57 results about "Training quality" patented technology

The invention discloses a quality-effect index and influence factor correlation analysis model construction method and system, and the method comprises the steps: carrying out the preprocessing of an original analysis data set, and then carrying out the standardization processing, so as to obtain a standard analysis data set which comprises standard quality-effect indexes and standard influence factor data; dimensionality reduction processing is carried out on the standard influence factors to extract key influence factors, the correlation degree between the key influence factors and the standard quality-effect indexes is quantified, and an initial correlation analysis model is constructed based on the correlation degree; and performing iterative training on historical training data according to a machine learning algorithm, and dynamically optimizing model parameters in the initial correlation analysis model to obtain a trained correlation analysis model. The method is used for solving the technical problems of inaccurate quality and efficiency evaluation, fuzzy influence factor relevance and the like in a training scene, systematic analysis and scientific prediction of the training quality and efficiency are realized, and quantitative decision support is provided for training scheme optimization and quality improvement.

The invention provides a large model data set quality evaluation method and system. The method comprises the following steps: S1, constructing a large model data set quality evaluation framework based on file quality, file description quality, file content quality, data labeling quality and model training quality; s2, constructing a large model data quality evaluation index based on the constructed large model data set quality evaluation framework; and S3, acquiring a large model data set, classifying the large model data set according to modal types and single evaluation indexes, and performing corresponding quality evaluation on the classified data set. According to the method, the data set quality of the large model can be comprehensively and systematically evaluated, and the high-quality data set construction of the large model can be effectively guided, so that the training efficiency of the large model is improved, errors and deviations in the training process are reduced, and the accuracy and generalization ability of the model are improved.

The invention relates to the technical field of vehicles, in particular to a vehicle energy management strategy training method and device, equipment, a medium and a product, and the method comprises the steps: obtaining the operation data of a vehicle model of a target vehicle; training an intelligent agent based on reinforcement learning by using the operation data, the intelligent agent comprising a state space and an action space, the state space receiving the operation data, and the action space outputting an energy management strategy of the target vehicle; evaluating the energy management strategy by using a pre-constructed reward function, wherein the reward function comprises a multi-stage fuel consumption reward, an SOC stabilization reward and an extreme state constraint penalty reward; and optimizing the output of the action space based on the output value of the reward function, and stopping training until a preset condition is met to obtain a target energy management strategy. The problems that the training quality is poor and the practical deployment feasibility of the energy management strategy obtained through training is poor in the process of training the intelligent agent to output the energy management strategy in the related technology are solved.

The application provides a logistics warehouse picking training and motion optimization method and system based on virtual reality, and relates to the technical field of intelligent warehousing. The method constructs a virtual warehouse environment linked with a database in real time and generates a picking task guide; collects depth images, inertial measurement data and physiological signals, and fuses to generate whole body posture data of a trainee; identifies non-standard actions based on the posture data and calculates work efficiency; combines the carrying weight, physiological state and motion data to perform ergonomics scoring, fatigue accumulation analysis and injury risk prediction; dynamically adjusts the task difficulty and controls the training process according to the training state; and finally generates a training evaluation report through motion data comparison. The application can improve the quality of logistics picking training and reduce the risk of work injury.

The invention discloses a human body acupoint automatic positioning method, system and device based on deep learning and a storage medium. A body part area containing an acupoint is quickly positioned through a target detection model; utilizing a key point detection model to accurately detect anatomical key point coordinates in the region; innovatively, firstly, according to a preset traditional Chinese medicine acupoint positioning rule knowledge base, mathematical calculation is conducted on the key point coordinates, and initial acupoint coordinates are obtained; then a local image area with the initial coordinates as the center is intercepted, a lightweight self-adaptive fine tuning network is input, fine offset compensation is conducted on the initial coordinates according to local image features, and final high-precision acupoint coordinates are output. In addition, the invention further provides a double-model cooperative training process, weighted screening is carried out on training samples by utilizing a region detection model so as to improve the training quality of a key point detection model, and traditional Chinese medicine knowledge and a modern artificial intelligence technology are deeply fused so as to improve the training quality of the key point detection model. The method overcomes the defects of black box, high data dependence, poor generalization and the like of a pure data driven model, has the advantages of high positioning precision, high reliability, good interpretability and the like, and is suitable for multiple fields of traditional Chinese medicine clinic, teaching, health service and the like.

The application discloses a real-time screenshot splicing-based operation key step playback method in a practical training room, relates to the technical field of data processing, and balances integrity and storage economy by aligning desktop screenshots with multi-dimensional event timestamps and combining operation strength adaptive sampling, which not only completely captures operation details, but also avoids invalid screenshot waste; relying on screenshot splicing and incremental storage technology, a globally consistent operation context is constructed to solve the information loss problem under window switching and complex operation; through intelligent screening and atlas indexing of key steps, accurate positioning of operation steps is realized, and playback step-level time compression and event superposition presentation are matched to greatly improve the backtracking efficiency; the playback consistency verification and rollback correction mechanism guarantee the accuracy of reproduction, and finally, the operation review effect in practical training teaching is optimized, the students are helped to quickly master core skills, and the training quality is significantly improved.

The invention discloses a military crane chassis fault reproduction and practical training system based on working condition simulation, and relates to the technical field of military equipment practical training, and the system comprises an entity chassis building unit which selects and uses entity parts of subsystems of a military automobile crane to be arranged and installed according to a real automobile, retains a key relation, presets an interface, defines a safety threshold value and the like; the dual-mode fault generation unit constructs an initial library, and an intelligent agent is embedded to generate a fault; the multi-dimensional data acquisition unit responds to the trigger signal to acquire multiple types of data; the fault diagnosis practical operation unit is used for trainees to operate; the training evaluation and fault iteration unit generates a report and feeds back optimization suggestions; according to the invention, a dual-mode fault generation unit is constructed, an initial fault library and an intelligent agent exploration capability are integrated, a screening algorithm is matched, and multiple fault scenes are provided; a complete practical training system is built, multi-dimensional data acquisition is achieved, practical operation and evaluation iteration are standardized, the practical training quality is improved through closed-loop design, and the military crane chassis fault practical training requirements are met.

The present application relates to a kind of breathing training systems using six-word formula, including user end hardware and cloud server, using the VR head-mounted integrated machine of user end hardware and elastic fabric waistband, guide and detect the breathing training condition of user, realize the accurate classification of abdominal breathing, chest breathing, mixed breathing by means of EMG, displacement, triple fusion calculation mode of flow, utilize mouth shape identification auxiliary phoneme confidence correction, accurately identify different exhale mode of six-word formula, introduce mouth shape identification, phoneme model, physiological sensing fusion multi-modal scoring mechanism, compared with traditional single chest and abdominal displacement monitoring scheme or single audio identification scheme, accurately evaluate the training quality of user from many aspects and comprehensive score, it is convenient for user to understand the improvement direction and progress space of self breathing training.

The invention discloses a practical training room desktop operation key step playback method based on real-time screenshot splicing, relates to the technical field of data processing, and aims to completely capture operation details, avoid invalid screenshot waste and balance integrity and storage economy through alignment of desktop screenshots and multi-dimensional event timestamps and adaptive sampling in combination with operation intensity. On the basis of screenshot splicing and incremental storage technologies, a globally consistent operation context is constructed, and the problem of information missing under window switching and complex operation is solved; through intelligent screening of key steps and map indexing, accurate positioning of operation steps is realized, step-level time compression and event superposition presentation of a playback end are matched, and the backtracking efficiency is greatly improved; a playback consistency verification and rollback correction mechanism guarantees reproduction accuracy, and finally, operation reproduction and efficient backtracking experience are performed with low storage transmission cost and high integrity, an operation replay effect in practical training teaching is optimized, students are assisted to quickly master core skills, and training quality is remarkably improved.

The invention provides a method for constructing a data quality evaluation model, and belongs to the technical field of quality evaluation, and the method comprises the steps: collecting a to-be-evaluated data set, and obtaining a business demand and an application scene of the to-be-evaluated data set; determining an assessment index vector of the to-be-assessed data set based on the business demand and the application scene of the to-be-assessed data set; based on the to-be-evaluated data set and the evaluation index vector, determining a training quality feature matrix and a test quality feature matrix; and constructing a quality evaluation model based on the training set and the training quality feature matrix, and evaluating and optimizing quality evaluation performance based on the evaluation index vector, the test set and the test quality feature matrix. The method can enhance the capturing capability of the model for the data quality characteristics, achieves the high-precision evaluation of dynamic weighting, improves the intelligence and stability of the quality evaluation model, and improves the adaptability of the model in the quality management and analysis scenes of complex data.

The invention discloses an electric energy metering device field calibration skill training method, system, equipment and medium, and relates to the technical field of electric energy metering device calibration training, and the method comprises the steps: generating a fault training scheme and a training operation instruction according to a fault scheme template library based on a training demand; in response to the practical training operation instruction, carrying out practical training safety pre-test, and after the test is passed, starting multi-skill item operation according to a fault practical training scheme sequence; generating a training report based on the monitoring data of the multi-skill project operation process; according to the invention, through a whole training period from fault scheme adaptation, operation process guidance to result feedback, training quality improvement of skills of the trainee from operation to standardized and accurate operation is realized, and operation standardization and accuracy are improved while operation fluency and efficiency of the trainee are improved.

This invention relates to the field of talent training quality assessment technology in medical and pharmaceutical colleges, and discloses a big data closed-loop assessment system for talent training quality in medical and pharmaceutical colleges. The system includes: a data acquisition module to collect multi-source training data; a preprocessing and fusion module to construct a talent training quality dataset; a big data closed-loop assessment module containing three-level indicator construction, multi-dimensional assessment, and result analysis units to complete the assessment and generate personalized reports; an intelligent feedback module to push and visualize the assessment results; a training optimization module to adapt rigid constraints and transform them into optimization measures; and a system iteration module to ensure the assessment system is updated synchronously with the pharmaceutical industry. This invention constructs a full-link closed loop, solving problems such as assessment disconnect and low accuracy, improving the quality of talent training and its industry adaptability, and adapting to the characteristics of medical and pharmaceutical college training.

The invention discloses an in-patient physician guidance physician training quality evaluation system based on comprehensive evaluation, and relates to the technical field of physician guidance training. Theoretical knowledge of each in-patient physician after training is finished is tested, and a personalized and targeted simulation scene is designed for each in-patient physician for simulation test; meanwhile, the actual work condition of each resident physician is scored, then the training quality of each resident physician is evaluated, and a targeted secondary training scheme is provided for each resident physician, so that multi-dimensional analysis of the training quality of the resident physician is realized, the reliability and authenticity of the training evaluation result of the resident physician are guaranteed, and the training efficiency of the resident physician is improved. And the training result and the secondary training scheme are sent to the guidance physician, reference is provided for knowing the guidance direction of the physician, more professional training guidance is provided for the resident physician, and therefore the training effect of the resident physician is improved again.

The invention relates to a catenary equipment training system and a control method. Comprising a high-simulation hardware simulation module, an intelligent teacher machine management module, a multi-terminal student practical operation system module, a WebAPI data communication server module, an automatic evaluation system module, a central control and display system module and a data storage module in bidirectional data connection with the WebAPI data communication server module. According to the catenary equipment training system, 1: 1 high-simulation hardware is adopted to simulate a typical structure of a catenary, the operation process and mechanical feedback are truly restored, and the training immersion and the practical operation pertinence are improved; multi-terminal cooperative training is supported, multi-person and multi-project training can be carried out at the same time, and the training requirements of trainees of different levels are met in combination with switching of multiple practical training strategies; through the technical means of high-simulation hardware simulation, intelligent teacher management, multi-terminal collaborative practical operation, automatic evaluation, data visualization and the like, standardization, intelligence and safety of special repair training of the overhead line system are achieved, and the training efficiency and the talent training quality are improved.

The invention belongs to the technical field of power system employee training, and provides a learning path planning and learning effect evaluation method and system based on a neural network, electronic equipment and a storage medium. The method comprises the steps of data acquisition, coding feature matrix construction, knowledge graph entity embedding generation, vector splicing, coding processing, matching score matrix calculation, curriculum personalized screening, knowledge level evaluation, learning demand analysis, optimal learning path screening and learning effect evaluation. According to the invention, through a two-stage learning path planning process combining actual demands and employee historical data, a learning demand list and a recommended course sequence are enabled to accurately fit employee knowledge shortages and business demands, and the path can be adjusted in real time according to the employee's own learning progress; in addition, due to the integration of the power grid business data, the training content hysteresis is avoided, and the training quality is improved.

The application discloses a client training method and device based on layered federated learning, and equipment. The method comprises the following steps: according to the estimated context reward between each client in the overlapping area and each overlapping edge server, the edge servers in the overlapping area are allocated to the clients to determine the client set in each edge server; according to the estimated context reward between each client in the client set and the corresponding edge server and the training quality of each client, the client participating in the training is determined, and the client participating in the training is trained according to the received training model; and the estimated context reward and the training quality between the client and the corresponding edge server are updated according to the training result of the client. The method effectively solves the attribution problem of the clients in the overlapping area of the edge servers, and designs a reasonable selection mechanism for the edge servers when training the clients, so that the training time and the training quality of the clients can be effectively guaranteed.

The application discloses a model training method and a virtual coach guidance strategy generation method, and relates to the technical field of computers. The method comprises the following steps: acquiring a current state of a user after the user performs a current training action; wherein the current state comprises any multiple items of an action quality score, a fatigue degree score, a self-satisfaction score, an understanding degree score and sensor data of a sports device; inputting the current state into a DQN model to obtain Q values of each strategy in a preset action space; selecting a current strategy from the action space based on the Q values of each strategy; performing the current strategy; acquiring a next state of the user after the user performs a next training action in response to the execution of the current strategy; and adjusting parameters of the DQN model based on the current state, the next state and the current strategy. The embodiment can provide more targeted training guidance for the user and improve the training quality of the user.

The application discloses a full-movement simulator training data collection and sharing system applying a blockchain technology, which comprises a data collection system and a data sharing system, wherein the data collection system is connected with the data sharing system, the data collection system comprises an attendance module, a course selection module, a simulator, and a blockchain module, the simulator comprises a simulator training module and a training data collection module, and the attendance module is connected with the course selection module. The weak trust problem of participants in simulator training is integrated, valuable data in training is collected and managed, the training subjects of pilots are recorded, the logs and operations of various conditions in simulator training are recorded, the flight data of pilots during training is acquired, preparation for training quality analysis and more effective training is made, the supervision of the CAAC supervision department and the record of the flight subjects of pilots are facilitated, and a complete data collection and sharing scheme is provided for all parties.

The utility model relates to the technical field of feedback training terminals, and discloses a train driving operation touch feedback training terminal which comprises an operation table, and a test frame is installed on the left side of the operation table. According to the train driving operation touch feedback training terminal, when the device carries out training feedback, a testing frame is installed on the left side of an operation table, sliding rails are arranged on the left side and the right side of the testing frame, a lead screw and a limiting rod are installed in the sliding rails on the two sides respectively, and sliding rods capable of sliding back and forth are installed outside the lead screw and the limiting rod; the sliding rod drives the top plate at the top to move back and forth, and the display is mounted on the top surface of the top plate, so that the display can be driven to move back and forth, the effect that the display moves back and forth to simulate a scene is achieved, and the problems that the real scene condition cannot be simulated more appropriately and the temporary strain capacity of a driver cannot be detected and trained accurately are avoided. Therefore, the training quality is reduced and the accuracy of the result fed back by the feedback terminal is reduced.

The invention discloses a large model pre-training data screening method and system based on conformal inference, terminal equipment and a medium, and relates to the technical field of model training data screening. The method comprises the following steps: collecting original unlabeled data to construct a target data set; selecting a part of samples from the target data set, and constructing a calibration set containing real quality labels with high-quality or low-quality labeled samples; calculating uncertainty scores of low-quality samples of the calibration set and to-be-screened samples of the target data set by using a pre-training quality classifier; calculating a conformal test value of each to-be-screened sample based on the two types of scores; and determining a test value condition through the dynamic threshold, and screening the to-be-screened sample meeting the condition as a high-quality sample. Statistical theory support is provided based on conformal inference, the proportion of low-quality samples after training data screening is controllable, the quality and the screening amount of the training data can be balanced, the obtaining cost of high-quality training data is reduced, and the large model pre-training requirement is met.

The application discloses a teaching evaluation method and system based on multi-agent collaborative evaluation closed loop, and belongs to the technical field of education informatization and artificial intelligence. The method comprises the following steps: collecting multi-source evaluation data of school teachers, enterprise tutors and student self-evaluation and mutual evaluation; constructing a multi-dimensional index system comprising theoretical knowledge, industrial practical ability, team cooperation ability and innovative thinking; adopting a fuzzy comprehensive evaluation method and a D-S evidence theory cascade two-stage fusion algorithm to fuse data, and dynamically adjusting the weights of each evaluation subject; establishing a correlation matrix of knowledge points and evaluation indexes, triggering an optimization suggestion generation engine when the evaluation result is lower than a threshold value, retrieving a course content optimization scheme from a knowledge graph / case library and pushing the scheme; iteratively executing the above steps to drive continuous optimization of course content, practical projects and training modes; and adopting a reinforcement learning algorithm to self-learn and optimize strategies. The application realizes the connection of evaluation and industrial demand and the self-adaptive evolution of the teaching system, and improves the talent training quality.

The invention discloses a big language model training method and device, an information processing method and device and electronic equipment. The method comprises the steps of obtaining a historical dialogue data set of customer services in a target field; a client role library is constructed based on the historical dialogue data set through the first large language model, a training data set is generated based on the client role library, the training data set comprises training data in multiple dialogue scenes, and the dialogue scenes in the multiple dialogue scenes comprise different client roles; a to-be-trained large language model is trained according to the training data set, a target large language model is obtained, the model performance of the first large language model is better than that of the to-be-trained large language model, and the target large language model is used for simulating the virtual customer role to output dialogue information with the to-be-trained customer service. According to the method and the device, the technical problem of relatively low training quality caused by lack of personalization and dynamism when customer service training is carried out according to standardized cases in related technologies is solved.

The application provides a ship electromechanical system operation training evaluation method and device based on a three-dimensional environment, and relates to the field of ship electromechanical system simulation. The method comprises the following steps: obtaining a simulation data set corresponding to a ship electromechanical system in a three-dimensional virtual environment; obtaining operation interaction data in different types of ship training consoles; preprocessing the simulation data set, and constructing a plurality of evaluation indexes based on the preprocessed simulation data set; inputting the operation interaction data into the multi-evaluation index model to output an index score corresponding to each evaluation index; inputting the index score into a comprehensive training evaluation model to output a training quality score corresponding to a target trainee; and outputting a training quality evaluation result of the target trainee based on the training quality score. The application solves the problem that the traditional ship operation training evaluation system lacks a coupling computer mechanism between operation behavior and physical simulation response, so that the evaluation result cannot reflect the comprehensive ability of the trainee to control the system.

The invention relates to the technical field of biological rehabilitation engineering and intelligent medical equipment, and discloses a children oral and facial muscle training compliance monitoring system based on multi-modal sensing, which is used for solving the problems of subjectivity, poor compliance and difficulty in remote monitoring of traditional training. The system comprises a wearable multi-mode sensing unit for acquiring surface myoelectricity, occlusal force, inertial measurement and image data; the AI recognition and compliance evaluation module processes data through a deep learning model to realize action recognition, multi-dimensional compliance scoring and personalized self-learning so as to objectively quantify the training quality; the feedback interaction module provides real-time guidance and instant feedback for the user to improve the degree of participation; and the remote management module is used for doctors to monitor data, adjust schemes and obtain analysis reports. Through objective and quantitative data analysis and intelligent interaction, the standardization level and interestingness of child oral and facial muscle training are effectively improved, efficient remote monitoring and personalized intervention are achieved, and the rehabilitation effect is remarkably improved.

The application provides a hospital information integration platform management method and system based on big data, and the application solves the residual image problem caused by the soft deletion only updating the state in the traditional CDC synchronization by constructing the data full life cycle association through the transaction intention log; the introduction of the dynamic confidence cleaning realizes the cleaning of the invalid data generated by the misoperation, meanwhile, the deletion records which need to be audited and traced in the normal business process are reserved, and the data quality and compliance are balanced; compared with the mode of manually writing filtering rules or fixed threshold judgment in the prior art, the method can adapt to the business scene change, reduce the processing burden of the downstream system, improve the data analysis accuracy and the AI model training quality, and reduce the business logic error risk caused by the missed filtering conditions.

The invention relates to the technical field of driving training cheating detection, and discloses a driving training cheating detection system based on somatosensory interaction, which comprises a multi-dimensional acquisition module and an intelligent detection module. According to the system, management data, driving training data and vehicle environment detection data of all trainees are acquired through a multi-dimensional acquisition module and are classified to form a data set, and an intelligent detection module judges the training progress of the trainees, evaluates whether identity information filled in by the trainees on site and biological information input by the trainees are correct or not, generates corresponding authentication scores and sends the authentication scores to a server. The intelligent detection module detects cheating behaviors in the driving training process of trainees in real time, generates corresponding cheating indexes and synchronous scores, strengthens the supervision of the training process, and improves the driving safety of the trainees. Through reference instruction information and reference positioning of adjacent traffic lights, the cheating behavior of counterfeiting GPS positioning is effectively prevented, and the examination fair training quality is high.

The invention relates to a practical training evaluation method and device, computer equipment and a storage medium. The method comprises the following steps: acquiring an attitude sequence of an operator in a current practical training process; wherein the attitude sequence comprises a preset number of three-dimensional attitude models of continuous time frames; determining a shape difference sequence of the attitude sequence according to the attitude sequence and the reference attitude of the attitude sequence; wherein the shape difference sequence comprises a shape difference operator corresponding to each time frame, and the shape difference operator corresponding to each time frame represents the shape difference between the three-dimensional attitude model of the time frame and the reference attitude; determining a target action category of the attitude sequence according to the shape difference sequence; according to the target action category and the action category mapping relation, evaluating the practical training process of the operator; wherein the action category mapping relation comprises action descriptions associated with different action categories. By adopting the method, real-time and objective practical training examination and evaluation can be realized, and the training quality and the operation safety are improved.

The invention relates to a respiratory training system using a six-character table, which comprises user-side hardware and a cloud server, a VR head display all-in-one machine using the user-side hardware and an elastic fabric waistband, guiding and detecting the respiratory training condition of a user, and calculating the respiratory training of the user by means of a triple fusion calculation mode of EMG, displacement and flow. The accurate classification of abdominal respiration, chest respiration and mixed respiration is realized, the mouth shape recognition is utilized to assist in phoneme confidence correction, different exhaling modes of a six-character table are accurately recognized, a multi-modal scoring mechanism integrating mouth shape recognition, a phoneme model and physiological sensing is introduced, and compared with a traditional single thoracic and abdominal displacement monitoring scheme or a single audio recognition scheme, the method has the advantages that the accuracy is high, and the accuracy is high. The training quality of the user is accurately evaluated from multiple aspects and comprehensive scores, and the user can conveniently know the improvement direction and progress space of breathing training of the user.

The invention provides an in-hospital information integration platform management method and system based on big data, and the method and system achieve the construction of data full life cycle association through a transaction intention log, and solve a ghosting problem caused by only updating a state through soft deletion in conventional CDC synchronization. Due to the introduction of dynamic cleaning confidence, invalid data generated by misoperation is cleaned, deletion records needing to be audited and traced in a normal business process are reserved, and the data quality and compliance are balanced; compared with the prior art in which filtering rules need to be written manually or fixed threshold judgment is needed, the method can adapt to business scene changes, reduce the processing burden of a downstream system, improve the data analysis accuracy and the AI model training quality, and reduce the risk of business logic errors caused by filtering condition omission at the same time.