to the difference threshold in the difference set, it indicates that the correlation impact degree of the key element deviates greatly from the normal situation, and it is necessary to continue to analyze the key elements of the current law enforcement case in depth to dig out potential abnormal factors or special circumstances. At this time, a deep analysis signal is output to guide further investigation and analysis.

[0048] The deep analysis signal is combined with the dynamic evolution law of the case elements in the digital twin model of the law enforcement case to predict the change trend of the key elements of the current law enforcement case to generate a first prediction data set. According to the deep analysis signal, the dynamic evolution characteristics of the key elements of the current law enforcement case are extracted, and a dynamic evolution model of the key elements is established in combination with the dynamic evolution law of the case elements in the digital twin model of the law enforcement case. The dynamic evolution model is used to predict the change trend of the key elements of the current law enforcement case in a preset time period to generate a first prediction data set. The first prediction data set includes case type evolution trend, spatiotemporal feature change trend, behavior mode change trend of the person involved in the case, and evidence chain correlation relationship change trend.

[0049] When the deep analysis signal is obtained, it means that there is a need for in-depth exploration of the key elements of the current law enforcement case. At this time, the dynamic evolution characteristics of the key elements of the current law enforcement case are carefully excavated from the digital twin model of the law enforcement case. Taking the case type as an example, it may be necessary to analyze whether there are signs of transformation from one type to another in the development process of the current case, such as from a general public security case to a case involving criminal offenses, which may be reflected in the escalation of the behavior involved, new evidence pointing to more serious illegal facts, etc. For the spatio-temporal characteristics, attention should be paid to whether the case site has a transfer trend, whether the case occurrence time shows regular changes, etc. For the behavior pattern of the person involved, whether there are significant changes in their action track, language expression, behavior habit, etc. in the development process of the case should be observed. For the evidence chain correlation, attention should be paid to whether the appearance of new evidence shakes or supplements the original evidence chain.

[0050] The dynamic evolution characteristics of the key elements of the current case extracted are integrated into the digital twin model of the law enforcement case, which accumulates a large number of historical case element dynamic evolution rules. For example, by analyzing a large number of historical cases of the same type, the general path and key node conditions of case type evolution are summarized, and then adjusted in combination with the special circumstances of the current case. Mathematical modeling, machine learning algorithms, and other technical means are used to establish a dynamic evolution model of the key elements. This model is a dynamic system that can reflect the changes of the key elements of the case over time or in the development process of the case, and it takes into account the interaction between various elements and the influence of external environmental factors.

[0051] Using the established dynamic evolution model, the changing trend of the key elements of the current law enforcement case within a preset time period is predicted. For the evolution trend of the case type, the possibility and direction of its transformation into other types of cases in the future are predicted; for the spatio-temporal characteristic change trend, the expansion or transfer range of the case site, the concentration or dispersion trend of the case occurrence time, etc. are estimated; for the behavior pattern change trend of the person involved, the subsequent possible actions and changes in psychological state are speculated; for the evidence chain correlation change trend, whether new evidence will appear, whether the existing evidence chain will be broken or further strengthened, etc. are judged. Through the prediction of the changing trend of these key elements, a first prediction data set is finally generated.

[0052] According to the first prediction data set, the development trend of the current law enforcement case is analyzed and a first risk prediction value is obtained, which includes the following steps: Based on the first prediction data set and the digital twin model, the influence degree of environmental factors on the key elements of the case is determined to obtain an environmental influence degree set; According to the environmental influence degree set and the first prediction data set, a second prediction data set of the current law enforcement case affected by the factors is predicted; The development trend of the current law enforcement case is determined based on the second prediction data set, and a first risk prediction value of the current law enforcement case is determined according to the development trend of the current law enforcement case.

[0053] The first prediction data set covers key information such as case type evolution trend, spatio-temporal feature change trend, suspect behavior pattern change trend, and evidence chain correlation change trend. At the same time, the digital twin model of the law enforcement case completely maps various elements and states of the case. On this basis, environmental factors such as population density, public order, public opinion atmosphere, and geographical environment of the case area are considered to analyze the effects of these environmental factors on the key elements of the case.

[0054] For example, if the population density of the case area is high, it may affect the movement trajectory of the suspect (spatio-temporal feature and behavior pattern), and may also interfere with the collection and preservation of evidence (evidence chain correlation); if the local public order is not good, it may increase the risk of case deterioration or derivation of other illegal criminal behaviors (case type evolution). By quantitatively analyzing these influences, an environmental influence degree set is obtained to represent the size and direction of the influence of different environmental factors on each key element.

[0055] According to the environmental influence degree set and the first prediction data set, the development of the law enforcement case considering the effects of environmental factors is further predicted. The environmental influence degree set clearly shows the influence mode and degree of environmental factors on key elements, while the first prediction data set gives the preliminary change trend of key elements without considering environmental factors.

[0056] The future trend of the key elements of the case is re-evaluated by combining the two. For example, originally it is predicted that the case type will not change, but considering the local public opinion fermentation that may cause public attention and pressure, leading to a more strict case handling procedure, which may cause the case type to evolve to a more serious direction. Through such comprehensive analysis, the second prediction data set is generated, which more comprehensively and accurately reflects the development trend of the case in the actual environment.

[0057] The development trend of the current law enforcement case is determined based on the second prediction data set. The development trend may include different situations such as case easing, maintaining the status quo, and deteriorating. For example, if it is predicted that the behavior pattern of the suspect is gradually stable, the evidence chain is continuously improved and clear, the case type has no deterioration trend, and the environmental factors have no adverse effects, it can be judged that the case development trend is to ease.

[0058] According to the determined development trend, the severity of the case, the potential harm, the processing difficulty and other factors are comprehensively considered to determine the first risk prediction value of the current law enforcement case. The risk prediction value can be represented by grades (such as low, medium, and high risk) or specific numerical values (such as a risk score of 0-10) to provide a visual risk reference for law enforcement departments to reasonably allocate resources and develop response strategies.

[0059] The distribution of law enforcement resources includes the number of law enforcement personnel, the equipment configuration of law enforcement, and the distribution density of law enforcement agencies.

[0060] The number of law enforcement personnel: Law enforcement personnel are the direct executors of law enforcement work, and their number directly affects the efficiency and effectiveness of case handling. During case handling, if the number of law enforcement personnel is sufficient, they can quickly respond to cases, conduct on-site investigation, collect evidence, track personnel, and timely control the situation to reduce the risk of case deterioration. For example, in large-scale activity security or major emergency cases, sufficient police deployment can effectively maintain order and ensure smooth case investigation. On the contrary, insufficient number of law enforcement personnel may lead to delayed case handling, missing the best investigation opportunity, and increasing the difficulty of case handling.

[0061] The equipment configuration of law enforcement: Advanced and complete law enforcement equipment is an important guarantee for improving law enforcement efficiency. For example, high-definition monitoring equipment helps to obtain clear image data of the scene, providing key clues for case investigation; advanced evidence detection equipment can more accurately analyze evidence to improve the evidence chain. Insufficient or outdated law enforcement equipment may limit the investigation depth and breadth of law enforcement personnel, affecting the progress and quality of case handling.

[0062] The distribution density of law enforcement agencies: Reasonable distribution density of law enforcement agencies can ensure effective coverage of the jurisdiction. In areas with high distribution density of law enforcement agencies, cases can be quickly responded to and timely handled. For example, in urban center areas, due to the concentration of population and business activities, a higher distribution density of law enforcement agencies can quickly respond to various cases. In remote areas, if the distribution of law enforcement agencies is sparse, the response time for cases may be prolonged, which may adversely affect the timeliness and effectiveness of case handling.

[0063] Social indicator data includes population density, public security status, and public opinion index.

[0064] Population density: Population density reflects the degree of concentration of personnel in the region. In high population density areas, the probability of case occurrence may be relatively high, and the impact range and speed of the case may be faster. For example, in crowded urban neighborhoods, cases such as theft and disputes are more likely to occur, and the concentration of personnel may lead to the case attracting public attention or secondary problems. At the same time, case handling in high population density areas may face more interference factors, increasing the difficulty of law enforcement.

[0065] Public order: Public order is a comprehensive reflection of the overall safety level of an area. In areas with good public order, the frequency and severity of cases are generally lower, and law enforcement work is relatively smooth. In contrast, in areas with poor public order, various illegal and criminal activities may be more rampant, and law enforcement departments face greater case pressure and may encounter more obstacles when handling cases, such as witnesses not daring to testify, illegal personnel resisting law enforcement, etc.

[0066] Public sentiment index: The public sentiment index reflects the public's attention to law enforcement cases and the direction of public opinion. In the current era of rapid information dissemination, law enforcement cases are easily subject to social public opinion. Positive public opinion helps to create a good law enforcement environment and gain public support and cooperation. However, negative public opinion or the uncontrolled fermentation of public opinion can bring great pressure to law enforcement departments, affect the progress of case handling, and even trigger social instability factors. Law enforcement departments need to closely monitor the public sentiment index and respond to public concerns in a timely manner to guide public opinion.

[0067] According to the distribution of law enforcement resources, social index data, and the first risk prediction value, a comprehensive prediction analysis result of the current law enforcement case is obtained, including the following steps: According to the distribution of law enforcement resources and social index data, a second risk prediction value of the current law enforcement case is predicted; According to the first risk prediction value and the second risk prediction value, the comprehensive prediction analysis result of the current law enforcement case is determined.

[0068] The distribution of law enforcement resources includes the number of law enforcement personnel, the equipment configuration of law enforcement, and the distribution density of law enforcement agencies. Social index data includes population density, public order, and public sentiment index. These factors will all have an impact on the handling of law enforcement cases.

[0069] Insufficient number of law enforcement personnel leads to delayed response and handling of cases, thereby increasing the risk of cases; outdated law enforcement equipment makes it difficult to obtain key evidence or effectively control the scene, also increasing the risk level; sparse distribution of law enforcement agencies makes it difficult to quickly reach the scene of the incident, also increasing the risk. High population density increases the potential likelihood of case occurrence and can easily trigger a chain reaction when handling; poor public order with frequent illegal and criminal activities increases the difficulty and uncertainty of case handling; high public sentiment index can easily trigger social attention and negative public opinion if not handled properly, bringing additional pressure to law enforcement.

[0070] By analyzing these law enforcement resource distribution and social index data, using appropriate analysis models (such as risk assessment models based on historical case data), the risk of the current law enforcement case is assessed to obtain a second risk prediction value. This value reflects the comprehensive impact of law enforcement resources and social environmental factors on case risk.

[0071] The first risk prediction value is based on the trend of changes in key elements of law enforcement cases, the influence of environmental factors on the cases, and the like, and mainly focuses on the development trend and inherent risks of the cases. The second risk prediction value focuses on the influence of law enforcement resources and external social environment on the cases.

[0072] The first risk prediction value and the second risk prediction value are comprehensively considered. A weighted average or the like can be used to give different weights to the two risk prediction values according to the characteristics and actual situation of different cases. For example, for some major cases that require a large amount of law enforcement resource investment, a higher weight can be given to the second risk prediction value; for some cases mainly affected by case itself factors, a higher weight is given to the first risk prediction value. Through comprehensive calculation and analysis, the comprehensive prediction analysis result of the current law enforcement case is determined, which provides a comprehensive and accurate basis for law enforcement department decision-making, and assists it in reasonably allocating resources and formulating scientific response strategies.

[0073] As shown in Fig. 3 The electronic device can include a processor 610, a communications interface 620, a memory 630, and a communications bus 640, wherein the processor 610, the communications interface 620, and the memory 630 complete mutual communication through the communications bus 640. The processor 610 can invoke the logic instructions in the memory 630 to execute a content data prediction analysis method based on digital twinning.

[0074] In addition, the logic instructions in the memory 630 described above can be implemented in the form of a software function unit and sold or used as an independent product, which can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, includes several instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the method described in various embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0075] On the other hand, the present application also provides a computer program product, which includes a computer program, the computer program can be stored on a non-transitory computer readable storage medium, and the computer program is executed by a processor, and the computer can execute a content data prediction analysis method based on digital twinning.

[0076] In yet another aspect, the present application also provides a non-transitory computer readable storage medium having stored thereon a computer program, which, when executed by a processor, implements a content data prediction analysis method based on digital twinning.

[0077] The device embodiments described above are merely illustrative, wherein the units described as separate components can or can not be physically separate, and the components displayed as units can or can not be physical units, i.e., can be located in one place, or can be distributed on multiple network units. Part or all of the modules can be selected to achieve the purpose of the present embodiment scheme according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0078] From the above description of the embodiments, those skilled in the art can clearly understand that the embodiments can be realized by means of software plus necessary universal hardware platforms, and of course can also be realized by hardware. Based on such understanding, the above technical solutions, essentially or in other words, the part that contributes to the prior art, can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in each embodiment or some parts of the embodiments.

[0079] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to some technical features; and these modifications or replacements do not make the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A content data prediction and analysis method based on digital twins, characterized in that: The method comprises the following steps: Collect multi-source heterogeneous data related to law enforcement cases, fuse and process the multi-source heterogeneous data to construct a digital twin model of law enforcement cases; Extract key elements from law enforcement cases based on the digital twin model, determine the set of correlation influence coefficients between key elements, and determine the normal range of variation of the correlation influence coefficients between key elements; After determining the correlation influence degree between the current law enforcement case and similar historical cases in terms of key elements based on the correlation influence coefficient set, a current correlation influence degree set is obtained. After processing the current correlation influence degree set, the set is compared with a preset normal influence range value to obtain a comparison result; The deep analysis signal in the comparison result is combined with the dynamic evolution law of the case elements in the digital twin model of the law enforcement case to predict the change trend of the key elements of the current law enforcement case to generate the first prediction data set; Analyze the development trend of current law enforcement cases based on the first prediction data set and obtain a first risk prediction value; The distribution of law enforcement resources corresponding to statistical law enforcement cases is analyzed, and comprehensive forecast analysis results are obtained by conducting comprehensive forecast analysis on current law enforcement cases based on the distribution of law enforcement resources, social indicator data and the first risk prediction value.

2. The content data prediction and analysis method based on digital twin according to claim 1 is characterized in that: The multi-source heterogeneous data includes basic case information, on-site investigation data, party information and historical similar case data.

3. The content data prediction and analysis method based on digital twin according to claim 2 is characterized in that: The key elements include case type, temporal and spatial characteristics, behavioral patterns of persons involved, and correlation relationships in the chain of evidence.

4. The content data prediction and analysis method based on digital twins according to claim 3, and determining the set of correlation influence coefficients between key elements and determining the normal variation range of the correlation influence coefficients between key elements, specifically comprising the following steps: Determine the mutual influence relationship between key factors and establish the correlation influence coefficient set between key factors; Obtain the correlation influence coefficients between key elements in historical normal cases and historical abnormal cases respectively, and construct a historical normal correlation influence coefficient set and a historical abnormal correlation influence coefficient set; According to the historical normal correlation influence coefficient set and the historical abnormal correlation influence coefficient set, the normal variation range value of the correlation influence coefficient between each key factor is determined.

5. The content data prediction and analysis method based on digital twin according to claim 4 is characterized in that: After processing the current set of associated impact levels, a comparison is made with a preset normal impact range value to obtain a comparison result, specifically including the following steps: If there is data with a higher correlation impact than the preset normal impact range, the key case elements corresponding to the data with a higher correlation impact than the preset normal impact range will be determined as risk warning signals of the comparison results; If all data in the current correlation influence degree set are smaller than the preset normal influence range value, the difference between each data in the current correlation influence degree set and the preset normal influence range value is calculated to obtain a difference value set; A difference threshold is preset. If there is data in the difference set that is greater than or equal to the difference threshold, the key elements of the current law enforcement case will continue to be deeply analyzed and a deep analysis signal of the comparison result will be output.

6. The content data prediction and analysis method based on digital twin according to claim 5 is characterized in that: The deep analysis signal is combined with the dynamic evolution law of case elements in the digital twin model of law enforcement cases to predict the changing trend of key elements of current law enforcement cases to generate the first prediction data set; Extract the dynamic evolution characteristics of the key elements of current law enforcement cases based on in-depth analysis signals, and establish a dynamic evolution model of key elements based on the dynamic evolution laws of case elements in the digital twin model of law enforcement cases; A first prediction data set is generated by using a dynamic evolution model to predict the changing trends of key elements of current law enforcement cases within a preset time period; wherein, the first prediction data set includes the evolution trend of case types, the changing trend of spatiotemporal characteristics, the changing trend of behavioral patterns of people involved in the case, and the changing trend of the correlation relationship of the evidence chain.

7. The content data prediction and analysis method based on digital twin according to claim 6 is characterized in that: Analyzing the development trend of current law enforcement cases based on the first prediction data set and obtaining a first risk prediction value specifically includes the following steps: An environmental impact degree set is obtained by determining the degree of influence of environmental factors on key elements of the case based on the first prediction data set and the digital twin model; Predicting a second prediction data set affected by factors affecting current law enforcement cases based on the environmental impact degree set and the first prediction data set; The development trend of the current law enforcement case is determined based on the second prediction data set, and the first risk prediction value of the current law enforcement case is determined according to the development trend of the current law enforcement case.

8. The content data prediction and analysis method based on digital twin according to claim 7 is characterized in that: The distribution of law enforcement resources includes the number of law enforcement personnel, the equipment of law enforcement equipment and the distribution density of law enforcement agencies.

9. The content data prediction and analysis method based on digital twin according to claim 8 is characterized in that: The social indicator data include population density, public security conditions and public opinion index.

10. The content data prediction and analysis method based on digital twin according to claim 9 is characterized in that: Based on the distribution of law enforcement resources, social indicator data and the first risk prediction value, a comprehensive forecast analysis of current law enforcement cases is conducted to obtain a comprehensive forecast analysis result, which specifically includes the following steps: Predict the second risk prediction value of current law enforcement cases based on the distribution of law enforcement resources and social indicator data; The comprehensive prediction and analysis results of the current law enforcement case are determined based on the first risk prediction value and the second risk prediction value.

Images