Artificial intelligence-based lecturer function grading method and system and storage medium

Abstract

The invention relates to the technical field of artificial intelligence, and particularly discloses a lecturer function grading method and system based on artificial intelligence and a storage medium, and the method comprises the steps: collecting the training information of a lecturer; analyzing the training information, and converting the training information into quantifiable index data; performing standardization processing on the quantifiable index data to form a lecturer feature vector set; and inputting the lecturer feature vector set into a pre-trained artificial intelligence evaluation model, and outputting a function level prediction result corresponding to the lecturer. The lecturer function grading method based on artificial intelligence is used for intelligently grading the lecturers, a more scientific, efficient and fair evaluation system can be provided, the teaching quality of the lecturers can be improved, professional growth of the lecturers can be promoted, and enterprise education resource allocation can be optimized.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an artificial intelligence-based lecturer function grading method, system and storage medium. Background Art

[0002] As companies pay more attention to employee competency training, they are conducting more and more training activities for their employees. In corporate training, the teaching quality of lecturers is of great significance to the improvement of employee competency. It can be seen that the evaluation of lecturers' functions is also an important part of corporate training. Accurately evaluating lecturers' teaching abilities by establishing a complete teaching quality evaluation system will help improve lecturers' teaching levels. Therefore, how to efficiently and accurately evaluate lecturers' abilities and suitability levels has become an important topic. Traditional manual evaluation methods have problems such as strong subjectivity and low efficiency, and are difficult to adapt to the needs of functional rating in large-scale training. Therefore, it is necessary to introduce intelligent technology to achieve objective and efficient grading of lecturer functions. Summary of the invention

[0003] Based on this, it is necessary to provide an artificial intelligence-based lecturer function grading method, system and storage medium to address the problems of strong subjectivity and low efficiency in the traditional manual evaluation method for lecturer function grading.

[0004] An artificial intelligence-based lecturer function grading method comprises collecting lecturer training information; analyzing the training information and converting the training information into quantifiable indicator data; standardizing the quantifiable indicator data to form a lecturer feature vector set; inputting the lecturer feature vector set into a pre-trained artificial intelligence evaluation model and outputting a prediction result of the function level corresponding to the lecturer.

[0005] In one embodiment, the collecting of the lecturer's training information includes obtaining the lecturer's teaching information based on the lecturer's training records; obtaining the trainees' evaluation information on the lecturer after the training; and obtaining the lecturer's teaching effectiveness information based on the trainees' career development after the training.

[0006] In one embodiment, the teaching information includes a lecture video of the lecturer, and the analyzing the training information and converting the training information into quantifiable indicator data includes converting the audio data in the lecture video into text data; analyzing the text data using a general natural language model to determine the language expression quality data of the lecturer; and analyzing the non-verbal behavior of the lecturer in the image data of the lecture video using computer vision technology to determine the behavior expression quality data of the lecturer.

[0007] In one of the embodiments, analyzing the training information and converting the training information into quantifiable indicator data further comprises using statistical analysis methods to extract quantifiable information from the evaluation information and determine the teaching evaluation data of the lecturer.

[0008] In one embodiment, the method for constructing the artificial intelligence evaluation model includes constructing an artificial intelligence evaluation model based on a machine learning algorithm; training the artificial intelligence evaluation model using a labeled historical lecturer function level data set; during the training process, using a cross-validation method to process the feature vector input into the artificial intelligence evaluation model, using a model ensemble learning method combined with the prediction results of multiple models to evaluate the performance of the artificial intelligence evaluation model, and adjusting the model parameters of the artificial intelligence evaluation model.

[0009] In one of the embodiments, after inputting the lecturer feature vector set into a pre-trained artificial intelligence evaluation model and outputting a prediction result of the functional level corresponding to the lecturer, the method further includes outputting an evaluation report for the lecturer based on the functional level prediction result; the evaluation report includes but is not limited to functional level, scores of various abilities, strengths analysis, improvement point analysis, and comparative analysis with reference standards.

[0010] In one of the embodiments, after inputting the lecturer feature vector set into a pre-trained artificial intelligence evaluation model and outputting the functional level prediction result corresponding to the lecturer, the method further includes adjusting the model parameters of the artificial intelligence evaluation model according to the matching degree between the functional level prediction result and the actual evaluation result of the lecturer; or adding the actual evaluation result of the lecturer to the historical lecturer functional level data set, and retraining the artificial intelligence evaluation model using the new historical lecturer functional level data set.

[0011] An artificial intelligence-based lecturer function grading system comprises a data acquisition module for collecting lecturer training information; a data analysis module connected to the data acquisition module for analyzing the training information and converting the training information into quantifiable indicator data; a data processing module connected to the data analysis module for standardizing the quantifiable indicator data to form a lecturer feature vector set; and a function grading module connected to the data processing module for inputting the lecturer feature vector set into a pre-trained artificial intelligence evaluation model and outputting a prediction result of the function level corresponding to the lecturer.

[0012] A computer device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the lecturer function grading method based on artificial intelligence described in any one of the above embodiments are implemented.

[0013] A computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the steps of the method for grading lecturer functions based on artificial intelligence described in any one of the above embodiments are implemented.

[0014] The above-mentioned AI-based lecturer function grading method analyzes the lecturer's training information during the training process, converts the training information into quantifiable indicator data, and standardizes the quantifiable indicator data to form a lecturer feature vector set. The lecturer feature vector set is input into the pre-trained AI evaluation model to output the prediction result of the lecturer's corresponding function level. Using the AI-based lecturer function grading method to intelligently grade lecturers can provide a more scientific, efficient and fair evaluation system, which is not only conducive to improving the teaching quality of lecturers, but also promotes the professional growth of lecturers and optimizes the allocation of corporate education resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the implementation methods of this specification or the technical solutions in the prior art, the drawings required for use in the implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some implementation methods recorded in this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0016] Figure 1 A schematic diagram of an application environment of a method for grading lecturer functions based on artificial intelligence in one of the embodiments of the present application;

[0017] Figure 2 A flowchart of a method for grading lecturer functions based on artificial intelligence in one of the embodiments of the present application;

[0018] Figure 3 This is a flow chart of a method for collecting training information of a lecturer in one of the embodiments of the present application;

[0019] Figure 4 This is a flow chart of a method for analyzing training information in one of the embodiments of the present application;

[0020] Figure 5 This is a schematic diagram of a method flow for constructing an artificial intelligence evaluation model in one of the embodiments of the present application;

[0021] Figure 6 This is a structural diagram of a lecturer function grading system based on artificial intelligence in one of the embodiments of the present application;

[0022] Figure 7 This is a diagram of the internal structure of a computer device in one of the embodiments of the present application. DETAILED DESCRIPTION

[0023] In order to facilitate the understanding of the present invention, the present invention will be described more fully below with reference to the relevant drawings. The preferred embodiments of the present invention are given in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to understand the disclosure of the present invention more thoroughly and comprehensively.

[0024] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0025] Figure 1 FIG. 1 is a schematic diagram of an application environment of a method for grading lecturers' functions based on artificial intelligence in one embodiment of the present application. The method for grading lecturers' functions based on artificial intelligence provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. Among them, the terminal 102 can be but is not limited to various personal computers, laptops, smart phones, tablets, Internet of Things devices and portable wearable devices. The Internet of Things devices can be smart speakers, smart TVs, smart air conditioners, smart car-mounted devices, etc. Portable wearable devices can be smart watches, smart bracelets, head-mounted devices, etc. The server 104 can be implemented with an independent server or a server cluster consisting of multiple servers.

[0026] It should also be noted that the relevant information (including but not limited to instructor training information, instructor personal information, etc.) and data (including but not limited to data used for display, data for analysis, etc.) involved in this disclosure are all information and data authorized by the user or fully authorized by all parties.

[0027] Figure 2 This is a flow chart of a method for grading a lecturer's function based on artificial intelligence in one embodiment of the present application. In one embodiment, Figure 2 As shown in the figure, a method for grading lecturer functions based on artificial intelligence is provided. Figure 1 Taking the application environment in as an example, the following steps S100 to S400 may be included.

[0028] Step S100: Collecting the training information of the instructor.

[0029] Step S200: Analyze the training information and convert the training information into quantifiable indicator data.

[0030] Step S300: Standardize the quantifiable indicator data to form a lecturer feature vector set.

[0031] Step S400: Input the lecturer feature vector set into a pre-trained artificial intelligence evaluation model, and output the prediction result of the lecturer's corresponding functional level.

[0032] Automatically obtain lecturer-related data from multiple channels. For example, you can collect lecturer's teaching videos, courseware quality, interaction records, student evaluations, test score changes before and after training, and other multi-source data from the enterprise training platform. The above information can be stored in Figure 1 The server 104 shown may also be placed on a cloud or other network server.

[0033] Extract key features that can reflect the instructor's teaching ability from the collected information, and convert relevant information into quantifiable indicators. For example, extract the instructor's explanation clarity and interactive activity from the teaching video, extract the instructor's knowledge transfer efficiency from the changes in students' test scores before and after training, and extract students' satisfaction with the instructor from students' feedback information.

[0034] Process multi-source data and construct a lecturer feature vector set. Input the lecturer feature vector set into the artificial intelligence evaluation model, use the artificial intelligence evaluation model to identify the characteristic patterns of lecturers at different functional levels, realize automatic classification, and output the prediction results of the lecturer's functional level. For example, lecturers are automatically classified into preset functional levels, such as junior, intermediate, and senior lecturers, and a detailed evaluation report is generated. The artificial intelligence evaluation model can be a pre-built and trained model based on deep learning or machine learning.

[0035] The embodiment of the present application provides a lecturer function grading method based on artificial intelligence, which analyzes the lecturer's ability from multiple dimensions of multiple lecturer training, and converts the relevant information into quantifiable indicators to construct a lecturer feature vector set that can be used to reflect the lecturer's teaching ability. Based on the lecturer feature vector set, the AI ​​(Artificial Intelligence) model is used to automatically assess the lecturer's functional level to improve the objectivity and accuracy of the assessment. The use of AI intelligence to achieve lecturer grading can provide a more scientific, efficient and fair evaluation system, which is not only conducive to improving teaching quality, but also promotes the professional growth of lecturers and optimizes the allocation of educational resources.

[0036] AI conducts evaluations based on preset standards and actual teaching performance data. Through data analysis, it can remove the influence of human emotional factors, provide more objective and accurate evaluation results, and reduce the bias of subjective judgment. Standardized evaluation processes and transparent data analysis can help ensure that each lecturer is evaluated under the same standards, enhancing the fairness and transparency of the evaluation system. In addition, in the application scenario of large-scale enterprise training, the above method can efficiently handle the evaluation of many lecturers, reduce the burden on the human resources department, and achieve a fast, unified and standardized grading process, which is especially suitable for large-scale enterprise training platforms.

[0037] Figure 3 This is a flowchart of a method for collecting lecturer training information in one of the embodiments of the present application. In one of the embodiments, collecting lecturer training information may include the following steps S110 to S130.

[0038] Step S110: Acquire the lecturer's teaching information according to the lecturer's training record.

[0039] Enterprises can conduct training activities for employees through the enterprise training system of the enterprise online school platform, and the enterprise training system can record the training status of the lecturer. Therefore, by using the training records of various activities in the enterprise training process stored in the enterprise online learning platform, various teaching information related to the lecturer can be automatically captured. For example, teaching information can include information such as courseware published by the lecturer, course schedule, list of students and their participation status, and course schedule can include information such as the number of trainings arranged by the lecturer, training time, and training topics.

[0040] Step S120: After the training, obtain the trainees' evaluation information on the instructor.

[0041] After each training session, students' evaluation information on the instructor can be obtained through multiple channels such as questionnaires, online forums, and social media comments. In a specific implementation, an electronic questionnaire system can be set up in the enterprise online learning platform to automatically send electronic questionnaires to trainees after each training session to collect quantitative scores and open feedback or evaluation information on the instructor's explanation clarity, classroom interaction, and practical guidance effects.

[0042] Step S130: Obtain the teaching effectiveness information of the instructor based on the career development of the trainees after the training.

[0043] By connecting to the corporate training system, the instructor's past training records, the comparison of trainees' skill test scores before and after training, and the long-term tracking of trainees' career development information in the corporate training system can be obtained as the basis for evaluating the instructor's teaching effectiveness. The trainees' career development status can be obtained by regularly issuing questionnaires to trainees and conducting regular skill tests.

[0044] The above-mentioned AI-based lecturer function grading method collects lecturer training information from multiple sources. AI can integrate and analyze a large amount of teaching data, including student feedback, teaching results, classroom interaction and other multi-dimensional information, thereby providing a more comprehensive assessment of lecturer capabilities from a lecturer performance dimension, rather than just based on a single test score or human observation.

[0045] Figure 4 This is a flow chart of a method for analyzing training information in one of the embodiments of the present application. In one of the embodiments, analyzing the training information and converting the training information into quantifiable indicator data may include the following steps S210 to S230.

[0046] Step S210: Convert the audio data in the teaching video into text data.

[0047] Use video processing software or tools to extract audio data from video files, or directly use the API interface to obtain audio from lecture videos. Pre-process the extracted audio, such as noise reduction, volume normalization, background noise removal, and speech clarity enhancement, to improve speech recognition. Use speech recognition technology to convert audio data into text data, or you can directly use the speech recognition API service. Post-process the text data generated based on the audio data, such as spelling correction, grammar correction, punctuation addition, and other operations.

[0048] Step S220: Analyze the text data using a general natural language model to determine the lecturer's language expression quality data.

[0049] Use a pre-trained natural language model to evaluate the semantic coherence and logic of text data, use the model's context-awareness to determine whether the term is used in the correct context, and use a generative model to generate similar text fragments and compare them with the original text to see whether the generated content is logical and professional. For example, the coherence score, term accuracy score, etc., are used to quantify the text data, thereby determining the lecturer's language expression quality data. In this embodiment, the language expression quality data may include data on dimensions such as the logic of the lecturer's language expression and the accuracy of the use of professional terms.

[0050] When evaluating the quality of the lecturer's language expression, the technology in the field of natural language processing is used to achieve this. In a specific factual way, the pre-trained BERT model can be used to evaluate the lecturer's language expression quality data. The BERT model has a strong context-aware ability and can accurately determine whether the term is used in the correct context, thereby effectively evaluating the semantic coherence and logic of the text data.

[0051] Furthermore, a professional dictionary can be constructed. The professional dictionary combines the vector space characteristics of the BERT model to compare and analyze the terms in the text data, calculate the accuracy of the term usage, and improve the accuracy of the model evaluation. The use of professional dictionaries not only improves the professionalism of the evaluation, but also provides strong support for the quantitative analysis of the quality of the lecturer's language expression.

[0052] When generating a model, you can select the GPT series model to generate similar text fragments and compare and analyze them with the original text. By calculating the logical coherence and semantic similarity between the generated content and the original text, we further verified the accuracy and professionalism of the lecturer's language expression.

[0053] Step S230: using computer vision technology to analyze the lecturer's non-verbal behavior in the image data of the lecture video to determine the lecturer's behavior expression quality data.

[0054] In this embodiment, non-verbal behavior can be body language, facial expressions and other behaviors. Through graphics processing technology, the video quality can be improved so that the lecturer's facial expressions and body movements can be clearly analyzed. Face recognition technology is used to locate the lecturer's face in the image data, facial expression recognition technology is used to identify the lecturer's facial expressions, such as smiling, frowning, etc., and human body posture estimation technology is used to capture the lecturer's body posture and movement. Facial key points (such as the position of eyes, nose, and mouth) are extracted to analyze the changes in expression, and the position information of body joints is extracted to analyze the lecturer's movements and postures.

[0055] In some specific implementations, a pre-trained sentiment analysis model can be used to identify the emotional state conveyed by the lecturer's facial expressions, and a behavior analysis model can be used to identify specific behavior patterns. Based on the lecturer's behavior patterns, the lecturer's teaching attitude can be evaluated. That is, by analyzing the teaching video, the lecturer's behavior expression quality data can be evaluated based on the lecturer's movements, expressions, etc. in the video data. The behavior expression quality data can include data on dimensions such as teaching enthusiasm and the ability to create a classroom atmosphere.

[0056] In one embodiment, analyzing the training information and converting the training information into quantifiable indicator data may include the following steps: using statistical analysis methods to extract quantifiable information from the evaluation information and determine the lecturer's teaching evaluation data.

[0057] After obtaining the students' evaluation information about the lecturer, the evaluation information can be pre-processed for subsequent analysis. The lecturer's teaching evaluation data can be determined by extracting key opinions, keywords, and features from the students' evaluation information and converting them into quantifiable indicators. For example, statistics can be collected on the satisfaction scores of each student for the lecturer, the frequency of improvement suggestions, and other data, or the students' opinions can be classified into different categories, such as "rich content", "lack of interaction", "equipment problems", etc., and the number of opinions in each category can be counted.

[0058] In one embodiment, standardizing the quantifiable indicator data to form a lecturer feature vector set may include the following steps: performing data cleaning on the quantifiable indicator data collected from various sources, such as denoising, unifying the text format, filling missing values, etc. Performing data standardization on each quantifiable indicator data, such as unifying the data format, normalizing, encoding, etc. Extracting features such as text features, behavioral features, and performance features from each quantifiable indicator data. Using feature selection technology to screen out more relevant features and integrate the screened features, combining the features of each lecturer into a vector, each element in the vector corresponds to the quantitative value of a feature, so as to construct a lecturer feature vector set that can be used to evaluate the lecturer's ability.

[0059] Figure 5 This is a flow chart of a method for constructing an artificial intelligence evaluation model in one embodiment of the present application. In one embodiment, the method for constructing an artificial intelligence evaluation model may include the following steps S10 to S30:

[0060] Step S10: Construct an artificial intelligence evaluation model based on a machine learning algorithm.

[0061] According to the processing method when evaluating the lecturer's ability, a suitable machine learning algorithm is selected to build an artificial intelligence evaluation model by comprehensively considering factors such as algorithm complexity, computational efficiency, and interpretability. In this embodiment, the machine learning algorithm used may include but is not limited to a support vector machine algorithm, a random forest algorithm, or a deep neural network algorithm.

[0062] Step S20: Use the labeled historical lecturer function level dataset to train the artificial intelligence evaluation model.

[0063] Divide the labeled historical lecturer function level dataset. For example, divide the dataset into a training set, a validation set, and a test set. The training set can be used to train the model, the validation set can be used to adjust the model parameters (such as hyperparameter tuning), and the test set can be used to evaluate the final performance of the model. After setting the training parameters such as the loss function, optimization algorithm, number of training rounds, and batch size, the selected model can be trained based on the dataset.

[0064] Step S30: During the training process, the feature vector of the input artificial intelligence evaluation model is processed using the cross-validation method, the performance of the artificial intelligence evaluation model is evaluated using the model ensemble learning method combined with the prediction results of multiple models, and the model parameters of the artificial intelligence evaluation model are adjusted.

[0065] When training the AI ​​evaluation model using the labeled historical lecturer function level dataset, the training set can be used to train the selected model. The validation set is used to adjust the model's hyperparameters to avoid overfitting. The cross-validation method is applied to the feature vector. During the training of the machine learning model, cross-validation is performed on the dataset containing all features, and based on the results of the cross-validation, the model parameters are adjusted or different types of models are selected to improve performance, so as to optimize the classification accuracy and recall rate, and ensure that the model's prediction results for the lecturer function level are both accurate and stable.

[0066] During the training process, you can also introduce ensemble learning strategies (such as Bagging, Boosting, and Stacking) to combine the prediction results of multiple sub-models to obtain better prediction results than a single model, so as to improve the robustness of the final classification decision. Specifically, evaluate the performance of the ensemble model on the validation set, and adjust the model parameters or select different weak learners based on the evaluation results. When the model's performance on the validation set reaches a satisfactory level, make a final evaluation on the test set. The use of ensemble learning strategies can effectively improve the prediction performance of artificial intelligence evaluation models.

[0067] In some other embodiments, other methods may be used to improve the robustness of the model to enhance the model's ability to maintain stable and accurate operation in the face of anomalies, interference, parameter changes, or changes in the external environment.

[0068] In the data preprocessing and cleaning stage, the model's adaptability to different data distributions is improved by removing noise, processing missing values, and standardizing data. In feature selection, carefully select and construct features to retain only the most useful information for model prediction to reduce the interference of irrelevant factors and enhance the generalization and robustness of the model. In the model integration stage, the model ensemble learning method is used to combine the prediction results of multiple models to reduce the bias and variance of a single model and improve the stability of the overall prediction. By applying regularization technology when training the model, overfitting can be prevented, making the model more robust and less susceptible to noise in the training data.

[0069] Adding adversarial examples to the training process can enhance the model's ability to resist malicious inputs and improve the model's performance in the face of unseen attacks or noise. Carefully adjusting the model's hyperparameters through methods such as grid search, random search, or Bayesian optimization can find a better configuration and improve the model's robustness.

[0070] When designing the AI ​​evaluation model, an anomaly detection module is built into the system that applies the AI-based lecturer function grading, and a corresponding processing strategy is formulated to respond in a timely manner when encountering abnormal situations to avoid system crashes or error propagation. Redundancy design is implemented in the system that applies the AI-based lecturer function grading, such as multi-server backup, dual-machine hot standby, etc., to ensure that single point failures do not affect the normal operation of the entire system. An effective monitoring system is established in the system that applies the AI-based lecturer function grading, and system performance and security status are regularly evaluated to promptly discover and fix potential problems and continuously iterate and optimize.

[0071] Through the comprehensive application of the above methods, the robustness of the model can be effectively improved, ensuring that it works stably and reliably in complex and changing environments.

[0072] In one of the embodiments, after inputting the lecturer's feature vector set into a pre-trained artificial intelligence evaluation model and outputting the functional level prediction result corresponding to the lecturer, the method may also include the following steps: outputting an evaluation report for the lecturer based on the functional level prediction result; the evaluation report includes but is not limited to functional level, scores for various abilities, strengths analysis, improvement point analysis, and comparative analysis with reference standards.

[0073] The processed lecturer feature vector set is input into the trained artificial intelligence evaluation model, and the model can output the corresponding functional level prediction for each lecturer. In addition, a detailed evaluation report can be automatically generated based on the prediction results. The evaluation report can include information such as functional level, various ability scores, advantages and improvement points analysis, and comparison with industry standards or best practices. Among them, the various ability scores can refer to the scores of lecturer data in various dimensions, such as professionalism scores, training effectiveness scores, and student satisfaction scores. Based on the scores of various abilities, the advantages and disadvantages of the lecturer are analyzed to give personalized improvement suggestions for the lecturer. When compared with the industry standard, the data of lecturers rated as excellent in various dimensions can be defined as industry standards in advance, or the industry standards can be collected and optimized with the help of a large model, and the lecturer's functional level, various ability scores and other data can be compared with the industry standard. Based on the specific performance and teaching style of the lecturer, personalized professional development suggestions are proposed to guide the lecturer on how to improve teaching skills, optimize course content or enhance student interaction, and promote the personal growth of the lecturer. It can be seen that the above method can provide personalized improvement suggestions for each lecturer and promote the continuous improvement of the lecturer's ability.

[0074] In one embodiment, after inputting the lecturer feature vector set into the pre-trained artificial intelligence evaluation model and outputting the prediction result of the functional level corresponding to the lecturer, the method may further include the following steps:

[0075] Adjust the model parameters of the artificial intelligence assessment model based on the matching degree between the function level prediction results and the actual assessment results of the lecturer; or

[0076] The actual evaluation results of lecturers are added to the historical lecturer function level dataset, and the artificial intelligence evaluation model is retrained using the new historical lecturer function level dataset.

[0077] In the embodiments provided in the present application, a feedback optimization mechanism can also be designed to continuously iteratively optimize the artificial intelligence evaluation model. Specifically, after the artificial intelligence evaluation model outputs the functional level prediction result, the matching degree of the functional level prediction result and the actual evaluation result of the lecturer can be compared, and the model parameters can be adjusted according to the error analysis to iteratively optimize the evaluation accuracy. Alternatively, information such as the matching degree between the functional level prediction result and the actual evaluation result of the lecturer is added as new data to the labeled historical lecturer functional level data set, and the artificial intelligence evaluation model is retrained based on the new historical lecturer functional level data set to iteratively optimize the evaluation accuracy.

[0078] When constructing an artificial intelligence evaluation model, cross entropy can be selected as the loss function of the artificial intelligence evaluation model, and cross entropy can be used to accurately measure the difference between the model prediction result and the true label to ensure that the model is continuously optimized during the training process. In addition, in terms of the optimization algorithm, an Adam optimizer with an adaptive learning rate adjustment mechanism can be used. The Adam optimizer can dynamically adjust the learning rate according to the data changes during the training process, thereby accelerating the convergence speed and reducing the risk of overfitting. In addition, for the model training process of the artificial intelligence evaluation model, for the setting of the number of training rounds and the batch size, after multiple experimental verifications, it was determined that when the number of training rounds is set to 50 rounds and the batch size is set to 64, better training results can be achieved. That is, preferably, when the artificial intelligence evaluation model is trained, the number of training rounds is set to 50 rounds and the batch size is set to 64. By setting the number of training rounds to 50 rounds and the batch size to 64, it is ensured that the model can fully learn the features in the training data, and effectively control the training time and resource consumption. In addition, the early stopping strategy can be combined with model training. When the performance on the validation set does not improve for several consecutive rounds, the training is terminated early. In other words, the early stopping strategy is used to avoid overfitting in model training.

[0079] The above-mentioned AI-based lecturer function grading method can respond to changes in training results in real time, dynamically adjust the lecturer's function positioning, and adapt to changes in corporate training needs. At the same time, by designing a feedback optimization mechanism, the evaluation model can be continuously optimized based on historical evaluation results and subsequent feedback on lecturer performance to improve grading accuracy. The AI ​​evaluation model can continuously learn from new data through machine learning and optimize the evaluation model, so that the evaluation standards and methods can evolve over time and maintain advancement and adaptability. It can be seen that the above method can not only ensure the efficiency and accuracy of the automatic grading of lecturer functions, but also promote the continuous optimization and personalized development of the training system, bringing significant benefits to corporate training management.

[0080] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0081] Based on the same inventive concept, the embodiment of the present application also provides an artificial intelligence-based lecturer function grading system for implementing the artificial intelligence-based lecturer function grading method involved above. The implementation scheme for solving the problem provided by the system is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more artificial intelligence-based lecturer function grading system embodiments provided below can refer to the limitations of the artificial intelligence-based lecturer function grading method above, and will not be repeated here.

[0082] Figure 6 This is a structural diagram of an artificial intelligence-based lecturer function grading system in one of the embodiments of the present application. In one of the embodiments, the artificial intelligence-based lecturer function grading system may include a data acquisition module 100, a data analysis module 200, a data processing module 300 and a function grading module 400.

[0083] The data collection module 100 can be used to collect the training information of the instructor.

[0084] The data analysis module 200 may be connected to the data collection module 100 , and the data analysis module 200 may be used to analyze the training information and convert the training information into quantifiable indicator data.

[0085] The data processing module 300 can be connected to the data analysis module 200, and the data processing module 300 can be used to perform standardization processing on the quantifiable indicator data to form a lecturer feature vector set.

[0086] The function grading module 400 can be connected to the data processing module 300, and the function grading module 400 can be used to input the lecturer's feature vector set into a pre-trained artificial intelligence evaluation model and output the prediction result of the lecturer's corresponding function level.

[0087] In one embodiment, the data collection module 100 can obtain the training information of the lecturer through multiple channels. According to the training records of the lecturer, the teaching information of the lecturer can be obtained; after the training, the evaluation information of the trainees on the lecturer can be obtained; according to the career development of the trainees after the training, the teaching effectiveness information of the lecturer can be obtained.

[0088] In one embodiment, the teaching information may include a lecture video of a lecturer. When the data analysis module 200 analyzes the training information, it may convert the audio data in the lecture video into text data; use a general natural language model to analyze the text data to determine the lecturer's language expression quality data; and use computer vision technology to analyze the lecturer's non-verbal behavior in the image data of the lecture video to determine the lecturer's behavior expression quality data.

[0089] In one embodiment, when the data analysis module 200 analyzes the training information, it can also use statistical analysis methods to extract quantifiable information from the evaluation information and determine the lecturer's teaching evaluation data.

[0090] In one embodiment, the lecturer function grading system based on artificial intelligence may further include a model training module. The model training module may be used to pre-build and train an artificial intelligence evaluation model. The model training module may build an artificial intelligence evaluation model based on a machine learning algorithm; train the artificial intelligence evaluation model using a labeled historical lecturer function level data set; during the training process, use a cross-validation method to process the feature vector input into the artificial intelligence evaluation model, use a model ensemble learning method to combine the prediction results of multiple models to evaluate the performance of the artificial intelligence evaluation model, and adjust the model parameters of the artificial intelligence evaluation model.

[0091] In one embodiment, the artificial intelligence-based lecturer function grading system may also include a feedback and optimization module, which may output an evaluation report for the lecturer based on the function level prediction result output by the function grading module 400; the evaluation report includes but is not limited to the function level, scores of various abilities, strengths analysis, improvement point analysis, and comparative analysis with reference standards.

[0092] In one embodiment, the feedback and optimization module may also adjust the model parameters of the artificial intelligence evaluation model according to the matching degree between the function level prediction result and the actual evaluation result of the lecturer; or

[0093] The actual evaluation results of lecturers are added to the historical lecturer function level dataset, and the artificial intelligence evaluation model is retrained using the new historical lecturer function level dataset.

[0094] This application has carried out a large number of experimental verifications on the above-mentioned artificial intelligence-based lecturer function grading method, and collected rich data as experimental support. The experimental data comes from the enterprise online learning platform, covering lecturers and students from different industries and levels, and is widely representative and practical.

[0095] In terms of data processing, preprocessing technology and feature extraction methods were used to construct a vector set containing multi-dimensional features of the instructor's comprehensive ability. Multiple AI evaluation models were trained using the constructed feature vectors, and the classification accuracy and robustness of the models were improved through cross-validation, model fusion and other methods.

[0096] The experimental results show that the artificial intelligence evaluation model constructed using the above-mentioned artificial intelligence-based lecturer function grading method can accurately classify lecturers into preset functional levels, and has achieved significantly better results than traditional evaluation methods in multiple evaluation indicators.

[0097] Each module in the above-mentioned artificial intelligence-based lecturer function grading system can be implemented in whole or in part through software, hardware and their combination. Each of the above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the corresponding operations of each of the above modules.

[0098] In one embodiment, Figure 7 The internal structure diagram of a computer device in one embodiment of the present application provides a computer device, which may be a server, and its internal structure diagram may be as follows: Figure 7 As shown. The computer device includes a processor, a memory and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a lecturer function grading method based on artificial intelligence is implemented.

[0099] Those skilled in the art will understand that Figure 7 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0100] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0101] In the description of this specification, the description with reference to the terms "some embodiments", "other embodiments", "ideal embodiments", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic descriptions of the above terms do not necessarily refer to the same embodiment or example.

[0102] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0103] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the invention patent. It should be pointed out that for ordinary technicians in this field, several modifications and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be based on the attached claims.

Claims

1. A lecturer function grading method based on artificial intelligence, characterized in that: include: Collect training information of lecturers; Analyzing the training information and converting the training information into quantifiable indicator data; Standardizing the quantifiable indicator data to form a lecturer feature vector set; The lecturer feature vector set is input into a pre-trained artificial intelligence evaluation model, and a prediction result of the functional level corresponding to the lecturer is output.

2. The method for grading lecturer's functions based on artificial intelligence according to claim 1 is characterized in that: The training information collected from the instructor includes: According to the training record of the lecturer, obtaining the teaching information of the lecturer; After the training, obtaining the trainees' evaluation information on the instructor; The teaching effectiveness information of the instructor is obtained based on the career development of the trainees after the training.

3. The method for grading lecturer's functions based on artificial intelligence according to claim 2 is characterized in that: The teaching information includes the lecturer's teaching video, and the analyzing the training information and converting the training information into quantifiable indicator data includes: Converting the audio data in the teaching video into text data; Analyzing the text data using a general natural language model to determine the language expression quality data of the lecturer; Computer vision technology is used to analyze the non-verbal behavior of the lecturer in the image data of the teaching video to determine the behavioral expression quality data of the lecturer.

4. The method for grading lecturer's functions based on artificial intelligence according to claim 2 is characterized in that: The analyzing the training information and converting the training information into quantifiable indicator data further comprises: Utilizing statistical analysis methods, quantifiable information is extracted from the evaluation information to determine the teaching evaluation data of the lecturer.

5. The method for grading lecturer's functions based on artificial intelligence according to claim 1 or 2, characterized in that: The method for constructing the artificial intelligence evaluation model includes: Build artificial intelligence evaluation models based on machine learning algorithms; The artificial intelligence evaluation model is trained using a labeled historical lecturer function level dataset; During the training process, the feature vector input into the artificial intelligence evaluation model is processed using a cross-validation method, and the performance of the artificial intelligence evaluation model is evaluated using a model ensemble learning method combined with the prediction results of multiple models to adjust the model parameters of the artificial intelligence evaluation model.

6. The method for grading lecturer's functions based on artificial intelligence according to claim 1 is characterized in that: After inputting the lecturer feature vector set into a pre-trained artificial intelligence evaluation model and outputting a prediction result of the functional level corresponding to the lecturer, the method further includes: An evaluation report for the lecturer is outputted according to the functional level prediction result; the evaluation report includes but is not limited to functional level, scores of various abilities, strengths analysis, improvement point analysis, and comparative analysis with reference standards.

7. The method for grading lecturer's functions based on artificial intelligence according to claim 1 is characterized in that: After inputting the lecturer feature vector set into a pre-trained artificial intelligence evaluation model and outputting a prediction result of the functional level corresponding to the lecturer, the method further includes: Adjusting the model parameters of the artificial intelligence evaluation model according to the matching degree between the function level prediction result and the actual evaluation result of the lecturer; or The actual evaluation results of the lecturer are added to the historical lecturer function level data set, and the artificial intelligence evaluation model is retrained using the new historical lecturer function level data set.

8. An artificial intelligence-based lecturer function grading system, characterized in that: include: Data collection module, used to collect instructors’ training information; A data analysis module, connected to the data acquisition module, for analyzing the training information and converting the training information into quantifiable indicator data; A data processing module, connected to the data analysis module, for performing standardization processing on the quantifiable indicator data to form a lecturer feature vector set; The function grading module is connected to the data processing module and is used to input the lecturer feature vector set into a pre-trained artificial intelligence evaluation model and output a prediction result of the function level corresponding to the lecturer.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the lecturer function grading method based on artificial intelligence described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the lecturer function grading method based on artificial intelligence described in any one of claims 1 to 7 are implemented.

Images