Intelligent self-adaptive driving skill evaluation question generation method and system

Abstract

The invention discloses an intelligent self-adaptive driving skill evaluation question generation method, process and system, and aims to improve the pertinence and accuracy of driving skill evaluation. The method comprises the steps of establishing a driving skill evaluation question bank, analyzing the driving skill level and weak links of an examinee, intelligently screening and generating evaluation questions suitable for the examinee, and evaluating and feeding back the driving skill of the examinee according to an evaluation result. According to the invention, through the intelligent question generation and evaluation process, personalized evaluation questions can be provided, examinees are helped to improve driving skills, and evaluation efficiency and accuracy are improved. According to the method, the efficiency and quality of driving test training are remarkably improved, individuation and intelligence of learning experience are promoted, revolutionary progress is brought to the driver training industry, the problems that an existing driving test question library is lagged in updating and lack of individualized learning schemes are solved, and the method has high social value and market prospects.

Description

Technical Field

[0001] The present invention relates to the technical fields of intelligent education, intelligent transportation, and information processing, and specifically to a method and system for generating questions for intelligent adaptive driving skill evaluation. Background Art

[0002] With the rapid development of the automotive industry, the training and evaluation of driving skills have become increasingly important. Traditional driving skill evaluation mainly relies on manually compiling and maintaining question banks. The traditional method is time-consuming, laborious, and inefficient, and it is difficult to meet the needs of candidates with different learning abilities and uneven experience levels. Therefore, developing an intelligent adaptive driving skill evaluation method, process, and system has important social significance and economic value in the driving training market. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a method and system for generating questions for intelligent adaptive driving skill evaluation to solve the problems existing in the prior art.

[0004] The technical solution of the present invention is as follows:

[0005] An intelligent adaptive driving skill evaluation question generation method includes the following steps:

[0006] S101: Establish a driving skill evaluation question bank according to the driving skill evaluation standards and requirements;

[0007] S102: Generate a personalized driving skill profile of the candidate according to the candidate's basic information and historical evaluation results;

[0008] S103: Intelligently screen and generate evaluation questions suitable for the candidate from the question bank according to the candidate's driving skill level and weak links;

[0009] S104: Push the generated evaluation questions to the candidate for evaluation, and evaluate and provide feedback on the candidate's driving skills according to the evaluation results, and update the candidate's personalized driving skill profile.

[0010] Further, the step S101 specifically includes:

[0011] S201: Collect and sort out the driving skill evaluation standards and requirements, including laws and regulations, safe driving knowledge, and practical operation skills;

[0012] S202: Construct a question bank framework, including question classification, difficulty level, question type, etc. Use a structured database management system to store and manage the question bank;

[0013] S203: Continuously update the question bank, establish a question bank update mechanism, regularly check and update the question bank content to reflect the latest laws and regulations and driving skill standards;

[0014] S204: Based on reinforcement learning technology, adjust the difficulty and type of questions according to the candidate's basic information and historical evaluation data, and automatically generate matching test questions according to different difficulties and types;

[0015] S205: Regularly maintain and update the question bank. By analyzing the candidate's feedback information and evaluation results, the intelligent agent predicts the future demand of the question bank and makes adjustments accordingly. [[ID=⑥]] [[ID=⑦]]

[0016] [[ID=⑧]]Further, step S102 specifically includes: [[ID=⑨]] [[ID=⑩]]

[0017] [[ID=⑪]]S301: Analyze the candidate's historical evaluation data using statistical analysis methods and machine learning algorithms; [[ID=⑫]] [[ID=⑬]]

[0018] [[ID=⑭]]S302: Conduct feature engineering, including extracting features related to driving skills, such as question type, difficulty, candidate's answering time, etc.; [[ID=⑮]] [[ID=⑯]]

[0019] [[ID=⑰]]S303: Use machine learning algorithms to model the features and evaluate the performance of the model through cross-validation methods; [[ID=⑱]] [[ID=⑲]]

[0020] [[ID=⑳]]S304: Generate a personalized driving skill profile of the candidate. [[ID=㉑]] [[ID=㉒]]

[0021] [[ID=㉓]]Further, step S103 specifically includes: [[ID=㉔]] [[ID=㉕]]

[0022] [[ID=㉖]]S401: Based on the candidate's personalized driving skill profile data, use a support vector machine (SVM) classification model to predict the suitability of questions to determine which questions are more suitable for the candidate; [[ID=㉗]] [[ID=㉘]]

[0023] [[ID=㉙]]S402: Apply the GAN method to automatically generate new questions that match the candidate's skill level and improve the quality of the generated questions through a discriminator. [[ID=㉚]] [[ID=㉛]]

[0024] [[ID=㉜]]Further, step S104 specifically includes: [[ID=㉝]] [[ID=㉞]]

[0025] [[ID=㉟]]S501: Use an online evaluation system to collect the candidate's actual performance data during the evaluation, including answering situation, score, and answering time; [[ID=㊱]] [[ID=㊲]]

[0026] [[ID=㊳]]S502: Combine the random forest algorithm to analyze the candidate's evaluation results to evaluate the candidate's overall driving skill level and weak links; [[ID=㊴]] [[ID=㊵]]

[0027] [[ID=㊶]]S503: According to the evaluation results, update the candidate's personalized driving skill profile and provide the candidate with a personalized driving skill evaluation report and improvement suggestions. [[ID=㊷]] [[ID=㊸]]

[0028] [[ID=㊹]]The present invention also provides an intelligent adaptive driving skill evaluation question generation system for implementing the above intelligent adaptive driving skill evaluation question generation method, including the following modules:

[0029] Question Bank Management Module: Responsible for storing and maintaining a driving skills assessment question bank that includes various question types and difficulty levels, and regularly updating the question bank content to reflect the latest laws, regulations, and driving skills standards.

[0030] Information Analysis Module: Collects, stores, and analyzes the basic information and historical assessment results of candidates, generates personalized driving skills portraits of candidates to identify their driving skills levels and weak links.

[0031] Question Generation Module: Based on the personalized driving skills portraits of candidates, intelligently filters and generates assessment questions suitable for candidates, ensuring that the questions cover different difficulties and types.

[0032] Assessment Evaluation Module: Collects assessment results and evaluates and provides feedback on candidates' driving skills, and analyzes candidates' assessment results using machine learning methods such as the random forest algorithm.

[0033] Online Assessment Module: Provides a user interface and a real-time feedback mechanism, allowing candidates to conduct online assessments and receive assessment results.

[0034] Data Storage Module: Responsible for storing and backing up data such as candidates' basic information, historical assessment results, personalized driving skills portraits, assessment questions, and assessment results. This module should be able to handle structured and unstructured data to ensure the long-term preservation and traceability of data.

[0035] Data Security and Privacy Protection Module: Responsible for protecting the security of candidates' information, assessment results, and question bank data, and taking encryption measures and access control strategies in accordance with national data security laws and regulations requirements to prevent data leakage, damage, or loss.

[0036] System Main Control Module: Responsible for managing and monitoring the operation of the entire system, including question bank updates, candidate information management, system configuration, etc., to ensure the stable operation of the system.

[0037] Data Integration Module: Responsible for integrating information from different data sources, such as candidate information, assessment results, and question bank data, to ensure data consistency and integrity among system modules.

[0038] Furthermore, the question bank module further includes:

[0039] A structured database for storing theoretical questions, simulation questions, and practical operation questions;

[0040] Question classification and tagging system for easy and precise retrieval;

[0041] An automatic update mechanism to regularly obtain new question content from relevant laws, regulations, and safe driving knowledge bases.

[0042] Furthermore, the data integration module further includes:

[0043] An integrated data function that integrates information from different data sources, such as candidate information, evaluation results, and question bank data;

[0044] A data cleaning and preprocessing function to ensure the accuracy and consistency of the data;

[0045] Data conversion and standardization to adapt to the data requirements of different modules;

[0046] Data access control and permission management to ensure that only authorized users can access sensitive data;

[0047] A data backup and recovery mechanism to prevent data loss and system failures.

[0048] The advantages of the present invention are as follows:

[0049] The present invention can generate evaluation questions suitable for a candidate according to the candidate's actual situation and needs, improving the pertinence and accuracy of the evaluation; by analyzing the candidate's driving skill level and weak links, it can provide personalized evaluation questions targeted to help the candidate improve their driving skills; adopting a systematic evaluation process can improve the efficiency and accuracy of the evaluation, reduce subjectivity and human intervention; by evaluating and providing feedback on the evaluation results, it can help the candidate understand their driving skill level and provide directions and suggestions for improvement. It significantly enhances the personalization and efficiency of the candidate's learning, makes the evaluation closer to the candidate's true ability, changes the limitations of traditional driving test question banks, and helps improve the overall quality of driving education. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 is a flowchart of the intelligent adaptive driving skill evaluation question generation method of the present invention;

[0051] Figure 2 is a core function structure diagram of the intelligent adaptive driving skill evaluation question generation system of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0052] The following further describes the specific embodiments of the present invention with reference to the accompanying drawings. It should be noted here that the description of these embodiments is for helping to understand the present invention, but does not constitute a limitation to the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0053] As Figure 1-2 shown:

[0054] An intelligent adaptive driving skill evaluation question generation method includes the following steps:

[0055] S101: Establish and maintain a driving skill assessment question bank that includes various question types and difficulty levels to meet the standards and requirements stipulated by the state for driving skill assessment;

[0056] S102: Collect and analyze the basic information and historical assessment data of candidates, generate personalized driving skill portraits of candidates to identify the driving skill levels and weak links of candidates;

[0057] S103: Based on the personalized driving skill portraits of candidates, intelligently screen and generate assessment questions suitable for candidates to ensure that the questions cover different difficulties and types;

[0058] S104: Push them to candidates through an online platform for assessment, collect the assessment results, evaluate and give feedback on the driving skills of candidates according to the assessment results, and update the personalized driving skill portraits of candidates.

[0059] The specific technical details and solutions for each step are as follows:

[0060] I: Establish a driving skill assessment question bank according to the driving skill assessment standards and requirements.

[0061] 1. Specifically, first collect and sort out the standards and requirements for driving skill assessment, including laws and regulations, safe driving knowledge, practical operation skills, etc.

[0062] 2. Then construct the framework of the question bank, including the classification of questions, difficulty levels, question types, etc. Use a structured database management system to store and manage the question bank. Specifically, the method for constructing the assessment question bank includes the following steps:

[0063] a. Initialize the framework of the question bank, including basic information such as the classification of questions, difficulty levels, question types, etc.;

[0064] b. Use a structured database management system (such as MySQL, etc.) to store and manage the question bank;

[0065] c. Define the database table structure, including question tables, classification tables, difficulty level tables, question type tables, etc.;

[0066] d. Store the detailed information of the questions in the question table, including question ID, question content, question answers, question types, question difficulty levels, etc.;

[0067] e. Store the classification information of the questions in the classification table, including classification ID, classification name, etc.;

[0068] f. Store the difficulty level information of the questions in the difficulty level table, including difficulty level ID, difficulty level name, etc.;

[0069] g. Store the question type information of the questions in the question type table, including question type ID, question type name, etc.;

[0070] h. Associate the question information with information such as classification, difficulty level, question type, etc. to form a complete question bank framework.

[0071] i. After the question bank is constructed, the maintenance and update of the question bank can be carried out to reflect the latest laws, regulations and driving skill standards;

[0072] j. Output the constructed intelligent adaptive driving skill evaluation question bank framework.

[0073] 3. Continuously update the question bank, establish a question bank update mechanism, regularly check and update the content of the question bank to reflect the latest laws, regulations and driving skill standards.

[0074] 4. Based on reinforcement learning technology, train an agent to learn how to generate test questions. The agent adjusts the difficulty and type of questions according to the basic information and historical evaluation data of the examinees, and automatically generates matching test questions according to different difficulties and types, so as to ensure the diversity and comprehensiveness of the question bank. Specifically, use the Q-learning algorithm in reinforcement learning to train the agent. The core calculation and update of the Q-value in the Q-learning algorithm are as follows:

[0075] Q(s,a)←Q(s,a)+α[reward+γMax a' Q(s',a')-Q(s,a)]

[0076] Among them: The binary function Q(s,a) represents the expected return of taking action a in the current state s; reward represents the immediate return immediately given by the environment when the state changes from s to s'; s' represents the next state; a' represents the possible action in the next state s'; γ represents the discount factor; α represents the learning rate.

[0077] 5. Regularly maintain and update the question bank. By analyzing the feedback information and evaluation results of the examinees, the agent predicts the future needs of the question bank and makes adjustments accordingly. Specifically, use the K-Means clustering analysis algorithm to predict the future needs of the question bank. The key formula in the clustering analysis is as follows:

[0078]

[0079] Among them: c k is the center point of cluster k. n k is the number of data points in cluster k. x i is data point i. The detailed calculation process includes:

[0080] 1. Collect the feedback information and evaluation result data of the candidates.

[0081] 2. Use the K-Means algorithm to perform clustering analysis on the data and divide the data into multiple clusters.

[0082] 3. Analyze the characteristics of each cluster and predict the future demand of the question bank.

[0083] 4. According to the prediction results, make corresponding adjustments and updates to the question bank.

[0084] II. Generate a personalized driving skill profile of the candidates based on their basic information and historical evaluation results.

[0085] 1. Candidate information collection. Use the provided user interface and form to allow candidates to fill in basic information online and verify the integrity of the information, such as age, driving experience, etc. Establish a basic profile of the candidates, including personal information and driving experience, etc. These information will be used to analyze the driving skill level and weak links of the candidates in the follow-up. Clean and organize the historical evaluation data of the candidates to form a complete data set. Among them:

[0086] Use the decision tree model to learn the data distribution and identify the points that are significantly different from the data distribution. Specifically, use the anomaly detection methods in the Scikit-Learn library of Python, such as IsolationForest or LocalOutlierFactor. Identify the outliers and adopt removal or resampling to ensure the data quality and model accuracy.

[0087] For numerical data, the mean and median of the data can be used to fill in the missing values. For categorical data, the mode (the most common value) is used to fill in the missing values. For time series data, use the time series analysis tools in the statsmodels library of Python and use the ARIMA time series analysis method to predict the missing values.

[0088] 2. Collection of historical evaluation results. Establish a historical evaluation result database to store the previous evaluation data of the candidates. The data includes the candidates' answers, scores, answering times for each question, etc., and ensure the accuracy and integrity of the data for subsequent statistical analysis and machine learning modeling.

[0089] 3. Analysis of the driving skill level and weak links of the candidates. Use statistical analysis methods to calculate indicators such as the average score and passing rate of the candidates.

[0090] The calculation formula for the average score indicator is as follows:

[0091]

[0092] The calculation formula for the passing rate index is as follows:

[0093]

[0094] Feature engineering: Extract features related to driving skills, such as question type, difficulty, and the time taken by candidates to answer questions. Specifically, use the data preprocessing tool, the Pandas library, to clean and transform the data. Select the most representative features using the information gain feature selection method, and use feature transformation methods such as normalization and standardization to process the features for model training.

[0095] Model training: Use machine learning algorithms to model the features and train a prediction model. Specifically, use the model training tool, the Scikit-Learn library, select the decision tree algorithm in machine learning algorithms for model training, and adjust model parameters such as the depth of the tree and the learning rate to improve the performance of the model.

[0096] Model evaluation: Use methods such as cross-validation to evaluate the performance of the model and optimize the model based on the evaluation results. Use the K-fold cross-validation technique to evaluate the performance of the model. Use the Matplotlib library in Python to visually display and analyze the model performance, and optimize the model according to the evaluation results, such as adjusting model parameters and selecting better features.

[0097] Predictive analysis: Use the trained model to predict the driving skill level and weak links of candidates. Generate a prediction report, including prediction results, weak links, etc. Visually display the prediction results in the form of bar charts, pie charts, etc. for easy understanding by candidates and coaches.

[0098] 4. Generation of candidates' personalized driving skill portraits. Among them, the candidates' personalized driving skill portraits include:

[0099] 1. Basic information: Basic information such as the age, gender, and driving experience of the candidate;

[0100] 2. Driving skill level: The driving skill level of the candidate, including theoretical knowledge and practical operation skills;

[0101] 3. Common error types: The types of errors that candidates often make in historical evaluations, and the impact of these errors on driving safety;

[0102] 4. Driving habits: The driving habits of the candidate, including driving speed, following distance, steering actions, etc.;

[0103] 5. Emergency response ability: The reaction speed and handling ability of the candidate in emergency situations;

[0104] 6. Degree of compliance with laws and regulations: The candidate's understanding and compliance with traffic laws and regulations;

[0105] 7. Safety awareness: The degree of importance that candidates attach to driving safety and their willingness to abide by traffic rules;

[0106] 8. Psychological quality: The psychological stability and coping ability of candidates under pressure and emergency situations;

[0107] 9. Personality characteristics of candidates: The personality traits, attitudes and behavior patterns of candidates, and the impact of these characteristics on driving skills;

[0108] 10. Learning needs of candidates: The needs and expectations of candidates in terms of improving driving skills, as well as the learning resources and guidance they hope to obtain.

[0109] The method for generating a personalized driving skill profile of candidates is as follows:

[0110] Input: Basic information of candidates, historical evaluation data

[0111] Output: Personalized driving skill profile of candidates

[0112] Main steps of the algorithm:

[0113] 1. Initialize the personalized driving skill profile of candidates, including basic information of candidates (such as age, gender, driving experience, etc.) and historical evaluation data (such as evaluation scores, common error types, etc.).

[0114] 2. Use statistical analysis methods and machine learning algorithms to analyze the historical evaluation data of candidates to identify the driving skill level and weak links of candidates.

[0115] 3. Conduct feature engineering, including extracting features related to driving skills, such as question types, difficulty levels, candidate answering times, etc.

[0116] 4. Use machine learning algorithms to model the features and evaluate the performance of the model through cross-validation methods.

[0117] 5. Generate a personalized driving skill profile for candidates based on the prediction results of the model, including the driving skill level, common error types and driving habits of candidates.

[0118] 6. Output the personalized driving skill profile of candidates to provide support for subsequent question generation and evaluation.

[0119] III. Intelligently screen and generate evaluation questions suitable for the candidate from the question bank according to the candidate's driving skill level and weak links.

[0120] 1. Based on the personalized driving skill portrait of the examinee, use the SVM classification model to predict the suitability of questions. The SVM model is used to predict the matching degree between each question and the examinee's skill level, so as to determine which questions are more suitable for the examinee. Specifically, for a linearly separable data set, SVM finds a hyperplane that maximizes the margin between data points of different classes. In question screening, this hyperplane can be represented as the matching degree between the question and the examinee's skill level. Among them:

[0121] For a linear SVM, the objective function for maximizing the margin is calculated as follows:

[0122]

[0123] s.t.y i (w T x i +b)≥1, i = 1, 2, …, m

[0124] For a non - linear data set, SVM uses the Kernel trick to map the data into a high - dimensional space and find the hyperplane with the maximum margin in the high - dimensional space. In question screening, the Kernel trick can map the question features into a high - dimensional space, enabling SVM to capture the non - linear relationship between the question and the examinee's skill level. For a non - linear SVM, the objective function for maximizing the margin is calculated as follows:

[0125]

[0126] s.t.y i (Φ(w) T Φ(x i )+b)≥1, i = 1, 2, …, m

[0127] 2. Apply the GAN method to generate questions. Use the generator of the GAN model to generate new questions that match the examinee's skill level, and use the discriminator to improve the quality of the generated questions.

[0128] Specifically, the purpose of the generator network is to generate new questions. Its output is a random noise vector, which is transformed through a series of operations to generate new questions. The calculation formula for the loss function of the generator is as follows:

[0129] [[ID=4l]]

[0130] Specifically, the purpose of the discriminator network is to distinguish between real questions and generated questions. Its input can be either real questions or generated questions, and the output is a probability value indicating the probability that the input is a real question. The calculation formula for the loss function of the discriminator is as follows:

[0131]

[0132] IV. Push the generated assessment questions to the candidates for assessment, evaluate and provide feedback on the candidates' driving skills based on the assessment results, and update the candidates' personalized driving skill profiles.

[0133] 1. Collect assessment results: Use an online assessment system to collect the actual performance data of candidates in the assessment, including answering situations, scores, answering times, etc., and update the candidates' personalized driving skill profiles.

[0134] 2. Evaluate the driving skill level of candidates: Analyze the assessment results of candidates by combining the random forest algorithm to evaluate the overall driving skill level and weak links of candidates. The main algorithm steps of the candidate driving skill evaluation method are as follows:

[0135] Input: Candidate assessment result data set

[0136] Output: Driving skill evaluation results of candidates

[0137] Main algorithm steps:

[0138] a. Data preprocessing: Clean the data and standardize the features;

[0139] b. Feature selection: Determine the features related to driving skills;

[0140] c. Data partitioning: Divide the data set into a training set and a test set;

[0141] d. Model construction: Use the random forest algorithm to train the model on the training set;

[0142] e. Model evaluation: Evaluate the model performance on the test set;

[0143] f. Skill evaluation: Predict the driving skill level of candidates according to the model;

[0144] g. Weak link analysis: Analyze the assessment results of candidates to find out the weak links;

[0145] h. Output results: Generate a driving skill evaluation report for candidates.

[0146] Through the above algorithm steps, the assessment results of candidates can be analyzed by combining the random forest algorithm, the driving skill level of candidates can be evaluated, and the weak links can be found to provide a personalized driving skill evaluation report for candidates.

[0147] 3. According to the evaluation results, update the candidates' personalized driving skill profiles, and provide a personalized driving skill evaluation report and improvement suggestions for candidates.

[0148] The present invention also provides an intelligent adaptive driving skill assessment question generation system, including:

[0149] Question Bank Management Module: Responsible for storing and maintaining a driving skills assessment question bank that includes various question types and difficulty levels, and regularly updating the question bank content to reflect the latest laws, regulations, and driving skill standards. The question bank module includes:

[0150] A structured database for storing theoretical questions, simulation questions, and practical operation questions;

[0151] Question classification and tagging system for quick retrieval and precise screening;

[0152] An automatic update mechanism to regularly obtain new question content from relevant laws, regulations, and safe driving knowledge bases.

[0153] Information Analysis Module: Collects, stores, and analyzes candidates' basic information and historical assessment results, generating personalized driving skill portraits of candidates to identify their driving skill levels and weak points.

[0154] Question Generation Module: Based on candidates' personalized driving skill portraits, intelligently screens and generates assessment questions suitable for candidates, ensuring that the questions cover different difficulties and types.

[0155] Assessment Evaluation Module: Collects assessment results and evaluates and provides feedback on candidates' driving skills, analyzing candidates' assessment results using machine learning methods such as the random forest algorithm.

[0156] Online Assessment Module: Provides a user interface and a real-time feedback mechanism, allowing candidates to conduct online assessments and receive assessment results.

[0157] Data Storage Module: Responsible for storing and backing up data such as candidates' basic information, historical assessment results, personalized driving skill portraits, assessment questions, and assessment results. This module should be able to handle structured and unstructured data to ensure the long-term preservation and traceability of data.

[0158] Data Security and Privacy Protection Module: Responsible for protecting the security of candidates' information, assessment results, and question bank data, adopting encryption measures and access control strategies to prevent data leakage, damage, or loss.

[0159] System Master Control Module: Responsible for managing and monitoring the operation of the entire system, including question bank updates, candidate information management, system configuration, etc., to ensure the stable operation of the system.

[0160] Data Integration Module: Responsible for integrating information from different data sources, such as candidate information, assessment results, and question bank data, to ensure data consistency and integrity among the system modules. The data integration module further includes:

[0161] Integrated data function to integrate information from different data sources, such as candidate information, assessment results, and question bank data;

[0162] Data cleaning and preprocessing functions to ensure data accuracy and consistency;

[0163] Data transformation and standardization to adapt to the data requirements of different modules;

[0164] Data access control and privilege management to ensure that only authorized users can access sensitive data;

[0165] Data backup and recovery mechanisms to prevent data loss and system failures.

[0166] Through the above technical solutions, the present invention can generate evaluation questions suitable for the examinee according to the actual situation and needs of the examinee, improving the pertinence and accuracy of the evaluation; by analyzing the driving skill level and weak links of the examinee, it can provide personalized evaluation questions targeted to help the examinee improve their driving skills; adopting a systematic evaluation process can improve the efficiency and accuracy of the evaluation, reduce subjectivity and human intervention; by evaluating and providing feedback on the evaluation results, it can help the examinee understand their own driving skill level and provide directions and suggestions for improvement. It significantly enhances the personalization and efficiency of the examinee's learning, makes the evaluation closer to the true ability of the examinee, changes the limitations of the traditional driving test question bank, and helps improve the overall quality of driving education.

[0167] The above has described the embodiments of the present invention in detail with reference to the accompanying drawings, but the present invention is not limited to the described embodiments. For those skilled in the art, without departing from the principles and spirit of the present invention, various changes, modifications, substitutions, and variations to these embodiments still fall within the protection scope of the present invention.

Claims

1. An intelligent adaptive driving skill evaluation question generation method, characterized in that It includes the following steps: S101: Establish a driving skill evaluation question bank according to the driving skill evaluation standards and requirements; S102: Generate a personalized driving skill profile of the candidate based on the candidate's basic information and historical evaluation results; S103: Intelligently screen and generate evaluation questions suitable for the candidate from the question bank according to the candidate's driving skill level and weak links; S104: Push the generated evaluation questions to the candidate for evaluation, evaluate and give feedback on the candidate's driving skills according to the evaluation results, and update the candidate's personalized driving skill profile.

2. The intelligent adaptive driving skill evaluation question generation method according to claim 1, wherein The specific steps of S101 include: S201: Collect and sort out the standards and requirements for driving skill evaluation, including laws and regulations, safe driving knowledge, and practical operation skills; S202: Build a question bank framework, including question classification, difficulty level, question type, etc. Use a structured database management system to store and manage the question bank; S203: Continuously update the question bank, establish a question bank update mechanism, regularly check and update the question bank content to reflect the latest laws and regulations and driving skill standards; S204: Based on reinforcement learning technology, adjust the difficulty and type of questions according to the candidate's basic information and historical evaluation data, and automatically generate matching test questions according to different difficulties and types; S205: Regularly maintain and update the question bank. By analyzing the candidate's feedback information and evaluation results, the intelligent agent predicts the future needs of the question bank and makes adjustments accordingly.

3. The intelligent adaptive driving skill evaluation question generation method according to claim 2, wherein: The specific steps of S102 include: S301: Analyze the candidate's historical evaluation data using statistical analysis methods and machine learning algorithms; S302: Conduct feature engineering, including extracting features related to driving skills, such as question type, difficulty, candidate answering time, etc.; S303: Use machine learning algorithms to model the features and evaluate the performance of the model through cross-validation methods; S304: Generate a candidate's personalized driving skill profile.

4. The intelligent adaptive driving skill evaluation question generation method according to claim 1, characterized in that: The specific steps of S103 include: S401: Based on the candidate's personalized driving skill profile data, use a support vector machine (SVM) classification model to predict the suitability of questions to determine which questions are more suitable for the candidate; S402: Apply the GAN method to automatically generate new questions that match the candidate's skill level, and improve the quality of the generated questions through a discriminator.

5. The intelligent adaptive driving skill evaluation question generation method according to claim 1, characterized in that: The specific steps of S104 include: S501: Use an online evaluation system to collect the candidate's actual performance data in the evaluation, including answering situation, score, and answering time; S502: Analyze the candidate's evaluation results in combination with the random forest algorithm to evaluate the candidate's overall driving skill level and weak links; S503: According to the evaluation results, update the candidate's personalized driving skill profile, and provide the candidate with a personalized driving skill evaluation report and improvement suggestions.

6. An intelligent adaptive driving skill evaluation question generation system for implementing the intelligent adaptive driving skill evaluation question generation method according to any one of claims 1-5, characterized in that, It includes the following modules: Question Bank Management Module: Responsible for storing and maintaining a driving skill evaluation question bank containing various question types and difficulty levels, and regularly updating the question bank content to reflect the latest laws and regulations and driving skill standards. Information Analysis Module: Collect, store, and analyze the basic information and historical evaluation results of candidates, generate personalized driving skill portraits of candidates to identify their driving skill levels and weak points. Question Generation Module: Based on the personalized driving skill portraits of candidates, intelligently screen and generate evaluation questions suitable for candidates, ensuring that the questions cover different difficulties and types. Evaluation and Assessment Module: Collect evaluation results and evaluate and provide feedback on candidates' driving skills, and analyze candidates' evaluation results using machine learning methods such as the random forest algorithm. Online Evaluation Module: Provide a user interface and a real-time feedback mechanism, allowing candidates to conduct online evaluations and receive evaluation results. Data Storage Module: Responsible for storing and backing up data such as candidates' basic information, historical evaluation results, personalized driving skill portraits, evaluation questions, and evaluation results. This module should be able to handle structured and unstructured data to ensure the long-term preservation and traceability of data. Data Security and Privacy Protection Module: Responsible for protecting the security of candidates' information, evaluation results, and question bank data. In accordance with national data security laws and regulations, encryption measures and access control strategies are adopted to prevent data leakage, damage, or loss. System Master Control Module: Responsible for managing and monitoring the operation of the entire system, including question bank updates, candidate information management, system configuration, etc., to ensure the stable operation of the system.

7. The intelligent adaptive driving skill evaluation question generation system according to claim 6, wherein Data Integration Module: Responsible for integrating information from different data sources, such as candidate information, evaluation results, and question bank data, to ensure data consistency and integrity among the various modules of the system. The question bank module further includes: A structured database for storing theoretical questions, simulation questions, and practical operation questions; A question classification and tagging system for easy retrieval and precise screening; 8. The intelligent adaptive driving skill evaluation question generation system according to claim 6, wherein: An automatic update mechanism to regularly obtain new question content from relevant laws and regulations and safe driving knowledge bases. The data integration module further includes: An integrated data function to integrate information from different data sources, such as candidate information, evaluation results, and question bank data; A data cleaning and preprocessing function to ensure data accuracy and consistency; Data conversion and standardization to adapt to the data requirements of different modules; Data access control and permission management to ensure that only authorized users can access sensitive data; A data backup and recovery mechanism to prevent data loss and system failures.

Images