Big data-based supervised person supervision relieving checking system for supervised reconstruction of control place

Abstract

The invention discloses a big data-based supervised person supervision relieving checking system for supervised place supervision transformation, and relates to the technical field of supervised place management. Comprising a data acquisition module, a data cleaning and verification module, a data integration module, a data analysis module, a risk assessment module, a real-time monitoring module, an early warning module and an intervention management module. Potential laws and trends are analyzed and mined through the data analysis module, then the supervisory risk relieving coefficient of each individual controlled person is calculated through the risk assessment module according to supervised data of each controlled person, and the supervisory risk relieving coefficient of each controlled person is compared with a threshold value of the early warning module. And then different intervention and management measures are taken according to different risk ranges, so that precise prediction and effective management of the controlled personnel are realized, the controlled personnel are helped to better return to the society, and the risk of illegal behaviors is reduced.

Description

Technical Field

[0001] The present invention relates to the field of supervision management technology, and in particular to a supervision reform and supervision verification system for supervised persons based on big data control. Background Art

[0002] In today's society, with the rapid development and widespread application of information technology, big data has become an important means to improve management efficiency and optimize decision-making processes in various fields. In the field of regulatory transformation, building a regulatory release verification system based on big data has important practical significance and urgency. Regulatory authorities are responsible for the transformation and education of regulated persons. However, traditional regulatory management methods have many challenges and shortcomings in the release verification of regulated persons. In the past, regulatory authorities often relied on paper files and manual records for information management of regulated persons. This method is not only inefficient, but also prone to inaccurate and incomplete information. When a supervised person is released from supervision, it is necessary to conduct a comprehensive review of his / her performance, reformation, and educational achievements at the place of supervision to determine whether he / she has met the standards for release from supervision. However, due to the dispersion and non-standardization of information, this process is often time-consuming and laborious, and it is difficult to ensure accuracy. In addition, traditional review methods make it difficult to conduct in-depth analysis of the behavioral patterns and psychological states of supervised persons. The behavioral performance of supervised persons at the place of supervision is an important basis for evaluating the effectiveness of their reformation, but it is difficult to discover potential patterns and trends by relying solely on manual observation and recording. It is also difficult to effectively identify and intervene in advance for some supervised persons who are at risk of committing illegal acts again.

[0003] The existing supervision and inspection system for controlled persons based on big data control areas has certain technical defects. The first is the lack of intelligence. There is a lack of intelligent algorithms to accurately predict and evaluate key indicators such as the risk of re-violation of controlled persons, resulting in the inability to make targeted interventions after the controlled persons are released from supervision, and the risk of re-violation cannot be reduced. In addition, it is difficult to obtain the archival data of controlled persons, data inconsistency between different systems, and obstacles to sharing and synchronous updating of controlled persons' data between different supervision areas, resulting in inaccurate grasp of the comprehensive situation of controlled persons when they are released from supervision, and untimely updating of performance data, reward and punishment data, psychological assessment data, etc. of controlled persons during their stay in the supervision area, which may lead to the inability to accurately judge their current status when checking the supervision. To this end, we propose a supervision and inspection system for controlled persons based on big data control areas for supervision and inspection. Summary of the invention

[0004] The purpose of the present invention is to provide a supervision reform and control person release verification system based on big data control.

[0005] In order to solve the problems raised in the above background technology, the present invention provides the following technical solutions: a system for checking the release of controlled persons for supervision and transformation based on big data control, the system for checking the release of controlled persons for supervision and transformation based on big data control includes a data acquisition module, a data cleaning and verification module, a data integration module, a data analysis module, a risk assessment module, a real-time monitoring module, an early warning module and an intervention management module;

[0006] Data collection module: used to collect detailed information of controlled persons;

[0007] Data cleaning and verification module: cleans and verifies the collected data;

[0008] Data integration module: Integrate data from different systems and channels to establish a unified data format and storage structure;

[0009] Data analysis module: Use data analysis algorithms and models to conduct in-depth analysis of the integrated data and explore potential patterns and trends;

[0010] Risk assessment module: uses intelligent algorithms to calculate the risk factor of the regulated person being released from supervision;

[0011] Real-time monitoring module: real-time tracking of various situations of the regulated persons during their stay at the regulated location to ensure the timeliness of the data;

[0012] Early warning module: receives the results of the risk assessment module and issues different early warning instructions according to the different threshold ranges of the assessment results;

[0013] Intervention management module: formulate and implement different detailed intervention management measures according to the different warning instructions issued by the early warning module.

[0014] As a further solution of the present invention: the data acquisition module is able to interface with various management systems in the supervision area. The data acquisition module has two modes: manual entry and data entry. For data entry, it supports text, image, audio and video entry methods, uses natural language processing technology and pattern matching algorithm to identify text data, and extracts information from the archival text of the controlled person, including the name of the controlled person, ID number, type of violation, control period and details of the violation. When processing image data, with the help of image processing technology and feature extraction algorithm, the facial features and body posture information of the controlled person are identified. When processing audio data, the emotional characteristics and language expression pattern data of the controlled person in the psychological assessment process are extracted through voice recognition technology and sentiment analysis algorithm, and a data folder is established. Each file in the data folder is named in the format of "controlled person name + ID number".

[0015] As a further solution of the present invention: the data cleaning and verification module adopts a duplicate data detection algorithm based on machine learning, and identifies and deletes duplicate data records by performing similarity calculation and pattern matching on the key fields of the basic information of the controlled persons and the illegal behavior information, thereby avoiding data redundancy. For the correction of erroneous data, the basic information of the controlled persons is cross-verified with the authoritative database of the law enforcement department through real-time comparison and data verification technology with the authoritative database of the law enforcement department. Once a difference is found, it is corrected according to the authoritative data. For data fields that may be missing, data interpolation and prediction algorithms are used to mark and remind, and a data cleaning and verification log is generated for the cleaned and verified data. The data cleaning and verification module will transmit the processed data to the data integration module.

[0016] As a further solution of the present invention: the data integration module can integrate the data of various management systems within the supervision area, and the various management systems within the supervision area include the supervision area management system, education reform system, psychological assessment system, labor reform system, medical health system and regulatory review system. The data integration module obtains data from these data sources through ETL data extraction tools and ODBC interface technology, and then extracts, converts and loads data from different data sources. For the data conversion stage, the extracted data is format converted, field mapped and data standardized. For the data loading stage, a column storage structure and data loading algorithm are used to quickly and accurately load the processed data into the data warehouse, and a unified data dictionary and metadata management mechanism are established to clarify the definition, source, format and update frequency of the data. The data integration module interacts with the data analysis module in real time.

[0017] As a further solution of the present invention: the data analysis module can receive the data integrated by the data integration module. The data analysis module adopts the density-based DBSCAN clustering algorithm and the AGNES hierarchical clustering algorithm to classify the controlled persons with the same illegal behavior type characteristics into one category, thereby helping to summarize the common problems and potential illegal behavior patterns in the group of controlled persons. Then the data analysis module combines the ARIMA model of the time series analysis method with the historical behavior data and transformation performance data of the controlled persons to predict their future behavior trends and risk changes. After conducting in-depth data analysis, the data analysis module adopts the Hadoop distributed computing framework to help improve the efficiency of data analysis, and then intuitively displays the analysis results in the form of charts. The data analysis module will transmit the analyzed data to the risk assessment module.

[0018] As a further solution of the present invention: the risk assessment module is provided with a deep learning algorithm and a complex risk assessment model. Through training of a large amount of historical data, the trained model learns and captures the relationship between the characteristics of the controlled person and the risk of re-violation, and calculates the type of violation, the performance of the supervision place, the age of the controlled person and the psychological assessment score according to the received type of violation, the performance of the supervision place, the age of the controlled person and the psychological assessment score, and then calculates the risk coefficient of the controlled person's release from supervision based on the calculated data. The specific calculation formula of the risk coefficient of the controlled person's release from supervision is as follows:

[0019]

[0020] Where: R represents the risk factor of the regulated person being released from supervision, M, N, G and O n represents the number of supervisions, the number of criminal violations, the number of administrative violations recorded by the law enforcement system, and the current age of release from supervision of the supervised person, respectively. i , Z i , Y i and T i represents the i-th supervision performance score, violation type score, psychological assessment score and violation age score, respectively. i 、b i 、c i and d i Represents the X set by the administrator i , Z i , Y i and T i The weight coefficient, K represents the scaling factor;

[0021] The risk assessment module transmits the calculated risk coefficient data of the controlled person's release from supervision to the early warning module.

[0022] As a further solution of the present invention: the real-time monitoring module is connected to the real-time interface of each management system in the supervision area to realize seamless acquisition of real-time performance data of the supervised personnel, and the performance data includes daily behavior (work and rest patterns, interpersonal communication), learning status (course participation, test scores) and labor performance (quality and efficiency of work completion). The real-time data is processed and analyzed by using streaming computing technology and combined with a complex event processing (CEP) engine to timely discover abnormalities and potential risk trends, and set personalized violation count thresholds based on the individual characteristics and historical performance of the supervised personnel. In addition, the real-time monitoring module is provided with data storage and query functions, and the distributed storage system HBase is used to save real-time monitoring data for a period of time, which can facilitate management personnel to trace and analyze the monitoring data. The real-time monitoring module interacts with the early warning module in real time.

[0023] As a further solution of the present invention: The warning module can receive the result data from the risk assessment module and make judgments based on a preset strict threshold range. The specific judgment categories are as follows:

[0024] When R ≤ R min At this time, it is determined to be a risk-free state, and the warning module does not take any intervention measures;

[0025] When R min < R ≤ R max At this time, it is determined to be a low-risk state, and the warning module issues a low-risk Class A warning instruction;

[0026] When R max < R, at this time, it is determined to be a high-risk state, and the warning module issues a high-risk Class B warning instruction.

[0027] As a further solution of the present invention: The intervention management module can receive the instruction data transmitted by the warning module;

[0028] When the intervention management module does not receive the instruction data, a normal release process is given to the regulated person, and conventional tracking and support measures are taken. Regularly (once every quarter), communicate with the regulated person by phone and online to understand their living and working conditions, and provide necessary information and resources, including job opportunity recommendations and social welfare policy notifications. At the same time, notify the situation to their所在 community to ensure that the community has a basic understanding and attention to them;

[0029] When the intervention management module receives a low-risk Class A warning instruction, it will conduct a face-to-face follow-up visit to the released person who has been released from supervision once a month, deeply understand the situation of the regulated person in terms of social adaptation, career development, and interpersonal relationships, and provide vocational training subsidies to encourage them to participate in courses to improve their employment ability. Jointly carry out public welfare activities with the community to promote their integration into society, establish a psychological support file for them to record changes in their psychological state, and provide short-term psychological counseling services if necessary;

[0030] When the intervention management module receives a high-risk Class B warning instruction, it will arrange for a dedicated person to conduct on-site visits to the released person who has been released from supervision at least twice a week, closely monitor their behavior and social activities, force them to participate in long-term psychological treatment and behavior correction courses, and conduct regular evaluations. Cooperate with relevant enterprises to provide them with specific job positions, and strengthen work supervision. Establish a real-time information sharing mechanism with law enforcement departments to timely master their dynamics. Economically, provide a living allowance for a certain period to ensure their basic living needs, reduce the inducement of illegal acts caused by economic pressure, and详细 record the whole process and effect of the intervention, and adjust the intervention strategy at any time according to the actual situation to minimize the risk of re-offending and promote their stable return to society.

[0031] By adopting the above technical solution, compared with the prior art, the beneficial effects of the present invention are:

[0032] 1. The present invention can collect the supervised information and personal information of the supervised persons through the data collection module, and after cleaning, checking and integrating through the data cleaning and verification module and the data integration module, analyze and mine potential rules and trends through the data analysis module, and then calculate the supervision release risk coefficient of each individual supervised person according to the supervision data of each supervised person through the risk assessment module, compare the supervision release risk coefficient of each supervised person with the threshold of the early warning module, and then make different intervention and management measures according to different risk ranges, so as to realize accurate prediction and effective management of supervised persons, help supervised persons return to society better, and reduce the risk of re-offending;

[0033] 2. The present invention can integrate the data obtained from different data sources through the setting of the data integration module, and establish a unified data format and storage structure, so as to solve the data inconsistency between different systems and the obstacles of sharing and synchronous updating of the data of controlled persons between different supervision places, so as to accurately grasp the comprehensive situation of the controlled persons when releasing the supervision, timely update the performance data, reward and punishment data and psychological assessment data of the controlled persons during the supervision place, and ensure that the current status of the controlled persons can be accurately judged when the supervision is lifted, thereby providing a more reliable basis for the supervision and inspection of the controlled persons, helping to reduce the risk of re-offending and promoting the controlled persons to better return to society;

[0034] 3. Through in-depth analysis of the data of the controlled persons, the system of the present invention can comprehensively understand the illegal behavior history, psychological characteristics and various data points of the supervised performance of each controlled person. Based on these detailed information, the supervisory management department can formulate a tailor-made personalized transformation plan for the controlled persons. This personalized transformation plan can better meet the individual needs of the controlled persons and improve their enthusiasm and initiative to participate in the transformation. Accurate transformation measures can more effectively solve the problems of the controlled persons, help them better understand their mistakes, correct bad behaviors, and thus improve the transformation effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 Schematic diagram of the system flow in an embodiment of the present invention. DETAILED DESCRIPTION

[0036] The specific embodiments of the present invention will be further described below in conjunction with the accompanying drawings. It should be noted that the description of these embodiments is used to help understand the present invention, but does not constitute a limitation of the present invention.

[0037] In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0038] A system for checking the release of supervised persons based on the supervision and transformation of the supervised areas by big data. When the supervised persons are released from supervision, their performance in the supervised areas, transformation conditions, etc. need to be fully checked. However, the previous method of relying on paper archives and manual records is not only inefficient, but also prone to inaccurate and incomplete information, making the verification process difficult and the accuracy difficult to guarantee. At the same time, it is difficult for the traditional method to deeply analyze the behavior patterns and psychological states of the supervised persons, and it is difficult to effectively identify and intervene in advance for the supervised persons with the risk of re-offending. In order to be able to target the actual situation of each individual supervised person, the data of each supervised person being supervised is intelligently analyzed to determine the possibility of the supervised person committing another illegal act after being released from supervision, and to intervene and identify in advance, and take corresponding intervention management measures to avoid the released supervised persons from committing another illegal act and going to prison, and at the same time, it can effectively protect the public security.

[0039] The existing supervision and reform based on big data control area supervision and verification system for controlled persons has certain technical defects. First of all, the intelligence level is insufficient. There is a lack of intelligent algorithms to accurately predict and evaluate key indicators such as the risk of re-violation of controlled persons, resulting in the inability to make targeted interventions after the controlled persons are released from supervision, and the risk of re-violation cannot be reduced. In addition, it is difficult to obtain the archival data of controlled persons, and there are data inconsistencies between different systems. There are obstacles to sharing and synchronous updating of controlled persons' data between different supervision areas, resulting in inaccurate grasp of the comprehensive situation of controlled persons when they are released from supervision, and the performance data, reward and punishment data, psychological assessment data, etc. of controlled persons during their stay in the supervision area are not updated in a timely manner, which may lead to the inability to accurately judge their current status when checking the release of supervision;

[0040] Therefore, in order to effectively solve the above problems, please refer to the attached Figure 1 , the present application proposes a system for verifying the release of controlled persons for supervision and transformation based on big data control. The system for verifying the release of controlled persons for supervision and transformation based on big data control includes a data acquisition module, a data cleaning and verification module, a data integration module, a data analysis module, a risk assessment module, a real-time monitoring module, an early warning module and an intervention management module;

[0041] Data collection module: used to collect detailed information of controlled persons;

[0042] Data cleaning and verification module: cleans and verifies the collected data;

[0043] Data integration module: Integrate data from different systems and channels to establish a unified data format and storage structure;

[0044] Data analysis module: Use data analysis algorithms and models to conduct in-depth analysis of the integrated data and explore potential patterns and trends;

[0045] Risk assessment module: uses intelligent algorithms to calculate the risk factor of the regulated person being released from supervision;

[0046] Real-time monitoring module: real-time tracking of various situations of the regulated persons during their stay at the regulated location to ensure the timeliness of the data;

[0047] Early warning module: receives the results of the risk assessment module and issues different early warning instructions according to the different threshold ranges of the assessment results;

[0048] Intervention management module: formulate and implement different detailed intervention management measures according to the different warning instructions issued by the early warning module.

[0049] In one embodiment of the present invention: the data acquisition module is able to interface with various management systems within the supervision area. The data acquisition module has two modes: manual entry and data entry. For data entry, it supports text, image, audio and video entry methods. Natural language processing technology and pattern matching algorithm are used to identify text data, and information is extracted from the archival text of the controlled person, including the name of the controlled person, ID number, type of violation, control period and details of the violation. When processing image data, the facial features and body posture information of the controlled person are identified with the help of image processing technology and feature extraction algorithm. When processing audio data, the emotional characteristics and language expression pattern data of the controlled person in the psychological evaluation process are extracted through voice recognition technology and sentiment analysis algorithm, and a data folder is established. Each file in the data folder is named in the format of "controlled person name + ID number".

[0050] In one embodiment of the present invention: the data cleaning and verification module adopts a duplicate data detection algorithm based on machine learning, and identifies duplicate data records by performing similarity calculation and pattern matching on key fields of the basic information of the controlled persons and the illegal behavior information, and deletes them, thereby avoiding data redundancy. For the correction of erroneous data, the basic information of the controlled persons is cross-verified with the authoritative database of the law enforcement department through real-time comparison and data verification technology with the authoritative database of the law enforcement department. Once a difference is found, it is corrected according to the authoritative data. For data fields that may be missing, data interpolation and prediction algorithms are used to identify and remind, and a data cleaning and verification log is generated for the cleaned and verified data. The data cleaning and verification module will transmit the processed data to the data integration module.

[0051] In one embodiment of the present invention: the data integration module can integrate the data of various management systems within the supervision area, and the various management systems within the supervision area include the supervision area management system, education reform system, psychological assessment system, labor reform system, medical health system and regulatory review system. The data integration module obtains data from these data sources through ETL data extraction tools and ODBC interface technology, and then extracts, converts and loads data from different data sources. For the data conversion stage, the extracted data is format converted, field mapped and data standardized. For the data loading stage, a column storage structure and data loading algorithm are used to quickly and accurately load the processed data into the data warehouse, and a unified data dictionary and metadata management mechanism are established to clarify the definition, source, format and update frequency of the data. The data integration module interacts with the data analysis module in real time.

[0052] In one embodiment of the present invention: the data analysis module can receive the data integrated by the data integration module. The data analysis module adopts the density-based DBSCAN clustering algorithm and the AGNES hierarchical clustering algorithm to classify the controlled persons with the same illegal behavior type characteristics into one category, thereby helping to summarize the common problems and potential illegal behavior patterns in the group of controlled persons. Then the data analysis module combines the ARIMA model of the time series analysis method with the historical behavior data and transformation performance data of the controlled persons to predict their future behavior trends and risk changes. After conducting in-depth data analysis, the data analysis module adopts the Hadoop distributed computing framework to help improve the efficiency of data analysis, and then intuitively displays the analysis results in the form of charts. The data analysis module will transmit the analyzed data to the risk assessment module.

[0053] In one embodiment of the present invention: the risk assessment module is provided with a deep learning algorithm and a complex risk assessment model. By training a large amount of historical data, the trained model learns and captures the relationship between the characteristics of the controlled person and the risk of re-violation, and calculates the type of violation, the performance of the supervision place, the age of the controlled person and the psychological assessment score according to the received type of violation, the performance of the supervision place, the age of the controlled person and the psychological assessment score, and then calculates the risk coefficient of the controlled person's release from supervision based on the calculated data. The specific calculation formula of the risk coefficient of the controlled person's release from supervision is as follows:

[0054]

[0055] Where: R represents the risk factor of the regulated person being released from supervision, M, N, G and O n represents the number of supervisions, the number of criminal violations, the number of administrative violations recorded by the law enforcement system, and the current age of release from supervision of the supervised person, respectively.i , Z i , Y i and T i respectively represent the performance score of the supervision area, the illegal behavior type score, the psychological assessment score, and the illegal behavior age score in the i-th time. a i , b i , c i and d i respectively represent the weight coefficients of X i , Z i , Y i and T i set by the management personnel, and K represents the scaling coefficient;

[0056] The risk assessment module will transmit the calculated data of the risk coefficient for releasing the supervised person to the warning module.

[0057] In an embodiment of the present invention: The real-time monitoring module interfaces with the real-time interfaces of each management system within the supervision area to achieve seamless acquisition of the real-time performance data of the supervised person. The performance data includes daily behaviors (work and rest patterns, interpersonal communication), learning situations (course participation, exam scores), and labor performance (work completion quality, efficiency). Using stream computing technology and combining with a complex event processing (CEP) engine, the real-time data is processed and analyzed to timely detect anomalies and potential risk trends. According to the individual characteristics and historical performance of the supervised person, a personalized threshold for the number of violations is set. Moreover, the real-time monitoring module is provided with a data storage and query function, and uses the distributed storage system HBase to save the real-time monitoring data for a period of time, which is convenient for the management personnel to trace and analyze the monitoring data. The real-time monitoring module interacts with the warning module in real time.

[0058] In an embodiment of the present invention: The warning module can receive the result data from the risk assessment module and make a judgment based on a preset strict threshold range. The specific judgment categories are as follows:

[0059] When R ≤ R min , it is determined as a risk-free state at this time, and the warning module does not take any intervention measures;

[0060] When R min < R ≤ R max , it is determined as a low-risk state at this time, and the warning module issues a low-risk Class A warning instruction;

[0061] When R max < R, it is determined as a high-risk state at this time, and the warning module issues a high-risk Class B warning instruction;

[0062] The warning module will transmit the warning instruction data to the intervention management module.

[0063] In one embodiment of the present invention: the intervention management module is capable of receiving instruction data transmitted by the early warning module;

[0064] When the intervention management module does not receive instruction data, it will give the controlled person a normal release process, take regular tracking and support measures, communicate with the controlled person regularly (once a quarter) by phone and online to understand his or her living and working conditions, and provide necessary information and resources, including job opportunity recommendations and social welfare policy notifications. At the same time, it will inform the community where the controlled person lives to ensure that the community has a basic understanding and concern for him or her;

[0065] When the intervention management module receives a low-risk Class A warning instruction, it will conduct a monthly face-to-face return visit to the deregulated personnel to gain an in-depth understanding of the social adaptation, career development, and interpersonal relationships of the regulated personnel, provide vocational training subsidies, encourage them to participate in courses to enhance their employability, and jointly carry out public welfare activities with the community to promote their integration into society. It will establish a psychological support file for them, record changes in their psychological status, and provide short-term psychological counseling services when necessary;

[0066] When the intervention management module receives a high-risk Class B warning instruction, it will arrange for special personnel to conduct on-site visits to the released supervisors at least twice a week, closely monitor their behavior and social activities, and force them to participate in long-term psychotherapy and behavior correction courses, and conduct regular evaluations. It will also cooperate with relevant companies to provide them with specific jobs, strengthen work supervision, establish a real-time information sharing mechanism with law enforcement agencies, keep abreast of their dynamics, and provide a certain period of living allowance financially to ensure their basic living needs and reduce the incentives for illegal behavior caused by economic pressure. The module will also record the entire process and effects of the intervention in detail, adjust the intervention strategy at any time according to actual conditions, minimize the risk of further illegal behavior, and promote their stable return to society.

[0067] Example 1, please refer to the attached Figure 1For the data acquisition module, manual data entry will also be designed in the actual application process. For the manually entered data, strict format verification and logic checking mechanisms need to be adopted, and regular expressions and logic judgment functions need to be used. For example, check whether the entered control period is in line with the scope prescribed by the law, whether the age of the controlled person and the time of the illegal behavior are within a reasonable logical range, etc. At the same time, in order to ensure the security of data during transmission and personal privacy, the Advanced Encryption Standard (AES) encryption algorithm is used to encrypt the data. In addition, an efficient data caching mechanism can be set in the data acquisition module. When facing a network failure or excessive data volume causing transmission interruption, the cache database can be used to temporarily store the untransmitted data. After the network returns to normal, the data can be continued through the breakpoint resume technology to ensure the integrity and accuracy of the data, and there will be no data loss or repeated transmission problems.

[0068] Example 2, please refer to the attached Figure 1 ,For the data cleaning and verification module, in addition to using ,duplicate data detection algorithms based on machine learning, real-time comparison and data verification ,technology with the authoritative database of law enforcement agencies, and using ,data interpolation and prediction algorithms to handle missing data, in actual ,applications, abnormal data detection can also be performed by setting reasonable data ranges and ,thresholds to perform abnormal detection on the collected data. For example, for the age data of the ,controlled persons, if a value obviously exceeds the reasonable range appears, the system will mark it as abnormal data and ,perform further verification and processing;

[0069] Data consistency check: Check the consistency of data from different systems and channels to ensure that the information of the same regulated person is consistent in various data sources. For example, check the consistency of the basic identity information and violation records of the regulated person in different systems;

[0070] Integration with external data: In addition to comparison with the authoritative database of law enforcement agencies, other relevant external data, such as social credit data, employment data, etc., can also be integrated to obtain more comprehensive information on controlled persons and improve the accuracy of risk assessment.

[0071] Example 3, please refer to the attached Figure 1For the data integration module, in addition to obtaining data from the data source through ETL data extraction tools and ODBC interface technology, and performing extraction, conversion and loading operations, in actual applications, data fusion technology can also be used to fuse multimodal data (such as text, images, audio, etc.) from different data sources to obtain more comprehensive and accurate information about the controlled persons. For example, the archival text of the controlled persons can be associated with image, audio and other data to explore deeper information, and data mapping and matching rules can be established to ensure that data from different data sources can be accurately associated and integrated. For data quality management, data quality can be monitored and evaluated in real time during the data integration process. If data quality problems are found, they can be cleaned, corrected and marked in time to ensure that the quality of the integrated data is reliable. For the association with external systems, the actual process can further expand the integration with external systems, including data interaction and sharing with relevant systems such as the judicial department and the social security department, to obtain more background information and social support resources about the controlled persons, and provide a more comprehensive basis for the transformation of the controlled persons and the management after the release of supervision.

[0072] Example 4, please refer to the attached Figure 1 For the real-time monitoring module, the real-time monitoring module is not limited to the data on the daily behavior, learning situation and labor performance of the supervised persons, but can also be extended to multi-dimensional data monitoring such as the health status and social relationships of the supervised persons. For example, through integration with the medical and health system, the physical indicators and disease information of the supervised persons can be obtained in real time, and potential health problems can be discovered in time. The abnormal behavior of the supervised persons can be obtained through video surveillance analysis, and the abnormal behavior and gathering situation of the supervised persons can be automatically identified by computer vision technology. The activity trajectory of the supervised persons can be monitored in real time through face recognition technology, and an alarm can be automatically triggered when abnormal gathering or violation behavior is found. In addition, the voice data of the supervised persons needs to be emotionally analyzed to understand their emotional state and psychological changes. The voice emotion recognition technology can be used to timely discover the possible emotional problems of the supervised persons so as to take corresponding psychological intervention measures.

[0073] Specifically, the present invention can collect the supervised information and personal information of the supervised persons with the help of the data acquisition module. After cleaning, verification and integration by the data cleaning and verification module and the data integration module, the data analysis module analyzes and mines the potential rules and trends. The risk assessment module then calculates the risk coefficient of release from supervision based on the supervised data of each supervised person, compares the coefficient with the threshold of the early warning module, and then implements different intervention and management measures according to different risk ranges, so as to achieve accurate prediction and effective management of the supervised persons, help the supervised persons return to society better, and reduce the risk of further violations.

[0074] Specifically, the present invention, by setting up a data integration module, can integrate the data obtained from different data sources, establish a unified data format and storage structure, solve the problem of data inconsistency between different systems and data sharing and synchronous updating of controlled persons between different supervision locations, ensure that the comprehensive situation of the controlled persons can be accurately grasped when the supervision is lifted, and timely update the performance data, reward and punishment data and psychological assessment data of the controlled persons during the supervision location, to ensure that their current status can be accurately judged when the supervision is lifted, thereby providing a more reliable basis for the supervision lift inspection of the controlled persons, which is conducive to reducing the risk of re-offending and promoting the controlled persons to better return to society.

[0075] Specifically, through in-depth analysis of the data of controlled persons, the system can comprehensively grasp the key data points of each controlled person's illegal behavior history, psychological characteristics, and regulatory performance. Based on these detailed information, the regulatory management department can formulate a tailor-made personalized reform plan for the controlled persons. This personalized reform plan can better meet the individual needs of the controlled persons and enhance their enthusiasm and initiative to participate in the reform. Accurate reform measures can more effectively solve the problems of the controlled persons, help them better recognize their own mistakes, correct bad behaviors, and thus improve the reform effect.

[0076] Working principle:

[0077] First, the data collection module can collect the supervised information and personal information of the supervised persons, and after cleaning, checking and integrating them through the data cleaning and verification module and the data integration module, the data analysis module analyzes and mines the potential rules and trends, and then uses the deep learning neural network in the risk assessment module to calculate and analyze the type of illegal behavior, supervision performance, age of the supervised persons and psychological assessment scores, and then uses the formula The risk coefficient of release from supervision of each individual controlled person is calculated, and then the early warning module can receive the data from the risk assessment module, and through interaction with the real-time monitoring module, obtain various data of the controlled persons in the supervision area in real time, and use this to determine the threshold size of the early warning module. The early warning module will then compare the received risk coefficient of release from supervision of the controlled persons with the threshold value therein, so as to determine the range of different categories, and issue different early warning instructions according to different ranges. Then the intervention management module can make different and detailed intervention management measures according to the received early warning instructions, help to carry out certain supervision on the released persons, realize accurate prediction and effective management of the controlled persons, help the controlled persons to better return to society, and reduce the risk of further violations. At this point, the entire workflow is completed.

[0078] At the same time, the present application uses specific words to describe the embodiments of the present application. For example, "one embodiment", "an embodiment", and / or "some embodiments" refer to a certain feature, structure or characteristic related to at least one embodiment of the present application. Therefore, it should be emphasized and noted that "one embodiment" or "an embodiment" or "an alternative embodiment" mentioned twice or more in different positions in this specification does not necessarily refer to the same embodiment. In addition, certain features, structures or characteristics in one or more embodiments of the present application can be appropriately combined.

[0079] Some aspects of the present application may be performed entirely by hardware, entirely by software (including firmware, resident software, microcode, etc.), or by a combination of hardware and software. The above hardware or software may be referred to as "data blocks", "modules", "engines", "units", "components" or "systems". The processor may be one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DAPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), processors, controllers, microcontrollers, microprocessors, or combinations thereof. In addition, various aspects of the present application may be expressed as computer products located in one or more computer-readable media, which include computer-readable program codes. For example, computer-readable media may include, but are not limited to, magnetic storage devices (e.g., hard disks, floppy disks, tapes ...), optical disks (e.g., compact disks CDs, digital versatile disks DVDs ...), smart cards, and flash memory devices (e.g., cards, sticks, key drives ...).

[0080] A computer-readable medium may include a propagated data signal containing computer program code, such as in baseband or as part of a carrier wave. The propagated signal may have a variety of manifestations, including electromagnetic, optical, etc., or a suitable combination. A computer-readable medium may be any computer-readable medium other than a computer-readable storage medium, which may be connected to an instruction execution system, device or apparatus to communicate, propagate or transmit a program for use. The program code on the computer-readable medium may be propagated through any suitable medium, including radio, cable, fiber optic cable, radio frequency signal, or similar medium, or a combination of any of the above mediums.

[0081] Similarly, it should be noted that in order to simplify the description disclosed in this application, so as to help understand one or more embodiments of the invention, in the above description of the embodiments of the present application, sometimes multiple features are merged into one embodiment, drawings or descriptions thereof. However, this disclosure method does not mean that the features required by the object of the present application are more than the features mentioned in the claims. In fact, the features of the embodiment are less than all the features of the single embodiment disclosed above. In some embodiments, numbers describing the number of components and attributes are used. It should be understood that such numbers used in the description of the embodiments are modified by the modifiers "approximately", "approximately" or "substantially" in some examples. Unless otherwise specified, "approximately", "approximately" or "substantially" indicate that the number allows a variation of ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may change according to the characteristics required by individual embodiments. In some embodiments, the numerical parameters should take into account the specified significant digits and adopt the general method of retaining the digits. Although the numerical domains and parameters used to confirm the breadth of their range in some embodiments of the present application are approximate values, in specific embodiments, such numerical settings are as accurate as possible within the feasible range.

[0082] Although the present invention is disclosed as above in terms of preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art may make possible changes and modifications without departing from the spirit and scope of the present invention. Therefore, any modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention without departing from the content of the technical solution of the present invention shall fall within the scope of protection defined by the claims of the present invention.

Claims

1. A system for verifying the removal of supervision of controlled persons for supervision and transformation based on big data control. Features: The control-based supervision and transformation control personnel release and verification system includes a data collection module, a data cleaning and verification module, a data integration module, a data analysis module, a risk assessment module, a real-time monitoring module, an early warning module, and an intervention management module; The data collection module is used to collect detailed information data of the controlled persons, and the collected detailed information data of the controlled persons is cleaned and verified by the data cleaning and verification module. The data integration module can integrate the cleaned and verified data, obtain data from these data sources through ETL data extraction tools and ODBC interface technology, and then extract, convert and load data from different data sources, and establish a unified data format and storage structure. The data analysis module can analyze the data integrated by the data integration module, and use the density-based DBSCAN clustering algorithm and AGNES hierarchical clustering algorithm to classify the controlled persons with the same illegal behavior type characteristics into one category to help dig out potential rules and trends. The risk assessment module can analyze the illegal behavior type, supervision performance, age of the controlled persons and psychological assessment data of the controlled persons through a deep learning model, calculate their scores based on these data, and then use the risk coefficient formula for the release of supervision of the controlled persons: The risk coefficient of release of supervision of each individual controlled person is calculated. The real-time monitoring module can track the various situations of the controlled person during the supervision period in real time to ensure the timeliness of the data, and use the early warning module to determine the threshold value based on the acquired data. Then, the early warning module compares the risk coefficient of release of supervision of each individual controlled person with the range to which it belongs, and then judges different risk states, and issues different early warning instructions according to different risk states. The intervention management module will implement different detailed intervention management measures according to the tricky instructions.

2. According to the big data control-based supervision and transformation control personnel release verification system according to claim 1, Features: The data acquisition module can interface with various management systems in the supervision area. The data acquisition module has two modes: manual entry and data entry. For data entry, it supports text, image, audio and video entry methods, uses natural language processing technology and pattern matching algorithm to identify text data, and extracts information from the archival text of the controlled person, including the name of the controlled person, ID number, type of violation, control period and details of the violation. When processing image data, with the help of image processing technology and feature extraction algorithm, the facial features and body posture information of the controlled person are identified. When processing audio data, the emotional characteristics and language expression pattern data of the controlled person in the psychological evaluation process are extracted through voice recognition technology and sentiment analysis algorithm, and a data folder is established. Each file in the data folder is named in the format of "controlled person name + ID number".

3. According to the big data control-based supervision and transformation control personnel release verification system of claim 1, Features: The data cleaning and verification module adopts a duplicate data detection algorithm based on machine learning. By performing similarity calculation and pattern matching on the basic information of the controlled persons and the key fields of the illegal behavior information, duplicate data records are identified and deleted, thereby avoiding data redundancy. For the correction of erroneous data, the basic information of the controlled persons is cross-verified with the authoritative database of the law enforcement department through real-time comparison and data verification technology with the authoritative database of the law enforcement department. Once a difference is found, it is corrected based on the authoritative data. For data fields that may be missing, data interpolation and prediction algorithms are used to mark and remind, and a data cleaning and verification log is generated for the cleaned and verified data. The data cleaning and verification module will transmit the processed data to the data integration module.

4. According to the big data control-based supervision and transformation control personnel release and verification system according to claim 1, Features: The data integration module can integrate the data of various management systems within the supervision area, which include the supervision area management system, education reform system, psychological assessment system, labor reform system, medical health system and regulatory review system. The data integration module obtains data from these data sources through ETL data extraction tools and ODBC interface technology, and then extracts, converts and loads data from different data sources. For the data conversion stage, the extracted data is format converted, field mapped and data standardized. For the data loading stage, a column storage structure and data loading algorithm are used to quickly and accurately load the processed data into the data warehouse, and a unified data dictionary and metadata management mechanism are established to clarify the definition, source, format and update frequency of the data. The data integration module interacts with the data analysis module in real time.

5. According to the big data control-based supervision and transformation control personnel release and verification system according to claim 1, Features: The data analysis module can receive the data integrated by the data integration module. The data analysis module adopts the density-based DBSCAN clustering algorithm and the AGNES hierarchical clustering algorithm to classify the controlled persons with the same illegal behavior type characteristics into one category, thereby helping to summarize the common problems and potential illegal behavior patterns in the controlled person group. Then the data analysis module combines the ARIMA model of the time series analysis method with the historical behavior data and transformation performance data of the controlled persons to predict their future behavior trends and risk changes. After conducting in-depth data analysis, the data analysis module adopts the Hadoop distributed computing framework to help improve the efficiency of data analysis, and then intuitively displays the analysis results in the form of charts. The data analysis module will transmit the analyzed data to the risk assessment module.

6. According to the big data control-based supervision and transformation control personnel release and verification system according to claim 1, Features: The risk assessment module is provided with a deep learning algorithm and a complex risk assessment model. Through training of a large amount of historical data, the trained model learns and captures the relationship between the characteristics of the controlled person and the risk of re-offending, and calculates the type of illegal behavior, the performance of the supervision place, the age of the controlled person and the psychological assessment score according to the received illegal behavior type, the performance of the supervision place, the age of the controlled person and the psychological assessment score, and then calculates the risk coefficient of the controlled person's release from supervision based on the calculated data; Where: R represents the risk factor of the regulated person being released from supervision, M, N, G and O n represents the number of supervisions, the number of criminal violations, the number of administrative violations recorded by the law enforcement system, and the current age of release from supervision of the supervised person, respectively. i , Z i , Y i and T i represents the i-th supervision performance score, violation type score, psychological assessment score and violation age score, respectively. i 、b i 、c i and d i Represents the X set by the administrator i , Z i , Y i and T i The weight coefficient, K represents the scaling factor; The risk assessment module transmits the calculated risk coefficient data of the controlled person's release from supervision to the early warning module.

7. According to the big data control-based supervision and transformation control personnel release and verification system according to claim 1, Features: The real-time monitoring module is connected to the real-time interface of each management system in the supervision area to achieve seamless acquisition of real-time performance data of the supervised persons. The performance data includes daily behavior (work and rest patterns, interpersonal communication), learning status (course participation, test scores) and labor performance (quality and efficiency of work completion). The real-time data is processed and analyzed by using streaming computing technology and combined with a complex event processing (CEP) engine to promptly detect abnormalities and potential risk trends, and set personalized violation count thresholds based on the individual characteristics and historical performance of the supervised persons. In addition, the real-time monitoring module is provided with data storage and query functions, and uses the distributed storage system HBase to save real-time monitoring data for a period of time, which can facilitate management personnel to trace and analyze the monitoring data. The real-time monitoring module interacts with the early warning module in real time.

8. According to claim 7, a system for verifying the removal of controlled persons from supervision for supervision and transformation based on big data control, Features: The early warning module can receive the result data from the risk assessment module and make judgments based on the preset strict threshold range. The specific judgment categories are as follows: When R≤R min When , it is judged as a risk-free state, and the early warning module does not take any intervention measures; When R min <R≤R max When the warning module is in low risk state, it issues a low risk Class A warning instruction. When R max < is less than R, it is determined as a high-risk state at this time, and the early warning module issues a high-risk Class B early warning instruction; The early warning module transmits the early warning instruction data to the intervention management module.

9. According to claim 8, a system for verifying the removal of controlled persons from supervision for supervision and transformation based on big data control, Features: The intervention management module is capable of receiving instruction data transmitted by the early warning module; When the intervention management module does not receive instruction data, it will give the controlled person a normal release process, take regular tracking and support measures, communicate with the controlled person regularly (once a quarter) by phone and online to understand his or her living and working conditions, and provide necessary information and resources, including job opportunity recommendations and social welfare policy notifications. At the same time, it will inform the community where the controlled person lives to ensure that the community has a basic understanding and concern for him or her; When the intervention management module receives a low-risk Class A warning instruction, it will conduct a monthly face-to-face return visit to the deregulated personnel to gain an in-depth understanding of the social adaptation, career development, and interpersonal relationships of the regulated personnel, provide vocational training subsidies, encourage them to participate in courses to enhance their employability, and jointly carry out public welfare activities with the community to promote their integration into society. It will establish a psychological support file for them, record changes in their psychological status, and provide short-term psychological counseling services when necessary; When the intervention management module receives a high-risk Class B warning instruction, it will arrange for special personnel to conduct on-site visits to the released supervisors at least twice a week, closely monitor their behavior and social activities, and force them to participate in long-term psychotherapy and behavior correction courses, and conduct regular evaluations. It will also cooperate with relevant companies to provide them with specific jobs, strengthen work supervision, establish a real-time information sharing mechanism with law enforcement agencies, keep abreast of their dynamics, and provide a certain period of living allowance financially to ensure their basic living needs and reduce the incentives for illegal behavior caused by economic pressure. The module will also record the entire process and effects of the intervention in detail, adjust the intervention strategy at any time according to actual conditions, minimize the risk of further illegal behavior, and promote their stable return to society.

Images