Target object transformation difficulty level prediction method, system and equipment and storage medium

Abstract

The invention provides a target object reconstruction difficulty level prediction method, system and device and a storage medium, and relates to the technical field of information processing, the method comprises the following steps: constructing an initial prediction factor database based on psychological assessment data and target object static information; constructing a machine learning algorithm model, and training the machine learning algorithm model according to the initial prediction factor database to obtain an initial prediction model; performing importance sorting on the predictive factors in the initial predictive factor database, and obtaining a middle-term predictive factor database according to a sorting result in combination with the dynamic information of the target object; training the initial prediction model according to the mid-term prediction factor database to obtain a target prediction model; and obtaining feature data of the target object, and inputting the feature data into the target prediction model to obtain a transformation difficulty prediction result. According to the method, the transformation difficulty degree of the target object can be objectively and comprehensively judged, and theoretical support, basis and reference are provided for transformation management and prevention and control work of the target object.

Description

Technical Field

[0001] The present invention relates to the field of information processing technology, and in particular to a method, system, device and storage medium for predicting the difficulty of transformation of a target object. Background Art

[0002] With the rapid development of information technology, informatization has penetrated into every corner of social life. The education and transformation of prison target objects can also be combined with advanced information technology. The purpose of transforming target objects is to achieve the reintegration of target objects into society and become law-abiding citizens after their release through a variety of correctional and training methods. With the close connection between information technology and prison management, key information such as the basic file information, psychological assessment status and transformation assessment status of target objects has been digitally managed, which has greatly improved the efficiency and accuracy of prison management.

[0003] However, in practical work, traditional evaluation of the quality of target object transformation still mostly adopts subjective evaluation or interview methods. The difficulty of transformation of target objects is mainly judged based on the work experience of prison districts and responsible police officers. There is a lack of an objective and comprehensive method to effectively judge the difficulty of transformation of target objects. Summary of the invention

[0004] In view of the above-mentioned deficiencies in the prior art, the present invention provides a method, system, device and storage medium for predicting the difficulty of transformation of a target object, which effectively solves the problem that the difficulty of transformation of a prison target object is mainly judged based on the work experience of the prison area and the responsible police officers.

[0005] In a first aspect, the present invention provides a method for predicting the degree of difficulty of transformation of a target object, the method comprising: Construct an initial predictor database based on psychological assessment data and static information of target subjects; Constructing a machine learning algorithm model, and training the machine learning algorithm model according to the initial prediction factor database to obtain an initial prediction model; Sorting the prediction factors in the initial prediction factor database by importance, and obtaining a mid-term prediction factor database based on the sorting results and the target object dynamic information; Training the initial prediction model according to the mid-term prediction factor database to obtain a target prediction model; Acquire characteristic data of the target object, input the characteristic data into the target prediction model, and obtain a transformation difficulty prediction result through the target prediction model.

[0006] Furthermore, the construction of an initial prediction factor database based on psychological assessment data and static information of target objects includes: Acquiring the psychological assessment data of the target object according to the psychological assessment document; Acquire static information of the target object from a target object management system; The initial prediction factor database is constructed according to the psychological assessment data and the target object static information.

[0007] Furthermore, the constructing of the machine learning algorithm model, training the machine learning algorithm model according to the initial prediction factor database, and obtaining the initial prediction model includes: Extracting target object data from the initial prediction factor database, and merging the target object data according to the target object numbers to obtain merged data; Calculate the total score of the psychological dimension according to the psychological assessment data in the merged data, classify the total score of the dimension, and obtain a prediction factor and a prediction label; Dividing the merged data into a training set and a test set according to the predicted labels; Construct a machine learning algorithm model, use the training set to train the machine learning algorithm model, and use the test set to test the trained machine learning algorithm model to obtain the initial prediction model.

[0008] Furthermore, the calculating the total score of the psychological dimension according to the psychological assessment data in the merged data and classifying the total score of the dimension includes: Obtaining the psychological assessment data of all the target objects; Calculating the evaluation result score corresponding to each psychological dimension in the psychological evaluation data, performing weighted addition on each of the evaluation result scores, and obtaining the dimension total score of each of the psychological dimensions; A classification threshold is set, and the psychological dimension is classified according to the classification threshold and the total score of the dimension.

[0009] Furthermore, the ranking of the prediction factors in the initial prediction factor database by importance includes: Selecting any of the predictive factors in the initial predictive factor database, randomly modifying the characteristic value of the predictive factor, and obtaining a modified data set; Inputting the modified data set into the initial prediction model for prediction to obtain a modified prediction value; Calculate the impact value of each of the prediction factors on the loss function after modification according to the modified prediction value and the actual prediction value of the actual data set; The sizes of the influence values ​​are compared, and the prediction factors are ranked in order of importance.

[0010] Furthermore, the training of the initial prediction model according to the mid-term prediction factor database to obtain a target prediction model includes: Dividing the data in the medium-term predictor database into a training set and a test set; In the training set, the predictive factors in the mid-term predictive factor database are used as features, and the degree of difficulty of transformation of the target object is used as a label to train the initial prediction model; The hyperparameters of the initial prediction model are adjusted by multi-fold crossover and Bayesian search to obtain an initial prediction model with an optimal area under the curve; A threshold of the initial prediction model is determined, and the initial prediction model is verified using the test set to obtain the target prediction model.

[0011] Further, the using the test set to verify the initial prediction model includes: Inputting the test set into the initial prediction model, calculating the evaluation index of the initial prediction model, wherein the evaluation index at least includes recall rate, area under the curve and accuracy; The performance of the initial prediction model is verified according to the evaluation index.

[0012] In a second aspect, the present invention provides a target object transformation difficulty prediction system, the system comprising: A data processing module is used to construct an initial predictor database based on psychological assessment data and static information of target subjects; A model building module, used to build a machine learning algorithm model, train the machine learning algorithm model according to the initial prediction factor database, and obtain an initial prediction model; A factor ranking module, used to rank the prediction factors in the initial prediction factor database according to their importance, and obtain a mid-term prediction factor database based on the ranking results combined with the target object dynamic information; A model training module, used for training the initial prediction model according to the mid-term prediction factor database to obtain a target prediction model; The result prediction module is used to obtain the characteristic data of the target object, input the characteristic data into the target prediction model, and obtain the transformation difficulty prediction result through the target prediction model.

[0013] In a third aspect, the present invention provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method for predicting the difficulty of target object transformation as described in the first aspect of the present invention.

[0014] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for predicting the difficulty of transformation of a target object as described in the first aspect of the present invention.

[0015] The present invention provides a method, system, device and storage medium for predicting the degree of difficulty of transformation of a target object. Based on the psychological evaluation data of the target object, a prediction factor database is constructed in combination with the static information and dynamic information of the target object, and the prediction factors are ranked in importance, which can reduce the computational complexity and model complexity. The target prediction model is obtained by training according to the prediction factor database, and the target prediction model is continuously updated and optimized through the dynamic information of the target object in the prison, thereby improving the accuracy and effectiveness of the model prediction. The present invention can objectively and comprehensively judge the degree of difficulty of transformation of the prison target object, and provide theoretical support, basis and reference for the transformation management and prevention and control of the prison target object. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0017] Figure 1 This is a first schematic diagram of a method for predicting the degree of difficulty of transformation of a target object provided by an embodiment of the present invention; Figure 2 This is a second schematic diagram of the process of the method for predicting the degree of difficulty of transformation of a target object provided by an embodiment of the present invention; Figure 3 This is a third schematic diagram of the process of the method for predicting the degree of difficulty of transformation of a target object provided by an embodiment of the present invention; Figure 4 This is a fourth schematic diagram of the process of the method for predicting the degree of difficulty of transformation of a target object provided by an embodiment of the present invention; Figure 5 This is a fifth schematic diagram of a method for predicting the degree of difficulty of transformation of a target object provided by an embodiment of the present invention; Figure 6 This is a sixth schematic diagram of a method for predicting the degree of difficulty of transformation of a target object provided by an embodiment of the present invention; Figure 7 is a schematic diagram of the structure of a target object transformation difficulty prediction system provided by an embodiment of the present invention; Figure 8 It is a structural schematic diagram of an electronic device provided by an embodiment of the present invention.

[0018] Description of main component symbols: 700. Target object transformation difficulty prediction system; 710. Data processing module; 720. Model building module; 730. Factor sorting module; 740. Model training module; 750. Result prediction module; 800. Electronic device; 810. Processor; 820. Communication interface; 830. Memory; 840. Communication bus. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be further clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. It should be noted that the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0020] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0022] The purpose of reforming the target object is to achieve the target object's reintegration into society after being released from prison through a variety of correctional and training methods, and to become a law-abiding citizen. With the close connection between information technology and prison management, key information such as the target object's basic file information, psychological assessment status, and reform assessment status have been digitally managed. However, in practical work, traditional evaluation of the quality of target object reform still mostly adopts subjective evaluation or interview methods. The difficulty of reforming the target object is mainly judged by the work experience of the prison area and the responsible police officers. There is a lack of an objective and comprehensive method to effectively judge the difficulty of reforming the target object.

[0023] Example 1 The embodiment of the present invention provides a method for predicting the difficulty of reforming a target object, which effectively solves the problem that the difficulty of reforming a prison target object is mainly judged based on the work experience of the prison area and the responsible police officers. Figure 1FIG. 1 is a first schematic diagram of a method for predicting the degree of difficulty of transformation of a target object provided by an embodiment of the present invention. Figure 1 As shown, the method comprises the following steps: S100: construct an initial prediction factor database based on psychological assessment data and static information of the target object.

[0024] Figure 2 2 is a second schematic diagram of the target object transformation difficulty prediction method provided by an embodiment of the present invention. Figure 2 As shown, constructing the initial predictor database includes the following steps: S110. Obtain psychological assessment data of the target object according to the psychological assessment document.

[0025] In the embodiment of the present invention, the psychological assessment document includes but is not limited to measurement questionnaires such as personality disorder diagnostic questionnaire, mental illness screening questionnaire, anxiety and depression scale, stress perception scale and childhood trauma experience questionnaire. Among them: The Personality Disorder Diagnostic Questionnaire (PDQ4+) is a personality disorder screening tool compiled based on the DSM-IV personality disorder diagnostic criteria, which can be used for self-examination or group screening. The PDQ4+ personality disorder test includes 12 types of personality disorders, namely paranoid, dissociative, schizoid, antisocial, borderline, histrionic, narcissistic, avoidant, dependent, obsessive-compulsive, passive-aggressive and depressive.

[0026] The Psychosis Screening Questionnaire (PSQ) is a scale used to assess whether an individual is at risk for mental illness. The questionnaire includes a series of questions involving the patient's mood, behavior, cognitive function, etc., in order to determine whether the patient is likely to have mental illnesses such as anxiety, depression, and schizophrenia.

[0027] The HAD scale is a commonly used scale to assess anxiety and depression symptoms. It contains a series of questions for assessing anxiety and depression and is often used in clinical and research settings to help assess a patient's mental health status.

[0028] The Perceived Stress Scale (CPSS) is a tool used to assess the level of stress an individual has felt in the past month. The questionnaire contains 10 items designed to assess an individual's perception and reaction to life events. These items cover a variety of factors that may cause stress, such as work, interpersonal relationships, and health.

[0029] The Childhood Traumatic Experiences Questionnaire (CTQ) is a commonly used self-report instrument used to assess various traumatic events such as depression, anxiety, and parenting styles that individuals may have experienced during childhood.

[0030] In the embodiment of the present invention, paper questionnaires can be distributed to the target subjects for filling out and collected, or questionnaires can be created using an online platform and shared with the target subjects through a link or a QR code. In order to ensure the integrity and accuracy of the data, invalid data can be reduced by setting mandatory questions and logical checks, and then scoring is performed according to the evaluation results of each psychological evaluation document to obtain psychological evaluation data.

[0031] S120: Obtain static information of the target object from the target object management system.

[0032] In the embodiment of the present invention, the static information of the target object can be directly obtained from the target object management system of the prison. The static information of the target object is the basic information of the target object, including but not limited to age, education level, marital status, sentence, crime type and other information.

[0033] S130. Construct an initial prediction factor database based on the psychological assessment data and the static information of the target object.

[0034] First, in order to ensure the comparability of psychological assessment data and static information of target objects of different dimensions and ranges, data standardization or normalization is required to eliminate the impact of different data units or magnitudes on the analysis results. Then, the table structure and fields of the initial predictor database are designed to ensure that each data item has a clear definition and coding method, and the psychological assessment data and static information of the target objects after data standardization or normalization are integrated according to the designed database structure.

[0035] S200, constructing a machine learning algorithm model, training the machine learning algorithm model according to an initial prediction factor database, and obtaining an initial prediction model.

[0036] Figure 3 FIG. 3 is a third schematic diagram of the process of the method for predicting the degree of difficulty of transformation of a target object provided by an embodiment of the present invention. Figure 3 As shown, the acquisition of the initial prediction model includes the following steps: S210 , extracting target object data from the initial prediction factor database, and merging the target object data according to the target object numbers to obtain merged data.

[0037] Optionally, first determine the unique number field used to identify each target object in the initial predictor database, such as the target object ID and user ID fields. Then extract all data related to the target object from the initial predictor database, select records containing the target object ID and related attributes (such as psychological assessment data and target object static information), and finally sort and merge the data according to the target object ID.

[0038] S220. Calculate the total score of the psychological dimension according to the psychological assessment data in the merged data, classify the psychological dimension according to the total score, and obtain prediction factors and prediction labels.

[0039] Figure 4 4 is a schematic diagram of the target object transformation difficulty prediction method provided by the embodiment of the present invention. Figure 4 As shown, the classification of psychological dimensions includes the following steps: S221. Obtain psychological assessment data of all target objects.

[0040] S222. Calculate the evaluation result score corresponding to each psychological dimension in the psychological evaluation data, perform weighted addition on each evaluation result score, and obtain the total score of each psychological dimension.

[0041] In the embodiment of the present invention, according to the content of the psychological assessment document, the scores of the assessment results of each psychological dimension are weighted and added to obtain the dimension total score of each psychological dimension. The dimension total score represents the performance of the target object in a specific psychological or behavioral characteristic. When calculating the dimension total score, if the assessment result score of a certain question of a target object is missing, the average score of other target objects in the prison is used to fill it.

[0042] S223. Set a classification threshold, and classify the dimension total score according to the classification threshold.

[0043] Set the classification threshold for the total score of each dimension. These classification thresholds can be determined based on domain knowledge, previous research, expert opinions, or data distribution characteristics. Use the selected classification threshold to classify the dimension total score, convert the continuous field into a categorical field, and finally obtain the predictor and prediction label.

[0044] S230 , dividing the merged data into a training set and a test set according to the predicted labels.

[0045] In the embodiment of the present invention, according to the ratio of predicted labels, 80% of the merged data is used as a training set and 20% of the merged data is used as a test set.

[0046] S240, construct a machine learning algorithm model, train the machine learning algorithm model using a training set, and test the trained machine learning algorithm model using a test set to obtain an initial prediction model.

[0047] Optionally, the machine learning algorithm model can adopt classification prediction models such as the lightweight gradient boosting decision algorithm model LightGBM and the support vector machine algorithm model SVM. In an embodiment of the present invention, a LightGBM model is constructed for classification prediction, the constructed LightGBM model is trained using a training set, and the trained LightGBM model is tested using a test set to obtain an initial prediction model.

[0048] S300, ranking the prediction factors in the initial prediction factor database by importance, and obtaining a mid-term prediction factor database based on the ranking results and the target object dynamic information.

[0049] Figure 5 FIG. 5 is a fifth schematic diagram of a method for predicting the degree of difficulty of transformation of a target object provided by an embodiment of the present invention. Figure 5 As shown in Figure 1, ranking the importance of predictors includes the following steps: S310 , selecting any prediction factor in the initial prediction factor database, randomly modifying the characteristic value of the prediction factor, and obtaining a modified data set.

[0050] S320: Input the modified data set into the initial prediction model for prediction to obtain a modified prediction value.

[0051] S330. Calculate the impact value of each prediction factor on the loss function after modification according to the modified prediction value and the actual prediction value of the actual data set.

[0052] S340. Compare the sizes of the various impact values ​​and rank the prediction factors in order of importance.

[0053] The purpose of ranking the importance of predictors is to quantify the contribution of each predictor in model prediction. In the embodiment of the present invention, the ranking is analyzed by measuring the change in the loss function when the characteristic value of the predictor is randomly modified or disturbed. The loss function is a function that measures the difference between the model prediction value and the true value. When a predictor has a significant impact on the prediction result, modifying the characteristic value of the predictor will cause a large change in the loss function. Ranking the importance of predictors can identify which predictors are more critical in the prediction task, so as to make more reasonable decisions in practical applications.

[0054] In an embodiment of the present invention, according to the sorting results combined with subjective experience in previous work, psychological assessment questions are screened to obtain new psychological assessment data, and a medium-term prediction factor database is obtained in combination with the dynamic information of the target object. The new psychological assessment data and the dynamic information of the target object are subjected to data standardization or normalization processing, thereby eliminating the influence on the analysis results caused by different data units or magnitudes. Then, the table structure and fields of the medium-term prediction factor database are designed to ensure that each data item has a clear definition and coding method, and the new psychological assessment data and the dynamic information of the target object after data standardization or normalization processing are integrated according to the designed database structure. The dynamic information of the target object includes but is not limited to information such as other target object support, disciplinary support, and points added or deducted in prison. In an embodiment of the present invention, other target object support and disciplinary support can be quantified by evaluation grades or evaluation scores.

[0055] S400: Train an initial prediction model according to a mid-term prediction factor database to obtain a target prediction model.

[0056] Figure 6 6 is a schematic diagram of a method for predicting the degree of difficulty of transformation of a target object provided by an embodiment of the present invention. Figure 6 As shown, the acquisition of the target prediction model includes the following steps: S410, dividing the data in the medium-term prediction factor database into a training set and a test set.

[0057] S420, using the prediction factors in the mid-term prediction factor database as features in the training set and the degree of difficulty of transformation of the target object as a label to train an initial prediction model; S430, using multi-fold crossover and Bayesian search to adjust the hyperparameters of the initial prediction model to obtain an initial prediction model with an optimal area under the curve; S440: Determine a threshold of the initial prediction model, verify the initial prediction model using a test set, and obtain a target prediction model.

[0058] In an embodiment of the present invention, the threshold of the initial prediction model can be determined based on empirical data and model conditions, the test set is input into the initial prediction model, and the evaluation index of the initial prediction model is calculated. The evaluation index includes but is not limited to indicators such as recall rate, area under the curve and accuracy. The performance of the initial prediction model is verified based on these evaluation indicators, so as to obtain the optimal target prediction model.

[0059] S500: Acquire characteristic data of the target object, input the characteristic data into a target prediction model, and obtain a transformation difficulty prediction result through the target prediction model.

[0060] In an embodiment of the present invention, the characteristic data includes real-time psychological assessment data of the target object, dynamic information of the target object, and static information of the target object. By inputting these characteristic data of each target object into the target prediction model, the prediction result of the difficulty of transformation of the target object can be obtained.

[0061] The method for predicting the degree of difficulty of reforming a target object provided by an embodiment of the present invention is based on the psychological assessment data of the target object, and then combines the static information and dynamic information of the target object to build a prediction factor database, and sorts the prediction factors by importance, which can reduce the computational complexity and model complexity. The target prediction model is obtained by training based on the prediction factor database, and the target prediction model is continuously updated and optimized through the dynamic information of the target object in prison, thereby improving the accuracy and effectiveness of the model prediction.

[0062] Example 2 Based on the same technical concept as the method embodiment of the above embodiment 1, the embodiment of the present invention provides a target object transformation difficulty prediction system. Figure 7 is a schematic diagram of the structure of a target object transformation difficulty prediction system provided by an embodiment of the present invention. Figure 7 As shown, the target object transformation difficulty prediction system 700 includes: The data processing module 710 is used to construct an initial prediction factor database based on the psychological assessment data and the static information of the target object.

[0063] The model building module 720 is used to build a machine learning algorithm model, train the machine learning algorithm model according to the initial prediction factor database, and obtain an initial prediction model.

[0064] The factor ranking module 730 is used to rank the prediction factors in the initial prediction factor database according to their importance, and obtain a mid-term prediction factor database based on the ranking results combined with the target object dynamic information.

[0065] The model training module 740 is used to train the initial prediction model according to the medium-term prediction factor database to obtain the target prediction model.

[0066] The result prediction module 750 is used to obtain characteristic data of the target object, input the characteristic data into the target prediction model, and obtain the transformation difficulty prediction result through the target prediction model.

[0067] The target object transformation difficulty prediction system provided by the embodiment of the present invention can objectively and comprehensively judge the transformation difficulty of prison target objects, and provide theoretical support, basis and reference for prison target object transformation management and prevention and control work.

[0068] It can be understood that the implementation method of the method for predicting the difficulty of target object transformation described in the above-mentioned embodiment 1 is also applicable to this embodiment and can achieve the same technical effect, so it will not be repeated here.

[0069] Example 3 Based on the same concept, an embodiment of the present invention further provides an electronic device, Figure 8 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention, such as Figure 8 As shown, the electronic device 800 may include: a processor 810, a communication interface 820, a memory 830 and a communication bus 840, wherein the processor 810, the communication interface 820, and the memory 830 communicate with each other through the communication bus 840. The processor 810 may call the logic instructions in the memory 830 to execute the steps of the method for predicting the degree of difficulty of transformation of the target object as described in the above embodiments. For example, it includes: S100, constructing an initial predictor database based on psychological assessment data and static information of target subjects; S200, constructing a machine learning algorithm model, training the machine learning algorithm model according to an initial prediction factor database, and obtaining an initial prediction model; S300, ranking the prediction factors in the initial prediction factor database by importance, and obtaining a mid-term prediction factor database based on the ranking results and the target object dynamic information; S400, training an initial prediction model according to a mid-term prediction factor database to obtain a target prediction model; S500: Acquire characteristic data of the target object, input the characteristic data into a target prediction model, and obtain a transformation difficulty prediction result through the target prediction model.

[0070] The processor 810 may be a central processing unit (CPU). The processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or a combination of the above chips.

[0071] In addition, the logic instructions in the above-mentioned memory 830 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0072] The memory 830 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created by the processor, etc. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0073] Example 4 Based on the same concept, an embodiment of the present invention further provides a computer-readable storage medium, which stores a computer program, and the computer program includes at least one code segment, which can be executed by a main control device to control the main control device to implement the steps of the method for predicting the degree of difficulty of transformation of a target object as described in the above embodiments. For example, it includes: S100, constructing an initial predictor database based on psychological assessment data and static information of target subjects; S200, constructing a machine learning algorithm model, training the machine learning algorithm model according to an initial prediction factor database, and obtaining an initial prediction model; S300, ranking the prediction factors in the initial prediction factor database by importance, and obtaining a mid-term prediction factor database based on the ranking results and the target object dynamic information; S400, training an initial prediction model according to a mid-term prediction factor database to obtain a target prediction model; S500: Acquire characteristic data of the target object, input the characteristic data into a target prediction model, and obtain a transformation difficulty prediction result through the target prediction model.

[0074] Based on the same technical concept, an embodiment of the present invention further provides a computer program, which is used to implement the above method embodiment when executed by a main control device.

[0075] The computer program may be stored in whole or in part on a computer-readable storage medium packaged with the processor, or may be stored in whole or in part on a memory not packaged with the processor.

[0076] Based on the same technical concept, an embodiment of the present invention further provides a processor, which is used to implement the above method embodiment. The above processor can be a chip.

[0077] In summary, the method, system, device and storage medium for predicting the degree of difficulty of transformation of a target object provided by the present invention are based on the psychological evaluation data of the target object, and then combined with the static information and dynamic information of the target object to construct a prediction factor database, and the prediction factors are ranked according to their importance, which can reduce the computational complexity and model complexity, and the target prediction model is obtained by training according to the prediction factor database, and the target prediction model is continuously updated and optimized through the dynamic information of the target object in the prison, thereby improving the accuracy and effectiveness of the model prediction. The present invention can objectively and comprehensively judge the degree of difficulty of transformation of the prison target object, and provide theoretical support, basis and reference for the transformation management and prevention and control of the prison target object.

[0078] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0079] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

[0080] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for predicting the difficulty of transformation of a target object, characterized in that: The method comprises: Construct an initial predictor database based on psychological assessment data and static information of target subjects; Constructing a machine learning algorithm model, and training the machine learning algorithm model according to the initial prediction factor database to obtain an initial prediction model; Sorting the prediction factors in the initial prediction factor database by importance, and obtaining a mid-term prediction factor database based on the sorting results and the target object dynamic information; Training the initial prediction model according to the mid-term prediction factor database to obtain a target prediction model; Acquire characteristic data of the target object, input the characteristic data into the target prediction model, and obtain a transformation difficulty prediction result through the target prediction model.

2. The method for predicting the difficulty of transformation of a target object according to claim 1, characterized in that: The initial prediction factor database is constructed based on the psychological assessment data and the static information of the target object, including: Acquiring the psychological assessment data of the target object according to the psychological assessment document; Acquire static information of the target object from a target object management system; The initial prediction factor database is constructed according to the psychological assessment data and the target object static information.

3. The method for predicting the difficulty of transformation of a target object according to claim 2, characterized in that: The step of constructing a machine learning algorithm model and training the machine learning algorithm model according to the initial prediction factor database to obtain an initial prediction model includes: Extracting target object data from the initial prediction factor database, merging the target object data according to the target object number to obtain merged data, wherein the target object data includes the psychological evaluation data of the target object and the target object static information; Calculate the total score of the psychological dimension according to the psychological assessment data in the merged data, classify the total score of the dimension, and obtain a prediction factor and a prediction label; Dividing the merged data into a training set and a test set according to the predicted labels; Construct a machine learning algorithm model, use the training set to train the machine learning algorithm model, and use the test set to test the trained machine learning algorithm model to obtain the initial prediction model.

4. The method for predicting the difficulty of transformation of a target object according to claim 3, characterized in that: The step of calculating the total score of the psychological dimension according to the psychological assessment data in the merged data and classifying the total score of the dimension includes: Obtaining the psychological assessment data of all the target objects; Calculating the evaluation result score corresponding to each psychological dimension in the psychological evaluation data, performing weighted addition on each of the evaluation result scores, and obtaining the dimension total score of each of the psychological dimensions; A classification threshold is set, and the dimension total score is classified according to the classification threshold.

5. The method for predicting the difficulty of transformation of a target object according to claim 1, characterized in that: The step of ranking the predictive factors in the initial predictive factor database by importance comprises: Selecting any of the predictive factors in the initial predictive factor database, randomly modifying the characteristic value of the predictive factor, and obtaining a modified data set; Inputting the modified data set into the initial prediction model for prediction to obtain a modified prediction value; Calculate the impact value of each of the prediction factors on the loss function after modification according to the modified prediction value and the actual prediction value of the actual data set; The sizes of the influence values ​​are compared, and the prediction factors are ranked in order of importance.

6. The method for predicting the difficulty of transformation of a target object according to claim 1, characterized in that: The step of training the initial prediction model according to the mid-term prediction factor database to obtain a target prediction model comprises: Dividing the data in the medium-term predictor database into a training set and a test set; In the training set, the predictive factors in the mid-term predictive factor database are used as features, and the degree of difficulty of transformation of the target object is used as a label to train the initial prediction model; The hyperparameters of the initial prediction model are adjusted by multi-fold crossover and Bayesian search to obtain an initial prediction model with an optimal area under the curve; A threshold of the initial prediction model is determined, and the initial prediction model is verified using the test set to obtain the target prediction model.

7. The method for predicting the difficulty of transformation of a target object according to claim 6, characterized in that: The using the test set to verify the initial prediction model includes: Inputting the test set into the initial prediction model, calculating the evaluation index of the initial prediction model, wherein the evaluation index at least includes recall rate, area under the curve and accuracy; The performance of the initial prediction model is verified according to the evaluation index.

8. A target object transformation difficulty prediction system, characterized in that: The system comprises: A data processing module is used to construct an initial predictor database based on psychological assessment data and static information of target subjects; A model building module, used to build a machine learning algorithm model, train the machine learning algorithm model according to the initial prediction factor database, and obtain an initial prediction model; A factor ranking module, used to rank the prediction factors in the initial prediction factor database according to their importance, and obtain a mid-term prediction factor database based on the ranking results combined with the target object dynamic information; A model training module, used for training the initial prediction model according to the mid-term prediction factor database to obtain a target prediction model; The result prediction module is used to obtain the characteristic data of the target object, input the characteristic data into the target prediction model, and obtain the transformation difficulty prediction result through the target prediction model.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: The processor executes the computer program to implement the method for predicting the difficulty of transformation of a target object as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for predicting the difficulty of transformation of a target object as described in any one of claims 1 to 7 is implemented.

Images