Public security big data-based key personnel abnormal behavior management and control prediction method and management and control prediction system

By extracting features from public security big data and constructing classification and prediction models, the shortcomings of traditional key personnel management methods have been addressed, enabling real-time and accurate prediction and control of abnormal behavior of key personnel, thereby improving the efficiency and accuracy of public security work.

CN119046825APending Publication Date: 2024-11-29ANHUI SUN CREATE ELECTRONICS
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410915674.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-09
Publication Date
2024-11-29

AI Technical Summary

Technical Problem

Traditional methods for managing key personnel rely on manual investigations, which are labor-intensive and difficult to achieve comprehensive, real-time, and accurate management. How to utilize big data technology for accurate prediction and timely management is an urgent problem to be solved.

Method used

By collecting and preprocessing data, extracting features, training models, predicting and controlling abnormal behavior, and using machine learning and deep learning algorithms to extract features from public security big data, we can build classification and prediction models to achieve real-time and accurate prediction and control of abnormal behavior of key personnel.

Benefits of technology

It improved the accuracy and efficiency of predicting abnormal behavior of key personnel, achieved real-time and precise control, and enhanced the efficiency and accuracy of public security work.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119046825A_ABST
    Figure CN119046825A_ABST
Patent Text Reader

Abstract

The embodiment of the invention discloses a management and control prediction method and system for abnormal behaviors of key persons based on public security big data, and relates to the technical field of public security big data processing and analysis, and the management and control prediction method comprises the steps of data collection and preprocessing, feature extraction, model training, abnormal behavior prediction and management and control measures. Through data acquisition and preprocessing, feature extraction, model training, abnormal behavior prediction and management and control measures, the prediction accuracy and efficiency of abnormal behaviors of key personnel can be improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of public security big data processing and analysis, and in particular to a control and prediction method and control and prediction system for abnormal behavior of key personnel based on public security big data. Background Art

[0002] In public security work, the control and prediction of key personnel is an important task. However, traditional key personnel control methods mainly rely on manual investigation and monitoring, which not only has a large workload, but also makes it difficult to achieve comprehensive, real-time and accurate control. Therefore, how to use modern information technology, especially big data technology, to achieve accurate prediction and timely control of abnormal behaviors of key personnel is an urgent problem to be solved.

[0003] With the rapid development of information technology, public security departments have accumulated a large amount of data, including identity information, trajectory data, case information, etc. of key personnel. These data make it possible to use big data technology to conduct management and prediction of key personnel.

[0004] However, how to extract useful information from these data and predict and control the abnormal behavior of key personnel is the main problem currently faced. Summary of the invention

[0005] In order to solve the problems existing in the prior art, the present invention provides a control and prediction method and a control and prediction system for abnormal behavior of key personnel based on public security big data, which can improve the accuracy and efficiency of predicting abnormal behavior of key personnel through data collection and preprocessing, feature extraction, model training, abnormal behavior prediction and control measures.

[0006] The technical solution adopted by the present invention to solve the above technical problems is as follows:

[0007] The present application provides a method for controlling and predicting abnormal behaviors of key personnel based on public security big data, the method comprising:

[0008] S1. Data collection and preprocessing: Through the data interface, obtain the information of key personnel from the public security big data, and preprocess the collected information data to prepare for providing feature data for subsequent model training;

[0009] S2. Feature extraction: Use machine learning and deep learning algorithms to extract features related to abnormal behaviors of key personnel from pre-processed information data;

[0010] S3. Model training: Use the extracted features and existing abnormal behavior cases of key personnel to build classification models and prediction models, and optimize the models through cross-validation methods;

[0011] S4. Prediction of abnormal behavior: input the key personnel data collected in real time into the classification model and prediction model, and predict the abnormal behavior of key personnel through the prediction model;

[0012] S5. Control measures: Take appropriate control measures based on the forecast results.

[0013] Optionally, in some embodiments of the present application, the main process of data collection and preprocessing is specifically:

[0014] Obtain information data of key personnel from public security big data through data interface;

[0015] Preprocess the acquired data to eliminate redundant and abnormal data and improve data quality to enhance the accuracy and efficiency of subsequent model training;

[0016] During the data collection process, scheduled tasks or real-time data updates are set to maintain the real-time and accuracy of the data;

[0017] The acquired data is stored in a database or file system for subsequent data analysis and model training.

[0018] Optionally, in some embodiments of the present application, methods for preprocessing data include cleaning, deduplication and normalization methods.

[0019] Optionally, in some embodiments of the present application, the feature extraction includes:

[0020] Determine the feature extraction method: Select the appropriate feature extraction method based on different data types and problem backgrounds;

[0021] Conduct exploratory analysis on data: Use statistical and visualization methods to conduct exploratory analysis on preprocessed data to understand the distribution, regularity, and correlation characteristics of the data, and provide guidance for subsequent feature extraction;

[0022] Feature selection: Perform feature selection on the data through machine learning and deep learning methods to select features related to the abnormal behavior of the key personnel;

[0023] Feature transformation: appropriately transform the selected features to improve the accuracy and generalization ability of the model;

[0024] Feature combination: Combine the selected features to obtain richer feature information.

[0025] Optionally, in some embodiments of the present application, the model training includes:

[0026] Select classification model and prediction model: Select appropriate classification model and prediction model according to the characteristics of the problem and data type;

[0027] Adjust model parameters as well as feature selection and transformation methods to improve the model’s predictive accuracy;

[0028] Input the extracted features and labeled data into the classification model and prediction model;

[0029] The model was trained and evaluated using cross-validation methods to determine the performance and accuracy of the model;

[0030] Based on the evaluation results, the model is adjusted and optimized.

[0031] Optionally, in some embodiments of the present application, the abnormal behavior prediction includes:

[0032] The classification model classifies the behavior of key personnel and determines whether it is abnormal behavior;

[0033] The predictive model predicts the future behavior of key personnel, and the prediction results are used to deploy control and tracking measures.

[0034] Optionally, in some embodiments of the present application, the control measures include:

[0035] Early warning: When the system predicts that key personnel have abnormal behavior, the system automatically triggers an early warning signal to remind relevant personnel to pay attention and deal with it in time;

[0036] Control: According to the prediction results, formulate corresponding control plans and control key personnel. At the same time, the system can automatically generate control instructions and pass them to the control personnel through the control terminal equipment;

[0037] Tracking: After the implementation of the control, the system tracks the actions of key personnel in real time, obtains their location, trajectory and other information, and transmits it to the system through the data interface for real-time processing and analysis;

[0038] Management and control effectiveness evaluation: After implementing management and control measures, the management and control effectiveness is evaluated through evaluation indicators. Based on the evaluation results, the management and control measures are optimized and adjusted to improve the management and control effectiveness.

[0039] The present application also provides a control and prediction system for abnormal behavior of key personnel based on public security big data, which is used to implement the above-mentioned control and prediction method. The system includes the following modules:

[0040] Data collection module: obtain information data of key personnel from public security big data;

[0041] Data preprocessing module: cleans, removes duplicates and normalizes the collected data to provide feature data for subsequent model training;

[0042] Feature extraction module: uses learning algorithms to extract features related to abnormal behaviors of key personnel from preprocessed data;

[0043] Model training module: Combine existing abnormal behavior cases of key personnel to build classification models and prediction models, and optimize the models;

[0044] Abnormal behavior prediction module: input the key personnel data collected in real time into the classification model and prediction model to predict the abnormal behavior of key personnel;

[0045] Control measures module: Take corresponding control measures based on the prediction results.

[0046] Compared with the prior art, the beneficial effects of the present invention are:

[0047] 1. Improve prediction accuracy and efficiency: Extract the features of key groups through machine learning, deep learning and other algorithms, and combine the features. It is possible to extract information by combining multiple features, and train the model on the extracted information, so that the corresponding model can be optimized according to the characteristics of the key group, adjust the corresponding parameters, improve the accuracy of the prediction, and thus improve the accuracy and efficiency of predicting abnormal behaviors of key personnel.

[0048] 2. Realize real-time and accurate management and control: By inputting the key personnel data collected in real time into the classification model and the prediction model, the key groups can be identified and predicted respectively. The classification model can determine whether the key groups have abnormal behavior, and the prediction model can predict the behavior of key personnel. The abnormal behavior of key personnel can be predicted in time, thereby realizing real-time and accurate management and control, and improving the efficiency and accuracy of public security work. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0050] Figure 1 A schematic diagram of a specific process of a method for controlling and predicting abnormal behaviors of key personnel based on public security big data provided in an embodiment of the present application;

[0051] Figure 2 A schematic diagram of a specific process of data collection and preprocessing provided in an embodiment of the present application;

[0052] Figure 3 A schematic diagram of a specific process of feature extraction provided in an embodiment of the present application;

[0053] Figure 4 A schematic diagram of the specific process of the control measures provided in the embodiments of the present application. DETAILED DESCRIPTION

[0054] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without making creative work belong to the protection scope of the present application. It is understood that the drawings are only provided for reference and illustration, and are not used to limit the present application. The connection relationship shown in the drawings is only for the convenience of clear description and does not limit the connection method.

[0055] Specifically, Figure 1 As shown, the embodiment of the present application provides a method for predicting and controlling abnormal behaviors of key personnel based on public security big data, and the method mainly includes data collection and preprocessing, feature extraction, model training, abnormal behavior prediction and control measures. The above five steps are carried out sequentially to facilitate the investigation and control tracking of abnormal behaviors by public security big data, specifically:

[0056] Through the data interface, the system can be connected to the public security big data to obtain the basic information of relevant key groups, including identity information, trajectory data and case information. The identity information includes the height, gender, ID card information and other basic information of the key groups; the trajectory data includes the position of the key groups moving in a unit time. The unit time can be set to one year, one month, or one week. The specific time length can be set according to the actual situation. The position of the key groups moving in a unit time can be collected through the trajectory data, which is convenient for detecting and tracking the key groups; the case information includes the record content of the cases committed by the key groups in the past time period.

[0057] After obtaining the basic information of key populations, the information needs to be preprocessed, which mainly includes cleaning, deduplication and normalization. In the embodiment of the present application, data cleaning mainly includes supplementation or discarding of missing values, data standardization, and processing of outliers, specifically removing duplicate, erroneous, missing or abnormal data to improve data quality. Specific operations include checking data integrity, correcting errors, filling missing values, deleting duplicate data, etc.; the supplementation and discarding of missing values ​​processes the original data to eliminate redundant data and abnormal data. Specifically, the data is converted into a standard scale to eliminate the impact of data differences on model training. For example, the data can be standardized to the range of [0,1] so that the differences between different data are within a smaller range, thereby improving the training efficiency and accuracy of the model; data deduplication is to remove duplicate records by comparing the unique identifiers in the data, such as ID number, mobile phone number, etc., to ensure data accuracy; data normalization is to convert the data into a standard scale to eliminate the impact of data differences on model training; outlier processing identifies and processes abnormal data through statistical, probabilistic and other methods, such as marking abnormal data as special categories or directly eliminating them; it can improve the overall quality of the data and facilitate subsequent feature extraction of the data to improve the accuracy and efficiency of subsequent model training.

[0058] Among them, in the process of data acquisition, due to the characteristics of the specific real-time activities of key groups, it is necessary to set up scheduled tasks or update data in real time to maintain the real-time and accuracy of the data. In the embodiment of the present application, the scheduled task is mainly processed by the open xxl-job distributed task scheduling platform. The scheduling platform has an interface to maintain scheduled tasks and trigger rules, which is very easy to manage, and the trajectory of key groups can be clearly tracked through the visual management interface, which is convenient for system identification and processing; real-time updated data can be processed through scheduled tasks to improve update efficiency.

[0059] The system stores the preprocessed data in the database and file system to facilitate data use.

[0060] Through the above steps, it is possible to efficiently and accurately obtain relevant data of key personnel from public security big data, providing reliable data support for subsequent abnormal behavior prediction and control.

[0061] After completing the above data collection and processing, the system performs feature extraction on the processed data. Feature extraction mainly includes:

[0062] 1. Determine the method of feature extraction: Choose the appropriate feature extraction method according to different data types and problem backgrounds. For example, for text data, you can use methods such as bag-of-words model and TF-IDF to extract features; for image data, you can use methods such as convolutional neural network to extract features;

[0063] 2. Conduct exploratory analysis on the data: Conduct exploratory analysis on the preprocessed data through statistical and visualization methods to understand the distribution, regularity, and correlation of the data, and provide guidance for subsequent feature extraction;

[0064] 3. Feature selection: Use machine learning, deep learning and other methods to select features related to abnormal behavior of key personnel. You can use filtering, packaging or embedded methods to select features. For example, you can use correlation analysis, chi-square test and other methods to select features with a high correlation with abnormal behavior;

[0065] 4. Feature transformation: appropriately transform the selected features to improve the accuracy and generalization ability of the model. For example, text data can be transformed into word vectors, or continuous variables can be transformed into categorical variables;

[0066] 5. Feature combination: Combine selected features to obtain richer feature information. For example, you can use a combination of multiple features as input, or use methods such as feature crossover to generate new features;

[0067] 6. Evaluate and optimize the extracted features: Use classification models or prediction models to evaluate and optimize the extracted features to determine the best feature combination and feature transformation method.

[0068] Through the above feature extraction steps, features related to abnormal behaviors of key personnel can be extracted from the preprocessed data, providing effective feature input for subsequent model training and prediction.

[0069] After the system extracts the features, it trains the model based on the features. The specific training process is as follows:

[0070] Using the extracted features and combining them with existing cases of abnormal behavior of key personnel, we build classification and prediction models, and optimize the models through cross-validation and other methods to improve the prediction accuracy of the models.

[0071] The model can be constructed based on the characteristics of the problem and the data type, such as classification models such as support vector machines, decision trees, random forests, and prediction models such as time series analysis and regression analysis. After selection, the features and labeled data are input into the classification model and prediction model. Before inputting the data, different types of data (such as text, images, etc.) need to be converted into numerical data for subsequent model training.

[0072] The classification model classifies the behavior of key personnel to determine whether it is abnormal behavior; the prediction model predicts the future behavior of key personnel, and the prediction results are used to deploy control and tracking measures.

[0073] After selection, the model parameters will be adjusted. Specifically, according to the actual situation, the model parameters such as learning rate, number of iterations, etc. will be adjusted to improve the accuracy and generalization ability of the model.

[0074] Finally, adjust the feature selection and transformation methods: According to the evaluation results, adjust the feature selection and transformation methods to improve the prediction accuracy of the model.

[0075] After the model is trained, abnormal behavior can be predicted. The key personnel data collected in real time can be input into the classification model and the prediction model, and the abnormal behavior of key personnel can be predicted through the prediction model.

[0076] The specific process is as follows:

[0077] The classification model classifies the behavior of key personnel and determines whether it is abnormal behavior. For example, it can determine whether the key personnel's activities are within the normal range based on their trajectory data, or determine whether there are abnormal social relationships based on their social network information.

[0078] The prediction model predicts the future behavior of key personnel, and the prediction results can be used for deployment, tracking and other measures. For example, the possible future activity trajectory can be predicted based on historical trajectory data, or the possible future behavior pattern can be predicted based on historical behavior patterns.

[0079] After the prediction results are output, the output results of the classification model and the prediction model are passed to the system, and the system takes corresponding control measures based on the prediction results. For example, if abnormal behavior of key personnel is predicted, the system can automatically trigger an early warning signal, or formulate a corresponding control plan based on the prediction results.

[0080] Finally, the model is updated. According to the actual application situation and prediction results, the classification model and prediction model are updated and optimized to improve the prediction accuracy and efficiency. For example, the model can be retrained according to the new data set, or the parameters and structure of the model can be adjusted according to the actual application situation.

[0081] Through the above specific implementation process, the abnormal behavior of key personnel can be predicted and controlled in real time, providing more accurate and efficient support for public security work.

[0082] Finally, control measures are used to deal with it, mainly by taking corresponding control measures based on the prediction results.

[0083] Control measures include:

[0084] Early warning: When the system predicts that key personnel have abnormal behavior, the system automatically triggers an early warning signal to remind relevant personnel to pay attention and deal with it in time;

[0085] Control: According to the prediction results, formulate corresponding control plans and control key personnel. At the same time, the system can automatically generate control instructions and pass them to the control personnel through the control terminal equipment;

[0086] Tracking: After the implementation of the control, the system tracks the actions of key personnel in real time, obtains their location, trajectory and other information, and transmits it to the system through the data interface for real-time processing and analysis;

[0087] Management and control effectiveness evaluation: After implementing management and control measures, the management and control effectiveness is evaluated through evaluation indicators. Based on the evaluation results, the management and control measures are optimized and adjusted to improve the management and control effectiveness.

[0088] Through the above specific implementation process, the abnormal behavior of key personnel can be predicted and controlled in real time, providing more accurate and efficient support for public security work.

[0089] The above embodiments all need to be carried out through a prediction system, specifically a control and prediction system for abnormal behaviors of key personnel based on public security big data. The system mainly includes the following modules: data acquisition module, data preprocessing module, feature extraction module, model training module, abnormal behavior prediction module and control measures module.

[0090] Data collection module: obtain information data of key personnel from public security big data;

[0091] Data preprocessing module: cleans, removes duplicates and normalizes the collected data to provide feature data for subsequent model training;

[0092] Feature extraction module: uses learning algorithms to extract features related to abnormal behaviors of key personnel from preprocessed data;

[0093] Model training module: Combine existing abnormal behavior cases of key personnel to build classification models and prediction models, and optimize the models;

[0094] Abnormal behavior prediction module: input the key personnel data collected in real time into the classification model and prediction model to predict the abnormal behavior of key personnel;

[0095] Control measures module: Take corresponding control measures based on the prediction results.

[0096] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A control and prediction method for abnormal behavior of key personnel based on public security big data, characterized in that: The method comprises: S1. Data collection and preprocessing: Through the data interface, obtain the information of key personnel from the public security big data, and preprocess the collected information data to prepare for providing feature data for subsequent model training; S2. Feature extraction: Use machine learning and deep learning algorithms to extract features related to abnormal behaviors of key personnel from pre-processed information data; S3. Model training: Use the extracted features and existing abnormal behavior cases of key personnel to build classification models and prediction models, and optimize the models through cross-validation methods; S4. Prediction of abnormal behavior: input the key personnel data collected in real time into the classification model and prediction model, and predict the abnormal behavior of key personnel through the prediction model; S5. Control measures: Take appropriate control measures based on the forecast results.

2. The control prediction method according to claim 1, characterized in that: The main process of data collection and preprocessing is specifically as follows: Obtain information data of key personnel from public security big data through data interface; Preprocess the acquired data to eliminate redundant and abnormal data and improve data quality to enhance the accuracy and efficiency of subsequent model training; During the data collection process, scheduled tasks or real-time data updates are set to maintain the real-time and accuracy of the data; The acquired data is stored in a database or file system for subsequent data analysis and model training.

3. The control prediction method according to claim 2, characterized in that: Methods for preprocessing data include cleaning, deduplication, and normalization methods.

4. The control prediction method according to claim 1, characterized in that: The feature extraction comprises: Determine the feature extraction method: Select the appropriate feature extraction method based on different data types and problem backgrounds; Conduct exploratory analysis on data: Use statistical and visualization methods to conduct exploratory analysis on preprocessed data to understand the distribution, regularity, and correlation characteristics of the data, and provide guidance for subsequent feature extraction; Feature selection: Perform feature selection on the data through machine learning and deep learning methods to select features related to the abnormal behavior of the key personnel; Feature transformation: transform the selected features to improve the accuracy and generalization ability of the model; Feature combination: Combine the selected features to obtain richer feature information.

5. The control prediction method according to claim 1, characterized in that: The model training includes: Select classification model and prediction model: Select appropriate classification model and prediction model according to the characteristics of the problem and data type; Adjust model parameters as well as feature selection and transformation methods to improve the model’s predictive accuracy; Input the extracted features and labeled data into the classification model and prediction model; The model was trained and evaluated using cross-validation methods to determine the performance and accuracy of the model; Based on the evaluation results, the model is adjusted and optimized.

6. The control prediction method according to claim 1, characterized in that: The abnormal behavior prediction includes: The classification model classifies the behavior of key personnel and determines whether it is abnormal behavior; The predictive model predicts the future behavior of key personnel, and the prediction results are used to deploy control and tracking measures.

7. The control prediction method according to claim 1, characterized in that: The control measures include: Early warning: When the system predicts that key personnel have abnormal behavior, the system automatically triggers an early warning signal to remind relevant personnel to pay attention and deal with it in time; Control: According to the prediction results, formulate corresponding control plans and control key personnel. At the same time, the system can automatically generate control instructions and pass them to the control personnel through the control terminal equipment; Tracking: After the implementation of the control, the system tracks the actions of key personnel in real time, obtains their location and trajectory information, and transmits it to the system through the data interface for real-time processing and analysis; Management and control effectiveness evaluation: After implementing management and control measures, the management and control effectiveness is evaluated through evaluation indicators. Based on the evaluation results, the management and control measures are optimized and adjusted to improve the management and control effectiveness.

8. A control and prediction system for abnormal behavior of key personnel based on public security big data, used to implement the control and prediction method as described in any one of claims 1 to 7, characterized in that: The system includes the following modules: Data collection module: obtain information data of key personnel from public security big data; Data preprocessing module: cleans, removes duplicates and normalizes the collected data to provide feature data for subsequent model training; Feature extraction module: uses learning algorithms to extract features related to abnormal behaviors of key personnel from preprocessed data; Model training module: Combine existing abnormal behavior cases of key personnel to build classification models and prediction models, and optimize the models; Abnormal behavior prediction module: input the key personnel data collected in real time into the classification model and prediction model to predict the abnormal behavior of key personnel; Control measures module: Take corresponding control measures based on the prediction results.