Intelligent simulation training method and device based on dynamic images

Abstract

The application relates to the technical field of artificial intelligence, and discloses an intelligent simulation training method and device based on dynamic portraits, which comprises the following steps: obtaining basic portrait information of a target student, generating training demand information of the target student based on the basic portrait information and a time sequence knowledge graph; obtaining real-time interaction information of the target student, determining corresponding evaluation analysis results based on the real-time interaction information and the basic portrait information; generating dynamic portrait information of the target student according to the evaluation analysis results and the basic portrait information, then generating simulation training parameters, and performing real-time simulation operation on the target student based on the simulation training parameters to obtain real-time simulation results corresponding to the target student; and generating simulation training evaluation results of the target student according to the real-time simulation results. It can be seen that the application can perform real-time simulation training to improve the real-time performance of simulation training, thereby being beneficial to improving the accuracy and reliability of simulation training, and further improving the training effect of simulation training.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and in particular to an intelligent simulation training method and apparatus based on dynamic profiling. Background Technology

[0002] With the rapid development of artificial intelligence and digital technologies, immersive training has become an important means to enhance users' advanced abilities such as complex decision-making, adaptability, and communication. Currently, intelligent training solutions for cultivating various professional skills are constantly emerging, including online course platforms and static role-playing simulations. These solutions can achieve large-scale instruction of basic knowledge and simple assessment of mastery, but their training modes are relatively singular and fixed, failing to meet users' practical training needs.

[0003] Therefore, it is particularly important to provide a new simulation training method to achieve dynamic and real-time training and improve the training effect for users. Summary of the Invention

[0004] This invention provides an intelligent simulation training method and apparatus based on dynamic profiling, which can improve the real-time performance of simulation training by conducting real-time simulation training, thereby improving the accuracy and reliability of simulation training and ultimately enhancing the training effect of simulation training.

[0005] The first aspect of this invention discloses an intelligent simulation training method based on dynamic profiling, the method comprising: Obtain basic profile information of the target student, and generate practical training requirement information corresponding to the target student based on the basic profile information and the pre-determined temporal knowledge graph. The real-time interaction information of the target student is obtained, and the evaluation and analysis results corresponding to the target student are determined based on the real-time interaction information and the basic profile information. The evaluation and analysis results include the simulated interaction analysis results of the target student. Based on the evaluation and analysis results and the basic profile information, dynamic profile information of the target student is generated; Based on the dynamic profile information, simulated training parameters corresponding to the target student are generated, and real-time simulation operations are performed on the target student based on the simulated training parameters to obtain the real-time simulation results corresponding to the target student. Based on the real-time simulation results, the simulated training evaluation results for the target trainees are generated.

[0006] As an optional implementation, in the first aspect of the present invention, before generating the simulated training evaluation result of the target trainee based on the real-time simulation result, the method further includes: Obtain real-time behavioral data corresponding to the target student, and determine the multimodal behavioral information corresponding to the target student based on the real-time behavioral data; Based on the multimodal behavioral information, an analysis operation is performed on the multimodal behavioral information based on the pre-determined temporal knowledge graph to obtain the multimodal evaluation result of the target student; The step of generating the simulated training evaluation results for the target trainee based on the real-time simulation results includes: Based on the real-time simulation results and the multimodal evaluation results, the simulated training evaluation results for the target trainees are generated.

[0007] As an optional implementation, in a first aspect of the present invention, generating dynamic profile information of the target trainee based on the evaluation analysis results and the basic profile information includes: Based on the evaluation and analysis results and the basic profile information, the target interaction data of the target learner is determined, wherein the target interaction data includes text interaction data, emotional interaction data, and temporal interaction data; Extract the interaction feature data corresponding to the target interaction data, and determine the multi-dimensional quantization result corresponding to the target student based on the interaction feature data and a pre-determined multi-dimensional quantization algorithm. Based on the multi-dimensional quantification results, dynamic profile information of the target student is generated.

[0008] As an optional implementation, in a first aspect of the present invention, the method further includes: Based on the simulated training evaluation results of the target trainees, determine whether the simulated training evaluation results meet the preset simulated training conditions; When it is determined that the simulation training evaluation result does not meet the preset simulation training conditions, analyze the target reasons corresponding to the simulation training evaluation result not meeting the preset simulation training conditions; Based on the stated cause, a target matching strategy corresponding to the stated cause is determined from a pre-defined matching strategy library, and the simulation training parameters corresponding to the target student are updated based on the stated target matching strategy.

[0009] As an optional implementation, in a first aspect of the present invention, determining a target matching strategy corresponding to the target cause from a pre-determined matching strategy library based on the target cause includes: Obtain the development goal information corresponding to the target student, and determine the simulation training path information corresponding to the target student based on the goal reasons and the development goal information; Based on the simulated training path information, a capability development path map corresponding to the target student is generated, and based on the capability development path map and the target reasons, the demand strategy information corresponding to the target student is determined. Based on the demand strategy information, a target matching strategy corresponding to the target reason is determined from a pre-determined matching strategy library.

[0010] As an optional implementation, in the first aspect of the present invention, after generating the dynamic profile information of the target student based on the evaluation analysis results and the basic profile information, the method further includes: Obtain the historical training data corresponding to the target student, and determine the historical evaluation results of the target student based on the historical training data; Based on the historical evaluation results, a training trend curve corresponding to the target student is generated, and based on the training trend curve, historical profile information of the target student is generated. Based on the historical profile information, update the dynamic profile information of the target student.

[0011] As an optional implementation, in a first aspect of the present invention, generating the simulated training evaluation result of the target trainee based on the real-time simulation result includes: Extract multimodal feature data from the real-time simulation results, wherein the multimodal feature data includes text feature data, speech feature data, and time-series feature data; Based on the multimodal feature data and the pre-determined multimodal fusion model, the fusion feature data of the target student is determined; Based on the fused feature data and the pre-determined weighted aggregation algorithm, the quantitative evaluation data of the target student is calculated; Based on the quantitative assessment data, the simulated training assessment results for the target trainees are generated.

[0012] A second aspect of this invention discloses an intelligent simulation training device based on dynamic profiling, the device comprising: The acquisition module is used to obtain basic profile information of the target students. The generation module is used to generate the practical training requirements information corresponding to the target trainee based on the basic profile information and the pre-determined temporal knowledge graph. The acquisition module is also used to acquire the real-time interaction information of the target student; The determining module is used to determine the evaluation analysis result corresponding to the target student based on the real-time interaction information and the basic profile information, wherein the evaluation analysis result includes the simulated interaction analysis result of the target student; The generation module is further configured to generate dynamic profile information of the target trainee based on the evaluation analysis results and the basic profile information; and generate simulation training parameters corresponding to the target trainee based on the dynamic profile information. The simulation module is used to perform real-time simulation operations on the target student based on the simulation training parameters, and obtain the real-time simulation results corresponding to the target student. The generation module is also used to generate the simulated training evaluation results of the target trainees based on the real-time simulation results.

[0013] As an optional implementation, in a second aspect of the present invention, the acquisition module is further configured to acquire real-time behavioral data corresponding to the target student before the generation module generates the simulated training evaluation result of the target student based on the real-time simulation result; The determining module is further configured to determine the multimodal behavioral information corresponding to the target student based on the real-time behavioral data; The device further includes: The analysis module is used to perform analysis operations on the multimodal behavior information based on the pre-determined temporal knowledge graph to obtain the multimodal evaluation results of the target learner. The specific method by which the generation module generates the simulated training evaluation results for the target trainee based on the real-time simulation results includes: Based on the real-time simulation results and the multimodal evaluation results, the simulated training evaluation results for the target trainees are generated.

[0014] As an optional implementation, in a second aspect of the present invention, the specific method by which the generation module generates the dynamic profile information of the target trainee based on the evaluation analysis results and the basic profile information includes: Based on the evaluation and analysis results and the basic profile information, the target interaction data of the target learner is determined, wherein the target interaction data includes text interaction data, emotional interaction data, and temporal interaction data; Extract the interaction feature data corresponding to the target interaction data, and determine the multi-dimensional quantization result corresponding to the target student based on the interaction feature data and a pre-determined multi-dimensional quantization algorithm. Based on the multi-dimensional quantification results, dynamic profile information of the target student is generated.

[0015] As an optional implementation, in a second aspect of the invention, the apparatus further includes: The judgment module is used to determine whether the simulated training evaluation results of the target trainee meet the preset simulated training conditions based on the simulated training evaluation results of the target trainee. The analysis module is further configured to analyze the target reasons corresponding to the simulation training evaluation results not meeting the preset simulation training conditions when the judgment module determines that the simulation training evaluation results do not meet the preset simulation training conditions. The determining module is further configured to determine, based on the target cause, a target matching strategy corresponding to the target cause from a pre-determined matching strategy library; The first update module is used to update the simulation training parameters corresponding to the target student based on the target matching strategy.

[0016] As an optional implementation, in a second aspect of the present invention, the specific method by which the determining module determines the target matching strategy corresponding to the target cause from a pre-determined matching strategy library based on the target cause includes: Obtain the development goal information corresponding to the target student, and determine the simulation training path information corresponding to the target student based on the goal reasons and the development goal information; Based on the simulated training path information, a capability development path map corresponding to the target student is generated, and based on the capability development path map and the target reasons, the demand strategy information corresponding to the target student is determined. Based on the demand strategy information, a target matching strategy corresponding to the target reason is determined from a pre-determined matching strategy library.

[0017] As an optional implementation, in a second aspect of the present invention, the acquisition module is further configured to acquire the historical training data corresponding to the target student after the generation module generates the dynamic profile information of the target student based on the evaluation analysis results and the basic profile information. The determining module is further configured to determine the historical evaluation results of the target student based on the historical training data; The generation module is further configured to generate a training trend curve corresponding to the target student based on the historical evaluation results, and generate historical profile information of the target student based on the training trend curve. The device further includes: The second update module is used to update the dynamic profile information of the target student based on the historical profile information.

[0018] As an optional implementation, in a second aspect of the present invention, the specific method by which the generation module generates the simulated training evaluation results of the target trainee based on the real-time simulation results includes: Extract multimodal feature data from the real-time simulation results, wherein the multimodal feature data includes text feature data, speech feature data, and time-series feature data; Based on the multimodal feature data and the pre-determined multimodal fusion model, the fusion feature data of the target student is determined; Based on the fused feature data and the pre-determined weighted aggregation algorithm, the quantitative evaluation data of the target student is calculated; Based on the quantitative assessment data, the simulated training assessment results for the target trainees are generated.

[0019] A third aspect of this invention discloses another intelligent simulation training device based on dynamic imagery, the device comprising: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute some or all of the steps in the intelligent simulation training method based on dynamic portrait according to any of the first aspects of the present invention.

[0020] The fourth aspect of the present invention discloses a computer storage medium storing computer instructions, which, when invoked, are used to execute some or all of the steps in the intelligent simulation training method based on dynamic portrait described in any of the first aspects of the present invention.

[0021] Compared with the prior art, the present invention has the following beneficial effects: In this embodiment of the invention, basic profile information of the target trainee is obtained; training requirements information of the target trainee is generated based on the basic profile information and a temporal knowledge graph; real-time interaction information of the target trainee is obtained; corresponding evaluation and analysis results are determined based on the real-time interaction information and the basic profile information; dynamic profile information of the target trainee is generated based on the evaluation and analysis results and the basic profile information, and simulation training parameters are generated; real-time simulation operations are performed on the target trainee based on the simulation training parameters to obtain the real-time simulation results corresponding to the target trainee; and simulation training evaluation results of the target trainee are generated based on the real-time simulation results. It is evident that implementing this invention can improve the real-time nature of simulation training by conducting real-time simulation training, thereby improving the accuracy and reliability of simulation training and ultimately enhancing the training effect of simulation training. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1This is a flowchart illustrating an intelligent simulation training method based on dynamic profiling disclosed in an embodiment of the present invention. Figure 2 This is a flowchart illustrating another intelligent simulation training method based on dynamic profiling disclosed in an embodiment of the present invention. Figure 3 This is a schematic diagram of the structure of an intelligent simulation training device based on dynamic profiling disclosed in an embodiment of the present invention; Figure 4 This is a schematic diagram of another intelligent simulation training device based on dynamic image disclosed in an embodiment of the present invention; Figure 5 This is a schematic diagram of the structure of another intelligent simulation training device based on dynamic image disclosed in an embodiment of the present invention. Detailed Implementation

[0024] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0025] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or end that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or ends.

[0026] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0027] This invention discloses an intelligent simulation training method and apparatus based on dynamic profiling, which can improve the real-time performance of simulation training by conducting real-time simulation training, thereby improving the accuracy and reliability of simulation training and ultimately enhancing the training effect. Detailed descriptions follow.

[0028] Example 1 Please see Figure 1 , Figure 1 This is a flowchart illustrating an intelligent simulation training method based on dynamic profiling disclosed in an embodiment of the present invention. Wherein, Figure 1 The described intelligent simulation training method based on dynamic profiling can be applied to intelligent simulation training devices based on dynamic profiling. These devices can be integrated into cloud servers or local servers; this embodiment of the invention does not impose any limitations. Figure 1 As shown, the intelligent simulation training method based on dynamic profiles can include the following operations.

[0029] 101. Obtain basic profile information of the target students, and generate practical training requirement information corresponding to the target students based on the basic profile information and the pre-determined temporal knowledge graph.

[0030] In this embodiment of the invention, optionally, the basic profile information of the target trainee may include one or more of the following: the target trainee's identity information, personal background information, historical training record information, assessment information, and work ability information.

[0031] In this embodiment of the invention, optionally, the pre-determined temporal knowledge graph can be generated based on a graph database and by parsing the training objectives and historical training information of the target learners through natural language processing and automatically extracting entities and relationships. The graph database may include the Neo4j graph database, and the natural language processing may include NLP processing. Neo4j is an open-source, high-performance graph database that uses nodes and relationships as its core abstractions to directly organize data into a "network" structure. Natural Language Processing (NLP) can be used to perform natural language processing on text.

[0032] In this embodiment of the invention, optionally, the generation of practical training requirement information corresponding to the target trainee based on basic profile information and a pre-determined temporal knowledge graph may include: A training risk assessment is performed on the basic profile information to obtain the training risk assessment results corresponding to the target trainees. The training risk assessment results may include information on the training shortcomings of the target trainees. Based on the results of the practical training risk assessment and the pre-determined time-series knowledge graph, the practical training needs scenarios and practical training objectives corresponding to the target students are determined. Based on the practical training needs scenario and the practical training objectives corresponding to the target students, generate practical training needs information corresponding to the target students.

[0033] In this embodiment of the invention, optionally, the training shortcomings information of the target trainee may include the target trainee's ability shortcomings in training. For example, if the target trainee's adaptability is weak, the training shortcomings information of the target trainee may include adaptability information, and the training requirement scenario may include a safety production accident handling scenario.

[0034] In this embodiment of the invention, optionally, the pre-determined temporal knowledge graph can also be used for domain knowledge question answering. However, the interaction is mainly one-way question answering, which makes it difficult to support in-depth discussions and strategy deductions in multiple rounds around a complex case. Therefore, it can also be based on a deduction engine using a local large language model and temporal knowledge graph. After the initial situation (such as a sudden event) is input, the engine will infer the subsequent development of the situation in real time based on the student's decisions at each step and the interaction of the digital human role, generating new challenges and plot branches.

[0035] 102. Obtain real-time interaction information of the target learners, and determine the corresponding evaluation and analysis results based on the real-time interaction information and basic profile information.

[0036] In this embodiment of the invention, the evaluation and analysis results include the simulated interaction analysis results of the target learners.

[0037] In this embodiment of the invention, optionally, the real-time interactive information of the target learner may include one or more of the target learner's interactive text information, voice tone information, and interactive response information; wherein, the interactive text information includes the target learner's text content information, semantic feature tag information, keyword information, and text structure information, wherein the text content information includes the complete text actively input and / or output by the target learner, including decision statements, supporting arguments, response scripts, and solution suggestions, etc.; the semantic feature tag information includes text semantic attributes extracted based on NLP algorithms, including core themes, logical relationships, and stance inclinations, etc.; the keyword information includes the frequency and proportion of keywords related to the core training capabilities (such as decision-making, communication, and risk-related words); the text structure information includes the text organization form, including whether it is presented in bullet points, the completeness of the logical chain, and the relevance between arguments and conclusions, etc.; the voice tone information includes one or more of the target learner's basic voice feature information, emotional tendency information, tone change information, and voice semantic matching information; wherein, the basic voice feature information includes the physical attributes of the voice, including speech rate, average pitch, pitch fluctuation amplitude, and speech clarity, etc.; the emotional tendency information includes emotion labels based on voice feature recognition. The system includes: a signature (calm, firm, anxious, angry, tense, etc.) and its corresponding intensity value; tone variation information including the tone adaptability of the voice in different interactive scenarios (such as reporting, responding to questions, and reassuring communication); voice semantic matching information including the consistency between the semantics and tone emotion after voice-to-text conversion; and interactive response information including the target learner's response duration, response-interaction interval duration, and interaction frequency. Among these, the response duration information includes the average duration of the target learner's voice response, the response-interaction interval duration information includes the time interval information of the target learner's response to the current scenario, and the interaction frequency information includes the number of times the target learner initiates communication, asks follow-up questions, and interrupts others.

[0038] In this embodiment of the invention, optionally, the above-mentioned assessment and analysis results for determining the target student based on real-time interaction information and basic profile information may include: By using pre-determined extraction parameters, feature extraction operations are performed on real-time interactive information and basic profile information to obtain interactive feature information; Based on the interaction feature information and basic profile information, feature fusion operation is performed to obtain the simulated interaction analysis results corresponding to the target student, and the evaluation analysis results of the target student are generated based on the simulated interaction analysis results. The pre-determined extraction parameters may include extracting semantic coherence features using a BERT model, extracting sentiment features using a CNN model, and extracting decision-making decisiveness features using an LSTM model.

[0039] In this embodiment of the invention, optionally, the simulated interaction analysis results of the target trainee may include the simulated interaction results of the target trainee during the training process, which are obtained by combining basic profile information and real-time interaction information.

[0040] In this embodiment of the invention, optionally, the evaluation and analysis results of the target learner may include the behavioral characteristic analysis results of the target learner; for example, the characteristic analysis results of the target learner may include the target learner's emotional stability characteristics, speech rate information, and word choice information.

[0041] In this embodiment of the invention, the evaluation and analysis results of the target trainees may further include the analysis results of dimensions such as the target trainees' decision-making decisiveness, logical consistency, emotional stability, and communication effectiveness.

[0042] 103. Based on the assessment and analysis results and basic profile information, generate dynamic profile information of the target trainees.

[0043] In this embodiment of the invention, optionally, the dynamic profile information of the target trainee may include the target trainee's real-time ability status, ability fluctuation trend, and training strategy information. Further optionally, the target trainee's real-time ability status may include the target trainee's decision-making quality, communication effectiveness, adaptability, and emotional stability; the ability fluctuation trend may include a line trend generated by the target trainee in several training sessions; and the training strategy information may include training strategy information derived from summarizing multiple rounds of decision-making behavior.

[0044] 104. Based on dynamic profile information, generate simulation training parameters corresponding to the target student, and perform real-time simulation operations on the target student based on the simulation training parameters to obtain the real-time simulation results corresponding to the target student.

[0045] In this embodiment of the invention, optionally, the simulated training parameters corresponding to the target trainee may include simulated scenario parameters and simulated interaction rule parameters corresponding to the target trainee. The simulated scenario parameters may include simulated scenario information, and the simulated interaction rule parameters may include simulated interaction round parameters, simulated interaction communication parameters, simulated interaction interface parameters, and simulated interaction handling parameters.

[0046] In this embodiment of the invention, optionally, generating simulated training parameters corresponding to the target trainee based on dynamic profile information may include: Based on dynamic profile information, the real-time ability status of the target trainee is determined, and the training matching rules for the target trainee are determined based on the real-time ability status. Based on the training matching rules for the target trainee, the simulated training parameters for the target trainee are generated. Among them, the training matching rules for target trainees include the target trainees' ability scenario matching rules, the scenario difficulty and ability level matching rules, and the interaction and ability status matching rules.

[0047] 105. Based on the real-time simulation results, generate the simulated training evaluation results for the target trainees.

[0048] In this embodiment of the invention, optionally, the quantitative score parameters corresponding to the target trainee are determined based on the real-time simulation results, and the simulated evaluation results of the target trainee are generated based on the quantitative score parameters and the pre-determined training evaluation level parameters.

[0049] In this embodiment of the invention, the method may further optionally include: generating a multi-dimensional ability assessment report for the target student based on the simulated training assessment results of the target student, wherein the multi-dimensional ability assessment report is used to display the target student's ability performance in each ability item.

[0050] In this embodiment of the invention, optionally, for example, 1. Context initialization and profile loading: The student selects or the system specifies an initial complex case (such as "major project public opinion crisis management"), and the system loads the student's basic profile and past ability assessment; the target student conducts immersive simulation interaction with one or more digital human characters (such as press conferences, crisis decision-making meetings), and the dynamic profile engine analyzes each interaction behavior of the student in real time and updates its current ability status profile; the context deduction engine generates new context changes and character reactions in real time based on the student's decision and the updated profile, making the simulation process full of uncertainty and adaptability; during the simulation, the system can provide adaptive guidance through digital human mentors or interface prompts when the student is judged to be in trouble or has cognitive bias based on the profile, rather than simply telling the answer; after the simulation training, a detailed ability assessment report is generated, and key decision points are reproduced for review and analysis, and (2) based on the performance and historical data, the system automatically generates a personalized ability development plan to clarify the training focus of the next stage.

[0051] It is evident that implementation Figure 1The described intelligent simulation training method based on dynamic profiling can acquire basic profiling information of target trainees, generate training requirement information for target trainees based on basic profiling information and temporal knowledge graphs; acquire real-time interaction information of target trainees, determine corresponding evaluation and analysis results based on real-time interaction information and basic profiling information; generate dynamic profiling information of target trainees based on evaluation and analysis results and basic profiling information, thereby generating simulation training parameters, and perform real-time simulation operations on target trainees based on simulation training parameters to obtain real-time simulation results for target trainees; and generate simulation training evaluation results for target trainees based on real-time simulation results. This method can accurately match the training needs of each target trainee, which is beneficial for improving training efficiency. This method enhances the accuracy and reliability of generating training needs information and simulated training parameters for each target learner, thereby improving the accuracy and reliability of obtaining real-time simulation results and simulated training evaluation results. Furthermore, it can combine temporal knowledge graphs and real-time interactive information to further improve the real-time performance and reliability of generated simulated training parameters, and dynamically adapt training parameters based on the real-time interactive information of target learners to obtain more accurate and timely simulated training results. This further enhances the real-time performance of simulated training, thereby improving the accuracy and reliability of simulated training and ultimately improving the training effect.

[0052] Example 2 Please see Figure 2 , Figure 2 This is a flowchart illustrating another intelligent simulation training method based on dynamic profiling disclosed in an embodiment of the present invention. Figure 2 The described intelligent simulation training method based on dynamic profiling can be applied to intelligent simulation training devices based on dynamic profiling. These devices can be integrated into cloud servers or local servers; this embodiment of the invention does not impose any limitations. Figure 2 As shown, this intelligent simulation training method based on dynamic profiling can include the following operations: 201. Obtain basic profile information of the target students, and generate practical training requirements information corresponding to the target students based on the basic profile information and the pre-determined temporal knowledge graph.

[0053] 202. Obtain real-time interaction information of the target learners, and determine the corresponding evaluation and analysis results based on the real-time interaction information and basic profile information.

[0054] 203. Based on the assessment and analysis results and basic profile information, generate dynamic profile information of the target trainees.

[0055] 204. Based on the dynamic profile information, generate the simulation training parameters corresponding to the target student, and perform real-time simulation operations on the target student based on the simulation training parameters to obtain the real-time simulation results corresponding to the target student.

[0056] In this embodiment of the invention, for a detailed description of steps 201-204, please refer to the other descriptions of steps 101-104 in Embodiment 1. This embodiment of the invention will not repeat them.

[0057] 205. Obtain real-time behavioral data corresponding to the target learners, and determine the multimodal behavioral information corresponding to the target learners based on the real-time behavioral data.

[0058] In this embodiment of the invention, optionally, the real-time behavior data corresponding to the target learner may include real-time text interaction behavior data, real-time voice interaction behavior data, and real-time timing operation interaction behavior data corresponding to the target learner; wherein, the real-time text interaction behavior data may include text content data, text input method data, and text modification record data input by the target learner; the real-time voice interaction behavior data may include the target learner's original audio stream data and voice triggering timing data; the real-time timing operation interaction behavior data may include decision-making interaction behavior data, interaction interval duration data, and operation frequency data.

[0059] In this embodiment of the invention, optionally, the determination of the multimodal behavioral information corresponding to the target learner based on real-time behavioral data may include: Perform data preprocessing operations on real-time behavioral data to perform data partitioning operations on the real-time behavioral data, thereby obtaining at least one data modality, wherein each data modality contains at least one piece of real-time behavioral data; Based on each data modality and the real-time behavioral data contained in each data modality, determine the modal data association between each data modality and the real-time behavioral data contained in that data modality, and based on all modal data associations, determine the multimodal behavioral information corresponding to the target learner.

[0060] In this embodiment of the invention, optionally, for example, the data modality may include text modality and speech modality. If it includes text modality, text vectors are extracted using a BERT pre-trained model (fine-tuned dataset: 100,000+ cadre training texts), core topics are determined by matching with a domain topic dictionary (such as an emergency management topic library), logical relationships are identified based on dependency parsing, and modal data association relationships corresponding to the text modality are obtained. If it includes speech modality, speech modality feature information such as speech rate, tone, and volume is extracted, and modal data association relationships corresponding to the speech modality are obtained based on the speech modality feature information.

[0061] 206. Based on the multimodal behavioral information, perform analysis operations on the multimodal behavioral information according to the pre-determined temporal knowledge graph to obtain the multimodal evaluation results of the target learners.

[0062] In this embodiment of the invention, optionally, the above-mentioned analysis operation on the multimodal behavioral information based on a pre-determined temporal knowledge graph to obtain the multimodal evaluation result of the target learner may include: Based on multimodal behavioral information and a pre-determined temporal knowledge graph, the modal assessment score information corresponding to each data modality is determined, and a consistency check operation is performed based on each modal assessment score information to obtain the consistency check result. Based on the consistency check result, the multimodal assessment result of the target learner is generated. The multimodal assessment result may include the core behavioral ability assessment result, the voice expression ability assessment result, and the temporal response ability assessment result.

[0063] 207. Based on the real-time simulation results and multimodal evaluation results, generate the simulated training evaluation results for the target trainees.

[0064] In this embodiment of the invention, optionally, the simulated training evaluation results of the target trainee may include the target trainee's comprehensive ability score, multi-dimensional ability radar chart data results, core strengths evaluation results, and core weaknesses evaluation results; furthermore, the simulated training evaluation results may also include the personalized improvement suggestion evaluation results corresponding to the target trainee.

[0065] It is evident that implementation Figure 2 The described intelligent simulation training method based on dynamic profiling can acquire real-time behavioral data of target trainees and determine multimodal behavioral information. Based on this multimodal behavioral information and a temporal knowledge graph, it performs analysis operations to obtain multimodal evaluation results. Finally, based on the real-time simulation results and the multimodal evaluation results, it generates simulation training evaluation results for the target trainees. By collecting real-time behavioral data, it expands the data collection dimensions, improving the accuracy and comprehensiveness of data collection. Furthermore, it enhances the density and breadth of data collection through the acquired multimodal data, avoiding the limitations of a single data modality, and providing a foundation for subsequent simulation training and... The evaluation results provide richer evidence, and the real-time data collection ensures the authenticity of the target learner's behavior. Furthermore, the multimodal evaluation results allow for the correction and analysis of simulation deviations, which helps improve the accuracy and reliability of subsequent simulation training and the obtained simulation training results. It also enables dynamic adaptation of training parameters based on the target learner's real-time interaction information to obtain more accurate and timely simulation training results. Moreover, it allows for real-time simulation training to improve the real-time nature of the training, thereby enhancing the accuracy and reliability of the simulation training and ultimately improving its effectiveness.

[0066] In an optional embodiment, dynamic profile information of the target trainee is generated based on the evaluation analysis results and basic profile information, including: Based on the assessment and analysis results and basic profile information, the target interaction data of the target learners is determined. The target interaction data includes text interaction data, emotional interaction data, and temporal interaction data. Extract the interaction feature data corresponding to the target interaction data, and determine the multi-dimensional quantization result corresponding to the target student based on the interaction feature data and the pre-determined multi-dimensional quantization algorithm. Based on the multi-dimensional quantitative results, dynamic profile information of the target students is generated.

[0067] In this optional embodiment, the determination of target interaction data for target learners based on the evaluation analysis results and basic profile information may include: The current interaction data of the target trainees is extracted from the evaluation and analysis results, and the historical interaction data of the target trainees is extracted from the basic profile information. Based on the current interaction data and the historical interaction data, target interaction data that is strongly related to the simulated training of the target trainees is selected. The text interaction data may include the target learner's interactive text data, text semantic data, and text structural feature data. The interactive text data may include text expression data, the text semantic data may include text topic data and text logical relationships, and the text structural feature data may include keyword density data. Emotional interaction data can include the target learner's voice emotion interaction data, emotion intensity data, and voice intonation feature data; The time-series interaction data can include the target learner's decision response time, interaction interval time, and real-time response time.

[0068] In this optional embodiment, the extraction of interaction feature data corresponding to the target interaction data, and the determination of the multi-dimensional quantization result corresponding to the target learner based on the interaction feature data and a pre-determined multi-dimensional quantization algorithm, may include: Extract the interaction feature data corresponding to the target interaction data. The interaction feature data includes text interaction feature data, sentiment interaction feature data, and temporal feature interaction data. Text interaction feature data may include semantic interaction feature data, keyword density feature data, and text interaction length data. Sentiment interaction feature data may include de-emotional interaction intensity data and tone feature data. Temporal feature interaction data may include response speed interaction feature data and interaction initiative feature data. Based on the interactive feature data and the pre-determined multi-dimensional quantization algorithm, a quantization calculation operation is performed on each interactive feature data to obtain the feature quantization result corresponding to each interactive feature data. Based on the feature quantization results corresponding to all interactive feature data, the multi-dimensional quantization result corresponding to the target student is determined.

[0069] In this optional embodiment, the above-mentioned generation of dynamic profile information of the target trainee based on multi-dimensional quantification results may include: Based on the multi-dimensional quantitative results, the profile composition information of the target trainees is determined. The profile composition information may include one or more of the following: the real-time ability status information, ability fluctuation trend information, and training strategy tendency information of the target trainees. Based on the profile information of the target trainees, the training characteristics and trends of the target trainees are determined, and dynamic profile information of the target trainees is generated based on the training characteristics and trends.

[0070] In this optional embodiment, for example, based on the basic profile information of the target trainee, the historical ability baseline of the target trainee is determined, and the real-time ability status of the target trainee is updated based on the multi-dimensional quantification results. Based on the real-time ability status and the historical ability baseline, the ability fluctuation value and ability fluctuation trend of the target trainee are calculated. By extracting the obtained text feature data and time series feature data, the training feature trend of the target trainee is updated by using the K-means clustering algorithm, thereby generating the dynamic profile information of the target trainee.

[0071] As can be seen, implementing this optional embodiment can determine the target interaction data of the target learner based on the evaluation analysis results and basic profile information, extract the interaction feature data corresponding to the target interaction data and the pre-determined multi-dimensional quantization algorithm, determine the multi-dimensional quantization result corresponding to the target learner, and thus generate the dynamic profile information of the target learner. By focusing on interaction feature data and multi-dimensional quantization algorithm, the accuracy and effectiveness of generating the dynamic profile information of the target learner can be improved. Furthermore, through feature extraction and data processing, the accuracy and reliability of determining the multi-dimensional quantization result of the target learner can be improved, thereby improving the accuracy and reliability of generating the dynamic profile information of the target learner. Moreover, the multi-dimensional quantization algorithm can ensure the objectivity and real-time nature of the generated dynamic profile information. It can also combine the real-time interaction information of the target learner to dynamically adapt the training parameters to obtain more accurate and timely simulated training results. Furthermore, it can improve the real-time nature of simulated training by conducting real-time simulated training, thereby improving the accuracy and reliability of simulated training and ultimately improving the training effect of simulated training.

[0072] In another alternative embodiment, the method further includes: Based on the simulated training evaluation results of the target trainees, determine whether the simulated training evaluation results meet the preset simulated training conditions; When it is determined that the simulation training evaluation results do not meet the preset simulation training conditions, analyze the target reasons corresponding to the simulation training evaluation results not meeting the preset simulation training conditions; Based on the target reason, a target matching strategy corresponding to the target reason is determined from a pre-determined matching strategy library, and the simulation training parameters corresponding to the target trainee are updated based on the target matching strategy.

[0073] In this optional embodiment, it is further possible that when it is determined that the simulation training evaluation result meets the preset simulation training conditions, the process can be terminated, or an iterative update operation can be performed on the predetermined time series knowledge graph based on the simulation training evaluation result and the simulation training parameters.

[0074] In this optional embodiment, the simulated training evaluation results of the target trainees may optionally include one or more of the target trainees' comprehensive ability evaluation results and evaluation results of various dimensions.

[0075] In this optional embodiment, the preset simulated training conditions may optionally include training capability conditions and training scenario adaptation conditions.

[0076] In this optional embodiment, the determination of whether the simulation training evaluation result meets the preset simulation training conditions may include: Determine the quantitative score result of the simulated training corresponding to the simulated training evaluation result, and determine whether the quantitative score result of the simulated training is greater than or equal to the preset threshold of the quantitative score result corresponding to the simulated training conditions; When the simulated training quantitative score result is determined to be greater than or equal to the preset quantitative score result threshold corresponding to the simulated training conditions, the simulated training evaluation result is determined to meet the preset simulated training conditions; when the simulated training quantitative score result is determined to be less than the preset quantitative score result threshold corresponding to the simulated training conditions, the simulated training evaluation result is determined to not meet the preset simulated training conditions.

[0077] In this optional embodiment, the reasons why the above-mentioned analysis simulation training evaluation results do not meet the preset simulation training conditions may include: Based on the simulation training evaluation results and the preset simulation training conditions, identify at least one target factor in which the simulation training evaluation results do not match the preset simulation training conditions, and based on all target factors, identify the target reasons corresponding to the simulation training evaluation results not meeting the preset simulation training conditions; wherein, the target factors may include one or more of the following: scenario influencing factors, ability dimension factors, emotional dimension factors, etc.

[0078] In this optional embodiment, the above-mentioned determination of the target matching strategy corresponding to the target cause from a pre-determined matching strategy library based on the target cause may include: Extract the cause keywords from the target cause, calculate the cause matching degree between the cause keywords and each matching strategy contained in the pre-determined matching strategy library, and filter out the highest matching degree among all cause matching degrees. The matching strategy corresponding to the highest matching degree is determined as the target matching strategy corresponding to the target cause.

[0079] In this optional embodiment, the target matching strategy may further include one or more of the following: training dimension enhancement strategy, training difficulty adjustment strategy, and training interaction optimization strategy; furthermore, the target matching strategy may also include one or more of the following: training implementation steps, training adjustment direction, and training adjustment range.

[0080] In this optional embodiment, the above-mentioned updating of the simulation training parameters corresponding to the target trainee based on the target matching strategy may include: Based on the target matching strategy, the training update parameters for the target trainees are determined. These training update parameters may include training difficulty update parameters and training scenario update parameters. Update the simulation training parameters corresponding to the target trainees based on the training update parameters of the target trainees.

[0081] As can be seen, implementing this optional embodiment can determine whether the preset simulation training conditions are met based on the simulation training evaluation results of the target trainees. If not, it analyzes the reasons why the simulation training evaluation results do not meet the preset simulation training conditions, determines the target matching strategy that matches the target reasons, and updates the simulation training parameters corresponding to the target trainees. It can accurately locate the problems of the target trainees in the simulation training based on the simulation training evaluation results and preset simulation training conditions, and can match the corresponding simulation training parameters for the target trainees, which is conducive to improving the training effect of the target trainees in the simulation training. It can also retain the configuration of the trainees' mastered strengths and optimize the weaknesses based on the simulation training evaluation results of the target trainees. Through multi-dimensional quantification algorithms, it ensures the objectivity and real-time nature of the generated dynamic profile information, and can dynamically adapt the training parameters by combining the real-time interaction information of the target trainees to obtain more accurate and timely simulation training results. Furthermore, it can improve the real-time nature of the simulation training by conducting real-time simulation training, thereby improving the accuracy and reliability of the simulation training and thus improving the training effect of the simulation training.

[0082] In another optional embodiment, based on the target cause, a target matching strategy corresponding to the target cause is determined from a pre-determined matching strategy library, including: Obtain the development goal information corresponding to the target students, and determine the simulation training path information corresponding to the target students based on the reasons for the goals and the development goal information; Based on the simulated training path information, a capability development path map corresponding to the target trainees is generated, and based on the capability development path map and the target reasons, the corresponding demand strategy information of the target trainees is determined. Based on the demand strategy information, the target matching strategy corresponding to the target reason is determined from the pre-determined matching strategy library.

[0083] In this optional embodiment, the development goal information corresponding to the target trainee may include the training goal information corresponding to the simulated training of the target trainee. For example, the development goal information corresponding to the target trainee may include one or more of the following: ability achievement goal information, simulated training phase duration information, and simulated training theme information.

[0084] In this optional embodiment, the above-mentioned determination of the simulated training path information corresponding to the target trainee based on the target reasons and development target information may include: Based on the reasons for the objectives, we determine the training shortcomings of the target trainees and the training target requirements of the target trainees based on the development goals. The training shortcomings include the ability shortcomings of the target trainees during the simulated training, and the training target requirements include the ability goals that the target trainees need to achieve during the simulated training. Based on the training shortcomings and training needs of the target trainees, we determine the simulated training themes and core objectives of each training stage for the target trainees. Based on the simulated training themes and core objectives of each training stage for the target trainees, we determine the simulated training path information for the target trainees. The information for determining the simulated training path for the target trainees includes the training theme for each stage, the core objectives of each simulated training stage, the training sequence for each simulated training stage, and the training duration for each simulated training stage.

[0085] In this optional embodiment, optionally, the above-mentioned generation of a capability development path map corresponding to the target student based on the simulated training path information, and the determination of the target student's corresponding demand strategy information based on the capability development path map and the target reasons, includes: Based on the simulated training path information, determine the simulated training stage information corresponding to the target student. The simulated training stage information corresponding to the target student may include the training stage node information and the training node duration information. Based on the simulated training phase information corresponding to the target trainees, a capability development path map corresponding to the target trainees is generated. The capability development path map corresponding to the target trainees may include a visualization chart of the capability development of the target trainees, which may include training phase node information, training phase duration information, and training capability manifestation information. Based on the capability development roadmap, the core objective information and expected effect information for each training stage corresponding to the target trainees are determined. Based on the reasons for the objectives, the shortcomings that the target trainees need to address in the training stage are determined. Based on the core objective information, expected effect information, and shortcomings, the corresponding demand strategy information for the target trainees is determined. Among them, the demand strategy information may include the direction of the training demand strategy for the target trainees, the priority of the training strategy demand, the training scenario demand information, and the training effect demand information.

[0086] In this optional embodiment, the process of determining the target matching strategy corresponding to the target cause from a pre-determined matching strategy library based on the demand strategy information may include: Extract the strategy keywords corresponding to the demand strategy information. Based on the strategy keywords, determine the target matching strategy that matches the strategy keywords from the pre-determined matching strategy library. The determined target matching strategy matches the training needs of the target trainees.

[0087] As can be seen, implementing this optional embodiment can obtain the development goal information corresponding to the target trainees, determine the simulated training path information based on the goal reasons and development goal information, generate the capability development path map corresponding to the target trainees based on the simulated training path information, and determine the demand strategy information corresponding to the target trainees based on the capability development path map and goal reasons, and then determine the matching goal matching strategy in the matching strategy library. This approach can comprehensively determine the simulated training path information and generate the capability development path map by combining the development goal information and goal reasons of the target trainees, which helps improve the comprehensiveness, accuracy, and reliability of the determined simulated training path information and capability development path map. Furthermore, it can also be based on... The capability development path map generates demand strategy information, which clarifies the priority and application scenarios of strategies at each stage. This helps improve the accuracy and reliability of determining the capability development path map and target matching strategies. It can also intuitively present the simulated training effects and processes of target trainees through capability development rounds, which helps improve the intuitiveness and reliability of simulated training for target trainees. Furthermore, it can dynamically adapt training parameters based on the real-time interaction information of target trainees to obtain more accurate and timely simulated training results. It can further improve the real-time nature of simulated training by conducting real-time simulated training, thereby improving the accuracy and reliability of simulated training and ultimately enhancing the training effect of simulated training.

[0088] In yet another optional embodiment, after generating dynamic profile information of the target trainee based on the evaluation analysis results and basic profile information, the method further includes: Obtain the historical training data corresponding to the target students, and determine the historical evaluation results of the target students based on the historical training data; Based on historical evaluation results, a training trend curve corresponding to the target trainees is generated, and based on the training trend curve, historical profile information of the target trainees is generated. Update the dynamic profile information of target students based on historical profile information.

[0089] In this optional embodiment, the historical training data corresponding to the target trainee may include the target trainee's historical training basic information, historical interaction data, historical multimodal data, and historical evaluation raw data.

[0090] In this optional embodiment, the target learner's historical evaluation results may include the target learner's historical evaluation scores for each round within a preset historical time period, historical scores for each dimension, and historical weaknesses.

[0091] In this optional embodiment, the training trend curve corresponding to the target trainee may optionally include the total ability trend curve, the ability trend curves of each dimension, and the weakness improvement trend curve corresponding to the target trainee.

[0092] In this optional embodiment, the above-mentioned determination of the target trainee's historical evaluation results based on historical training data may include: Based on historical training data, determine the historical multi-dimensional evaluation results corresponding to the target trainees. The historical multi-dimensional evaluation results include the historical round evaluation results and historical evaluation statistics of the target trainees. Based on historical multi-dimensional assessment results, determine the historical assessment results of the target students.

[0093] In this optional embodiment, the process of generating a training trend curve corresponding to the target trainee based on historical evaluation results, and generating historical profile information of the target trainee based on the training trend curve, may include: Based on historical assessment results, determine the historical ability characteristics of the target trainees, and based on the historical ability characteristics, determine the visual trend curve of the target trainees and then generate the corresponding practical training trend curve of the target trainees. Based on the training trend curve, the historical training characteristics of the target trainees are determined, and based on the historical training characteristics, the historical profile information of the target trainees is generated. Among them, historical training characteristics information may include the target trainees' historical training ability assessment information, historical ability growth rate information, and historical weakness characteristics information.

[0094] In this optional embodiment, the updated dynamic profile information of the target student may optionally include the historical profile information corresponding to the target student.

[0095] As can be seen, implementing this optional embodiment can obtain historical training data corresponding to the target trainees and determine their historical evaluation results. Based on the historical evaluation results, it generates training trend curves and historical profile information for the target trainees. It then updates the dynamic profile information of the target trainees based on this historical profile information. This approach, combining historical training data to determine historical evaluation results and generate historical profile information, improves the accuracy and reliability of the generated historical profile information. It also enhances the personalization and matching degree of the generated historical profile information. Furthermore, it integrates the historical profile information to update the dynamic profile information, improving the accuracy and comprehensiveness of the obtained dynamic profile information. By combining the training trend curve with the historical profile, it can uncover long-term patterns behind the scores, enabling the dynamic profile to accurately match the trainees' long-term characteristics. This further allows for real-time simulation training, improving the real-time nature of the simulation training, thereby enhancing the accuracy and reliability of the simulation training and ultimately improving its effectiveness.

[0096] In another optional embodiment, based on the real-time simulation results, a simulated training evaluation result for the target trainee is generated, including: Extract multimodal feature data from real-time simulation results, including text feature data, speech feature data, and time-series feature data; Based on multimodal feature data and a pre-determined multimodal fusion model, the fusion feature data of the target learner is determined; Based on the fusion feature data and the pre-determined weighted aggregation algorithm, the quantitative evaluation data of the target students is calculated; Based on quantitative assessment data, simulated training assessment results for the target trainees are generated.

[0097] In this optional embodiment, the text feature data may include text semantic feature data, text keyword feature data, and text length feature data; the speech feature data may include speech rate feature data, speech tone feature data, speech volume feature data, speech loudness feature data, speech emotion feature data, and speech clarity feature data; the temporal feature data may include response speed feature data, interaction frequency feature data, and interaction scenario adaptation feature data.

[0098] In this optional embodiment, the determination of the target learner's fusion feature data based on multimodal feature data and a pre-determined multimodal fusion model may include: Multimodal feature data is input into a pre-determined multimodal fusion model to perform fusion processing on the multimodal feature data, thereby obtaining the fused feature data of the target learner.

[0099] In this optional embodiment, the pre-determined multimodal fusion model may include an input layer for uniformly mapping text / speech / temporal features to a 256-dimensional vector; a fusion layer for calculating intramodal feature associations, intermodal feature associations, and assigning dynamic weights; and an output layer including a fully connected layer for mapping the fused features to a 512-dimensional high-dimensional vector.

[0100] In this optional embodiment, the above-mentioned calculation of the quantitative evaluation data of the target trainee based on the fused feature data and a pre-determined weighted aggregation algorithm may include: Based on the fused feature data and a pre-determined weighted aggregation algorithm, the feature data weights for each modality are determined. Then, weighted aggregation calculations are performed on each feature data based on its weight, yielding the quantitative assessment data for the target learner. This allows for weighted summation of the fused feature vectors according to the target learner's ability assessment dimensions, transforming the high-dimensional vectors into standardized dimensional scores and a comprehensive score, thus achieving quantitative assessment.

[0101] In this optional embodiment, the above-mentioned generation of simulated training evaluation results for the target trainees based on quantitative evaluation data may include: Based on the quantitative evaluation data, determine the quantitative parameters of the simulated training for the target trainees, and generate the simulated training evaluation results for the target trainees based on the quantitative parameters of the simulated training. The quantitative parameters for the simulated training of the target trainees may include the quantitative scoring results corresponding to the characteristic data of each modality. The evaluation results of the simulated training of the target trainees may include the quantitative score report of the target trainees, the analysis of the target trainees' strengths and weaknesses, the review of key decisions, and the improvement suggestions for the target trainees.

[0102] As can be seen, implementing this optional embodiment can extract multimodal feature data from real-time simulation results and combine it with a multimodal fusion model to determine the fusion feature data of the target trainee. Based on the fusion feature data and a weighted aggregation algorithm, it calculates the quantitative evaluation data of the target trainee and generates the simulated training evaluation results of the target trainee. It can extract multimodal feature data to characterize the training ability of the target trainee in multiple dimensions. Each multimodal feature directly corresponds to the core ability dimension of the target trainee. It can also improve the accuracy and reliability of the evaluation results by combining with a multimodal fusion model. Furthermore, it can assign dynamic weights to different features through an attention mechanism in the multimodal fusion model, which is conducive to improving the flexibility of generating simulated training evaluation results of the target trainee and the matching degree with the real-time situation. It can also further improve the accuracy and reliability of generating simulated training evaluation results of the target trainee through a weighted aggregation algorithm. In addition, it can improve the real-time performance of simulated training by conducting real-time simulated training, thereby improving the accuracy and reliability of simulated training and thus improving the training effect of simulated training.

[0103] Example 3 Please see Figure 3 , Figure 3 This is a schematic diagram of the structure of another intelligent simulation training device based on dynamic imagery disclosed in an embodiment of the present invention. For example... Figure 3 As shown, the intelligent simulation training device based on dynamic portraits may include: Module 301 is used to acquire basic profile information of the target student. The generation module 302 is used to generate training requirement information corresponding to the target trainees based on basic profile information and a pre-determined temporal knowledge graph. The acquisition module 301 is also used to acquire real-time interactive information of the target students; The determination module 303 is used to determine the evaluation and analysis results corresponding to the target student based on real-time interaction information and basic profile information. The evaluation and analysis results include the simulated interaction analysis results of the target student. The generation module 302 is also used to generate dynamic profile information of the target trainee based on the evaluation and analysis results and basic profile information; and to generate simulation training parameters corresponding to the target trainee based on the dynamic profile information. Simulation module 304 is used to perform real-time simulation operations on the target student based on simulation training parameters, and obtain the real-time simulation results corresponding to the target student. The generation module 302 is also used to generate simulated training evaluation results for the target trainees based on the real-time simulation results.

[0104] It is evident that implementation Figure 3The described device can acquire basic profile information of target trainees, generate training requirement information for target trainees based on basic profile information and temporal knowledge graphs; acquire real-time interaction information of target trainees, determine corresponding evaluation and analysis results based on real-time interaction information and basic profile information; generate dynamic profile information of target trainees based on evaluation and analysis results and basic profile information, thereby generating simulated training parameters, and perform real-time simulation operations on target trainees based on simulated training parameters to obtain real-time simulation results for target trainees; and generate simulated training evaluation results for target trainees based on real-time simulation results. This allows for precise matching of training requirements for each target trainee, which is beneficial for improving the generation of training requirements for each target trainee. By accurately and reliably obtaining training needs information and simulation training parameters from trainees, the accuracy and reliability of real-time simulation results and simulation training evaluation results can be improved. Furthermore, by combining time-series knowledge graphs and real-time interactive information, the real-time performance and reliability of generated simulation training parameters can be further enhanced. This also helps to improve the real-time performance and reliability of simulation training evaluation results. In addition, by combining real-time interactive information from trainees, training parameters can be dynamically adapted to obtain more accurate and timely simulation training results. Furthermore, by conducting real-time simulation training, the real-time performance of simulation training can be improved, thereby enhancing the accuracy and reliability of simulation training and ultimately improving the training effect of simulation training.

[0105] In an optional embodiment, such as Figure 4 As shown, the acquisition module 301 is also used to acquire the real-time behavioral data of the target student before the generation module 302 generates the simulated training evaluation result of the target student based on the real-time simulation result. The determination module 303 is also used to determine the multimodal behavioral information corresponding to the target learner based on real-time behavioral data; The device also includes: Analysis module 305 is used to perform analysis operations on multimodal behavioral information based on a pre-determined temporal knowledge graph to obtain the multimodal evaluation results of the target learner. The specific methods by which the generation module 302 generates the simulated training evaluation results for the target trainees based on the real-time simulation results include: Based on the real-time simulation results and multimodal evaluation results, the simulated training evaluation results of the target trainees are generated.

[0106] It is evident that implementation Figure 4The described device can acquire real-time behavioral data corresponding to the target learner and determine multimodal behavioral information. Based on the multimodal behavioral information and temporal knowledge graph, it performs analysis operations on the multimodal behavioral information to obtain multimodal evaluation results. Based on the real-time simulation results and multimodal evaluation results, it generates simulated training evaluation results for the target learner. By collecting real-time behavioral data, it expands the data acquisition dimensions, which helps improve the accuracy and comprehensiveness of data acquisition. It also increases the density and breadth of data acquisition through the collected multimodal data, avoiding the limitations of a single data modality and providing richer evidence for subsequent simulated training and evaluation results. Furthermore, the real-time nature of data acquisition ensures the authenticity of the target learner's behavior. It can also correct and analyze the deviations in simulation results through multimodal evaluation results, which helps to further improve the accuracy and reliability of subsequent simulated training and the obtained simulated training results. Moreover, it can dynamically adapt training parameters by combining the target learner's real-time interaction information to obtain more accurate and timely simulated training results. Finally, it can improve the real-time nature of simulated training by conducting real-time simulated training, thereby improving the accuracy and reliability of simulated training and ultimately enhancing the training effect.

[0107] In another alternative embodiment, such as Figure 4 As shown, the specific methods by which the generation module 302 generates dynamic profile information of the target trainee based on the evaluation analysis results and basic profile information include: Based on the assessment and analysis results and basic profile information, the target interaction data of the target learners is determined. The target interaction data includes text interaction data, emotional interaction data, and temporal interaction data. Extract the interaction feature data corresponding to the target interaction data, and determine the multi-dimensional quantization result corresponding to the target student based on the interaction feature data and the pre-determined multi-dimensional quantization algorithm. Based on the multi-dimensional quantitative results, dynamic profile information of the target students is generated.

[0108] It is evident that implementation Figure 4The described device can determine the target interaction data of the target learner based on the evaluation analysis results and basic profile information, extract the interaction feature data corresponding to the target interaction data and the pre-determined multi-dimensional quantization algorithm, determine the multi-dimensional quantization result corresponding to the target learner, and then generate the dynamic profile information of the target learner. By focusing on interaction feature data and multi-dimensional quantization algorithm, it can improve the accuracy and effectiveness of generating the dynamic profile information of the target learner. Furthermore, through feature extraction and data processing, it helps to improve the accuracy and reliability of determining the multi-dimensional quantization result of the target learner, thereby improving the accuracy and reliability of generating the dynamic profile information of the target learner. It can also ensure the objectivity and real-time nature of generating the dynamic profile information through multi-dimensional quantization algorithm, and can dynamically adapt training parameters by combining the real-time interaction information of the target learner to obtain more accurate and timely simulated training results. Furthermore, it can improve the real-time nature of simulated training by conducting real-time simulated training, thereby improving the accuracy and reliability of simulated training and ultimately improving the training effect of simulated training.

[0109] In yet another alternative embodiment, such as Figure 4 As shown, the device also includes: The judgment module 306 is used to determine whether the simulated training evaluation results meet the preset simulated training conditions based on the simulated training evaluation results of the target trainees. The analysis module 305 is also used to analyze the target reasons corresponding to the simulation training evaluation results not meeting the preset simulation training conditions when the judgment module 306 determines that the simulation training evaluation results do not meet the preset simulation training conditions. The determination module 303 is also used to determine the target matching strategy corresponding to the target cause from a pre-determined matching strategy library based on the target cause; The first update module 307 is used to update the simulation training parameters corresponding to the target trainees based on the target matching strategy.

[0110] It is evident that implementation Figure 4The described device can determine whether preset simulation training conditions are met based on the simulation training evaluation results of the target trainee. If not, it analyzes the reasons why the simulation training evaluation results do not meet the preset simulation training conditions, determines the target matching strategy that matches the target reasons, and updates the simulation training parameters corresponding to the target trainee. It can accurately locate the problems of the target trainee in the simulation training based on the simulation training evaluation results and preset simulation training conditions, and can match the target trainee with corresponding simulation training parameters, which is conducive to improving the training effect of the target trainee in the simulation training. It can also retain the configuration of the trainee's mastered strengths and optimize the weakness dimensions based on the simulation training evaluation results of the target trainee. Through multi-dimensional quantification algorithms, it ensures the objectivity and real-time nature of the generated dynamic profile information, and can dynamically adapt the training parameters by combining the real-time interaction information of the target trainee to obtain more accurate and timely simulation training results. Furthermore, it can conduct real-time simulation training to improve the real-time nature of simulation training, thereby improving the accuracy and reliability of simulation training, and thus improving the training effect of simulation training.

[0111] In yet another alternative embodiment, such as Figure 4 As shown, the specific methods by which the determining module 303 determines the target matching strategy corresponding to the target cause from a pre-determined matching strategy library include: Obtain the development goal information corresponding to the target students, and determine the simulation training path information corresponding to the target students based on the reasons for the goals and the development goal information; Based on the simulated training path information, a capability development path map corresponding to the target trainees is generated, and based on the capability development path map and the target reasons, the corresponding demand strategy information of the target trainees is determined. Based on the demand strategy information, the target matching strategy corresponding to the target reason is determined from the pre-determined matching strategy library.

[0112] It is evident that implementation Figure 4The described device can acquire development goal information corresponding to target trainees, determine simulated training path information based on the reasons for the goals and the development goal information, generate a capability development path map corresponding to the target trainees based on the simulated training path information, determine the target trainees' corresponding demand strategy information based on the capability development path map and the reasons for the goals, and then determine the matching target matching strategy in the matching strategy library. It can comprehensively determine the simulated training path information and generate the capability development path map by combining the target trainees' development goal information and the reasons for the goals, which helps improve the comprehensiveness, accuracy, and reliability of the determined simulated training path information and capability development path map. Furthermore, it can also be based on capability development... The path map generates demand strategy information, which clarifies the priority and application scenarios of strategies at each stage. This helps improve the accuracy and reliability of determining the capability development path map and target matching strategies. It can also intuitively present the simulated training effects and processes of target trainees through capability development rounds, which helps improve the intuitiveness and reliability of simulated training for target trainees. Furthermore, it can dynamically adapt training parameters by combining the real-time interaction information of target trainees to obtain more accurate and timely simulated training results. It can further improve the real-time nature of simulated training by conducting real-time simulated training, thereby improving the accuracy and reliability of simulated training and ultimately enhancing the training effect of simulated training.

[0113] In yet another alternative embodiment, such as Figure 4 As shown, the acquisition module 301 is also used to acquire the historical training data corresponding to the target student after the generation module 302 generates the dynamic profile information of the target student based on the evaluation analysis results and the basic profile information. Module 303 is also used to determine the historical evaluation results of target trainees based on historical training data; The generation module 302 is also used to generate a training trend curve corresponding to the target trainee based on the historical evaluation results, and to generate historical profile information of the target trainee based on the training trend curve. The device also includes: The second update module 308 is used to update the dynamic profile information of the target student based on historical profile information.

[0114] It is evident that implementation Figure 4The described device can acquire historical training data corresponding to target trainees and determine their historical evaluation results. Based on the historical evaluation results, it generates a training trend curve and historical profile information for the target trainees. It then updates the dynamic profile information of the target trainees based on this historical profile information. This ability to combine historical training data to determine historical evaluation results and generate historical profile information improves the accuracy and reliability of the generated historical profile information, as well as its personalization and matching degree. Furthermore, it integrates historical profile information to update the dynamic profile information, enhancing the accuracy and comprehensiveness of the obtained dynamic profile information. By combining the training trend curve with the historical profile, long-term patterns behind the scores can be uncovered, enabling the dynamic profile to accurately match the trainees' long-term characteristics. This further allows for real-time simulation training, improving the real-time nature of the simulation training and thus enhancing its accuracy and reliability, ultimately improving the training effect.

[0115] In yet another alternative embodiment, such as Figure 4 As shown, the specific methods by which the generation module 302 generates the simulated training evaluation results for the target trainees based on the real-time simulation results include: Extract multimodal feature data from real-time simulation results, including text feature data, speech feature data, and time-series feature data; Based on multimodal feature data and a pre-determined multimodal fusion model, the fusion feature data of the target learner is determined; Based on the fusion feature data and the pre-determined weighted aggregation algorithm, the quantitative evaluation data of the target students is calculated; Based on quantitative assessment data, simulated training assessment results for the target trainees are generated.

[0116] It is evident that implementation Figure 4The described device can extract multimodal feature data from real-time simulation results and combine it with a multimodal fusion model to determine the fusion feature data of the target trainee. Based on the fusion feature data and a weighted aggregation algorithm, it calculates the quantitative evaluation data of the target trainee and generates the simulated training evaluation results of the target trainee. It can extract multimodal feature data to characterize the training ability of the target trainee from multiple dimensions. Each multimodal feature directly corresponds to the core ability dimension of the target trainee. It can also improve the accuracy and reliability of the evaluation results by combining with the multimodal fusion model. Furthermore, it can assign dynamic weights to different features through an attention mechanism in the multimodal fusion model, which is conducive to improving the flexibility of the generated simulated training evaluation results of the target trainee and the matching degree with the real-time situation. It can also further improve the accuracy and reliability of the generated simulated training evaluation results of the target trainee through a weighted aggregation algorithm. In addition, it can improve the real-time performance of the simulated training by conducting real-time simulation training, thereby improving the accuracy and reliability of the simulated training and thus improving the training effect of the simulated training.

[0117] Example 4 Please see Figure 5 , Figure 5 This is a schematic diagram of the structure of another intelligent simulation training device based on dynamic imagery disclosed in an embodiment of the present invention. For example... Figure 5 As shown, the intelligent simulation training device based on dynamic portraits may include: Memory 401 storing executable program code; Processor 402 coupled to memory 401; The processor 402 calls the executable program code stored in the memory 401 to execute some or all of the steps in any of the intelligent simulation training based on dynamic portraits in Embodiment 1 of the present invention.

[0118] Example 5 This invention discloses a computer storage medium storing computer instructions. When these computer instructions are invoked, they are used to execute some or all of the steps in any of the intelligent simulation training methods based on dynamic images disclosed in Embodiment 1 of this invention.

[0119] The device embodiments described above are merely illustrative. The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0120] Through the detailed description of the above embodiments, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, including read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-Erasable Programmable Read-Only Memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0121] Finally, it should be noted that the above embodiments are merely preferred embodiments of the present invention and are only used to illustrate the technical solutions of the present invention, not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for intelligent simulation training based on dynamic profiling, characterized in that, The method includes: Obtain basic profile information of the target student, and generate practical training requirement information corresponding to the target student based on the basic profile information and the pre-determined temporal knowledge graph. The real-time interaction information of the target student is obtained, and the evaluation and analysis results corresponding to the target student are determined based on the real-time interaction information and the basic profile information. The evaluation and analysis results include the simulated interaction analysis results of the target student. Based on the evaluation and analysis results and the basic profile information, dynamic profile information of the target student is generated; Based on the dynamic profile information, simulated training parameters corresponding to the target student are generated, and real-time simulation operations are performed on the target student based on the simulated training parameters to obtain the real-time simulation results corresponding to the target student. Based on the real-time simulation results, the simulated training evaluation results for the target trainees are generated.

2. The intelligent simulation training method based on dynamic profiling according to claim 1, characterized in that, Before generating the simulated training evaluation result for the target trainee based on the real-time simulation result, the method further includes: Obtain real-time behavioral data corresponding to the target student, and determine the multimodal behavioral information corresponding to the target student based on the real-time behavioral data; Based on the multimodal behavioral information, an analysis operation is performed on the multimodal behavioral information based on the pre-determined temporal knowledge graph to obtain the multimodal evaluation result of the target student; The step of generating the simulated training evaluation results for the target trainee based on the real-time simulation results includes: Based on the real-time simulation results and the multimodal evaluation results, the simulated training evaluation results for the target trainees are generated.

3. The intelligent simulation training method based on dynamic profiling according to claim 1 or 2, characterized in that, The step of generating dynamic profile information for the target trainee based on the evaluation analysis results and the basic profile information includes: Based on the evaluation and analysis results and the basic profile information, the target interaction data of the target learner is determined, wherein the target interaction data includes text interaction data, emotional interaction data, and temporal interaction data; Extract the interaction feature data corresponding to the target interaction data, and determine the multi-dimensional quantization result corresponding to the target student based on the interaction feature data and a pre-determined multi-dimensional quantization algorithm. Based on the multi-dimensional quantification results, dynamic profile information of the target student is generated.

4. The intelligent simulation training method based on dynamic profiling according to claim 2, characterized in that, The method further includes: Based on the simulated training evaluation results of the target trainees, determine whether the simulated training evaluation results meet the preset simulated training conditions; When it is determined that the simulation training evaluation result does not meet the preset simulation training conditions, analyze the target reasons corresponding to the simulation training evaluation result not meeting the preset simulation training conditions; Based on the stated cause, a target matching strategy corresponding to the stated cause is determined from a pre-defined matching strategy library, and the simulation training parameters corresponding to the target student are updated based on the stated target matching strategy.

5. The intelligent simulation training method based on dynamic profiling according to claim 4, characterized in that, The step of determining the target matching strategy corresponding to the target cause from a pre-determined matching strategy library based on the target cause includes: Obtain the development goal information corresponding to the target student, and determine the simulation training path information corresponding to the target student based on the goal reasons and the development goal information; Based on the simulated training path information, a capability development path map corresponding to the target student is generated, and based on the capability development path map and the target reasons, the demand strategy information corresponding to the target student is determined. Based on the demand strategy information, a target matching strategy corresponding to the target reason is determined from a pre-determined matching strategy library.

6. The intelligent simulation training method based on dynamic profiling according to claim 1, characterized in that, After generating the dynamic profile information of the target student based on the evaluation analysis results and the basic profile information, the method further includes: Obtain the historical training data corresponding to the target student, and determine the historical evaluation results of the target student based on the historical training data; Based on the historical evaluation results, a training trend curve corresponding to the target student is generated, and based on the training trend curve, historical profile information of the target student is generated. Based on the historical profile information, update the dynamic profile information of the target student.

7. The intelligent simulation training method based on dynamic profiling according to claim 1, characterized in that, The step of generating the simulated training evaluation results for the target trainee based on the real-time simulation results includes: Extract multimodal feature data from the real-time simulation results, wherein the multimodal feature data includes text feature data, speech feature data, and time-series feature data; Based on the multimodal feature data and the pre-determined multimodal fusion model, the fusion feature data of the target student is determined; Based on the fused feature data and the pre-determined weighted aggregation algorithm, the quantitative evaluation data of the target student is calculated; Based on the quantitative assessment data, the simulated training assessment results for the target trainees are generated.

8. An intelligent simulation training device based on dynamic imagery, characterized in that, The device includes: The acquisition module is used to obtain basic profile information of the target students. The generation module is used to generate the practical training requirements information corresponding to the target trainee based on the basic profile information and the pre-determined temporal knowledge graph. The acquisition module is also used to acquire the real-time interaction information of the target student; The determining module is used to determine the evaluation analysis result corresponding to the target student based on the real-time interaction information and the basic profile information, wherein the evaluation analysis result includes the simulated interaction analysis result of the target student; The generation module is further configured to generate dynamic profile information of the target trainee based on the evaluation analysis results and the basic profile information; and generate simulation training parameters corresponding to the target trainee based on the dynamic profile information. The simulation module is used to perform real-time simulation operations on the target student based on the simulation training parameters, and obtain the real-time simulation results corresponding to the target student. The generation module is also used to generate the simulated training evaluation results of the target trainees based on the real-time simulation results.

9. An intelligent simulation training device based on dynamic imagery, characterized in that, The device includes: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the intelligent simulation training method based on dynamic portrait as described in any one of claims 1-7.

10. A computer storage medium, characterized in that, The computer storage medium stores computer instructions, which, when invoked, are used to execute the intelligent simulation training method based on dynamic portrait as described in any one of claims 1-7.

Images