Nurse training examination system based on OKR

Abstract

The invention discloses a nurse training examination system based on OKR. According to the system, a nurse can effectively master clinical theoretical knowledge and skills through the system, the nursing service quality is improved, medical errors are reduced, the work satisfaction degree and occupational development of the nurse are improved, modernization of nursing education is greatly promoted through implementation of the system, and the development of the nurse is promoted. The invention discloses an OKR-based nurse training examination system, and aims to ensure that continuous education and professional development of nurses can meet the goals of medical institutions, and improve the nursing quality and patient nursing results at the same time. By integrating the OKR into the training and assessment process, the progress can be better tracked, nurses are helped to improve professional skills, finally, the medical service quality is improved, and benefits are brought to patients. According to the system, a customized and high-efficiency nurse training system is created, and contributions are made to the development of the nursing industry.

Description

Technical Field

[0001] This invention belongs to the field of nurse training and examination technology, specifically an OKR-based nurse training and examination system. Background Technology

[0002] Nurses are healthcare professionals who perform a full range of clinical procedures or specialized nursing activities related to medical care. Their mission is to alleviate patient suffering and protect lives. Their work is multifaceted and demands a high level of expertise, requiring continuous learning and training to improve their nursing skills and quality of care. To enhance their professional competence and prepare for various professional examinations, nurses need to study various medical and nursing books, regulations, job descriptions, emergency plans, and operational guidelines. Currently, the assessment of nurses' clinical competence, including periodic evaluations and the three-tiered examination system, is primarily based on theoretical examinations.

[0003] However, most training and examinations are conducted through face-to-face instruction, which makes the automation efficiency of the entire teaching and examination system insufficient and prone to cheating and other adverse effects during the examination process. Summary of the Invention

[0004] The purpose of this invention is to provide a nurse training and examination system based on OKR in order to solve the problems mentioned above.

[0005] The technical solution adopted in this invention is as follows: an OKR-based nurse training and examination system, the OKR-based nurse training and examination system comprising:

[0006] General function modules;

[0007] Training Center Module;

[0008] Examination Center;

[0009] Personal Center Module;

[0010] The examination center is equipped with teacher-side and student-side modules.

[0011] In a preferred embodiment, the general function module is internally configured with: user registration and login, system main page design, notification and announcement publishing, and a communication circle function to facilitate nurses to exchange experiences and knowledge.

[0012] In a preferred embodiment, the training center module includes: 2. Training introduction page; 3. Support for multi-format training courses (MP4, Word, Excel, PPT, PDF); 4. Viewing time statistics and interactive quizzes; 5. Training test papers and examination system; and 6. Automatic issuance of qualification certificates.

[0013] In a preferred embodiment, the examination center module is internally equipped with a simulation practice function, a formal examination system including anti-cheating measures and facial recognition technology, a make-up exam module, and examination record management.

[0014] In a preferred embodiment, the personal center module includes: personal information management, my KPI tracking and display, and OKR completion status.

[0015] In a preferred embodiment, the training center module includes functions such as training management framework, system management, personnel management, and training quality management, with training quality management being the focus of system construction. Intelligent goal management and process management are the basic elements of training quality management in this system. Goal management is achieved through goal quality control and task progress management. The nursing department presets training goals at the individual and nursing unit levels. The system automatically compares the correctness of the teaching goals submitted by nursing units, dynamically displays the completion status of nursing unit teaching tasks and individual training tasks, and can remind users to complete tasks. Closed-loop management of teaching activities and full-process, full-element analysis provide support for training process management. Training and assessment information is synchronized between the teaching and learning parties, connecting the teaching management data of the nursing department, major departments, specialties, and nursing units. It will also automatically generate vertical and horizontal comparison results to complete the continuous improvement of training quality through "evaluation-analysis-improvement".

[0016] In a preferred embodiment, the training center module features internal resource sharing: the system integrates multi-disciplinary teaching resources and enables resource sharing. Resource management is achieved through classification, quality control, and access control. The system covers resources such as operational skill standards and videos, industry standards, courseware resources, question banks, case studies, and evaluation forms. Resource quality management is implemented through resource review and user feedback. Users can browse and download teaching resources within their authorized scope.

[0017] In a preferred embodiment, the teacher-side module includes three aspects: efficient question bank management, diversified test paper generation, and diversified marking methods. Question bank management: Diverse question types include multiple-choice, fill-in-the-blank, true / false, short-answer, essay, and analytical questions. Teachers can manage the question bank via the network, such as adding, editing, and deleting questions. Adding questions can be done manually one by one or automatically in batches using templates. When adding questions, the difficulty, score, and answer can be set, with difficulty levels ranging from 1 to 5. Diverse test paper generation: Manual test paper generation and intelligent test paper generation. Manual test paper generation: Primarily, teachers manually select questions from the question bank according to exam requirements and assemble them into a suitable test paper. Manual test paper generation is inefficient, susceptible to human error, and prone to question leakage. Intelligent test paper generation: Teachers can set the composition and distribution of the test paper, the difficulty level of the questions, the generation logic, and the test duration. The system automatically extracts and combines questions from the question bank according to the generation requirements. Intelligent test paper generation is highly efficient and the randomness of question selection reduces the risk of test paper leakage to a certain extent. Clicking the "Customize" button in the intelligent test paper generation function will display the following customization options, allowing users to select the difficulty level, question type, number of questions, and question score according to their needs. Clicking "Generate" will generate the test questions based on the customized requirements.

[0018] In a preferred embodiment, the teacher-side module has two grading modes: automatic and manual. Teachers can choose manual grading to independently grade the papers, or they can use automatic grading for one-click grading. The automatic grading principle varies depending on the type of question. Exam questions mainly include objective and subjective questions: multiple choice, true / false, fill-in-the-blank, and short answer questions. Objective questions (multiple choice, true / false, and fill-in-the-blank) are graded by directly matching the answers to the correct answers; if a perfect match is found, points are awarded; otherwise, no points are awarded. Subjective questions (fill-in-the-blank and short answer) are graded by using TF feature vectors and the simhash algorithm to calculate the similarity of the Chinese text, thus assigning corresponding scores based on the similarity. When comparing the similarity of two feature vectors using the simhash algorithm, Hamming distance is used. Intelligent grading and score analysis are achieved by directly comparing the answers to multiple choice and true / false questions, thus realizing automatic grading. In addition, the teacher-side module can also perform comprehensive analysis of students' exam scores: exam analysis, exam situation analysis, and student analysis.

[0019] In a preferred embodiment, the student-side module employs two methods to prevent cheating and achieve intelligent proctoring: first, monitoring the examination machine by using front-end JavaScript to prevent cheating during web-based examinations by disabling keyboard copy-paste, screen switching, and enabling full-screen mode; second, monitoring the examinee by using MediaPipe, OpenCV, and TensorFlow to perform facial recognition verification for system login and real-time facial monitoring during the examination. Before the exam, in addition to a username and password, the system uses ArcFace technology provided by ArcSoft AI to compare the examinee's facial information in the facial database with the examinee's login facial information. Only after a successful facial match can the examinee enter the examination system. During the exam, the facial detection for cheating is mainly based on two types. Specifically, 1) Real-time facial recognition based on ArcFace technology. The camera is on throughout the exam, and facial information of the examinee is collected periodically for recognition. If a face is obscured, moves out of the camera area, or the camera captures a face image of a non-examinee for more than 2 seconds, the system determines that the examinee has potential cheating behavior and issues a warning. 2) Binocular depth measurement based on the Mediapipe iris tracking model detects the distance between the examinee's eyes and the exam screen. This data is used to determine if the examinee is looking around or moving away from (or closer to) the camera. Under normal exam conditions, the distance between the examinee's eyes and the exam screen is within a certain range. If the examinee looks around or moves their face away from the exam screen during the exam, and this continues for more than two days, the depth values ​​'s' of the left and right irises are used to determine if the examinee is potentially cheating and a warning is issued. Simultaneously, the exam machine's intelligent monitoring during the exam includes: initially preventing screen switching by setting the exam paper to full screen; mouse monitoring to ensure the mouse moves within the valid area of ​​the exam screen to prevent cheating through a second screen; and keyboard monitoring to check for the use of prohibited shortcut keys, such as screen switching, copy / paste (which is disabled during the exam), and zooming. If the examinee engages in any of these invalid exam behaviors, a warning will be issued, and the number of warnings will accumulate. When the number of warnings reaches a set threshold, the examinee will be forced to submit their exam.

[0020] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0021] 1. In this invention, nurses can effectively master clinical theoretical knowledge and skills through this system to improve the quality of nursing services, reduce medical errors, and increase nurses' job satisfaction and career development. The implementation of this system will greatly promote the modernization of nursing education. The purpose of the OKR-based nurse training and examination system is to ensure that nurses' continuing education and professional development align with the goals of medical institutions, while simultaneously improving nursing quality and patient outcomes. By integrating OKR into the training and assessment process, progress can be better tracked, helping nurses improve their professional skills and ultimately enhancing the quality of medical services for the benefit of patients. This system creates a customized, high-efficiency nurse training system, contributing to the development of the nursing industry.

[0022] 2. This invention incorporates a nursing education and assessment tool based on the OKR (Objectives and Key Results) framework. OKR is a popular management strategy used to set and track goals and their outcomes, commonly employed in businesses and organizations to improve efficiency and measurability of results. Applying OKR to a nurse training and examination system can help nursing education institutions and medical facilities ensure that nurses' training and development are aligned with organizational goals. By providing comprehensive online educational resources and functions, it strengthens nurses' clinical theoretical knowledge and practical skills. The system will utilize advanced information technology, combined with actual clinical nursing needs, to provide nurses with a personalized and precise learning experience to improve the quality of nursing services. Attached Figure Description

[0023] Figure 1 This is a system block diagram of the present invention. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0025] Reference Figure 1 ,

[0026] An OKR-based nurse training and examination system, comprising:

[0027] The system includes a general function module, a training center module, an examination center module, and a personal center module. The examination center is internally configured with teacher and student modules.

[0028] The general functional modules include: user registration and login, system main page design, notification and announcement publishing, and a communication circle function to facilitate nurses to exchange experiences and knowledge.

[0029] The training center module includes: 1. Training introduction page; 2. Support for multi-format training courses (MP4, Word, Excel, PPT, PDF); 3. Viewing time statistics and interactive quizzes; 4. Training test papers and examination system; 5. Automatic issuance of qualification certificates.

[0030] The exam center module includes a simulation practice function, a formal exam system, anti-cheating measures and facial recognition technology, a make-up exam module, and exam record management.

[0031] The Personal Center module includes: Personal Information Management, My KPI Tracking and Display, and OKR Completion Status.

[0032] The training center module includes functions such as training management framework, system management, personnel management, and training quality management, with training quality management being the key focus of the system's construction. Intelligent goal management and process management are the fundamental elements of this system's training quality management. Goal management is achieved through goal quality control and task progress management. The nursing department presets training goals at the individual and nursing unit levels. The system automatically compares the correctness of the teaching goals submitted by nursing units, dynamically displays the completion status of nursing unit teaching tasks and individual training tasks, and can remind users to complete tasks. Closed-loop management of teaching activities and full-process, full-element analysis provide support for training process management. Training and assessment information is synchronized between the teaching and learning parties, connecting teaching management data from the nursing department—major departments—specialties—nursing units. It will also automatically generate longitudinal and horizontal comparison results to achieve continuous improvement of training quality through "evaluation—analysis—improvement".

[0033] The training center module features internal resource sharing: the system integrates and shares teaching resources across multiple disciplines. Resource management is achieved through categorization, quality control, and access control. The system covers operational skill standards and videos, industry standards, courseware resources, question banks, case studies, evaluation forms, and other resources. Resource quality management is implemented through resource review and user feedback. Users can browse and download teaching resources within their authorized scope.

[0034] The teacher-side module comprises three aspects: efficient question bank management, diverse test paper generation, and diverse marking methods. Question Bank Management: Diverse question types include multiple-choice, fill-in-the-blank, true / false, short-answer, essay, and analytical questions. Teachers can manage the question bank online, such as adding, editing, and deleting questions. Adding questions can be done manually one by one or automatically in batches using templates. When adding questions, teachers can set the difficulty level, score, and answer key, with difficulty levels ranging from 1 to 5. Diverse Test Paper Generation: Manual and intelligent test paper generation. Manual Test Paper Generation: Teachers manually select questions from the question bank and assemble them into a suitable test paper according to exam requirements. Manual generation is inefficient, susceptible to human error, and prone to question leakage. Intelligent Test Paper Generation: Teachers can set the test paper composition, question difficulty level, generation logic, and exam duration. The system automatically extracts and combines questions from the question bank to generate the test paper according to these requirements. Intelligent test paper generation is highly efficient and offers greater randomness in question selection, thus reducing the risk of test paper leakage to some extent. Clicking the "Customize" button in the intelligent test paper generation system will display the following customization options, allowing users to select question difficulty, question type, number of questions, and question score according to their needs. Clicking "Generate" will generate the test questions based on the customized requirements.

[0035] The teacher-side module offers both automatic and manual grading modes. Teachers can choose manual grading to independently grade papers, or use automatic grading for one-click grading. The automatic grading principle varies depending on the type of question. Exam questions mainly include objective and subjective questions: multiple choice, true / false, fill-in-the-blank, and short answer questions. Objective questions (multiple choice, true / false, and fill-in-the-blank) are graded by directly matching the answers to the correct answers; a perfect match earns points, otherwise no points are awarded. Subjective questions (fill-in-the-blank and short answer) are graded using TF feature vectors and the simhash algorithm to calculate the similarity of Chinese text, thus assigning corresponding scores based on the similarity. The simhash algorithm uses Hamming distance to compare the similarity of two feature vectors. Intelligent grading and score analysis are achieved by directly comparing answers for multiple choice and true / false questions, thus enabling automatic grading. Furthermore, the teacher-side module allows for comprehensive analysis of student exam results: exam analysis, exam situation analysis, and student analysis.

[0036] The student-side module employs two methods to prevent cheating and achieve intelligent proctoring: first, monitoring the examination machine, using front-end JavaScript to disable keyboard copy-paste, screen switching, and implement full-screen mode to prevent cheating during web-based exams; second, monitoring the examinee, using MediaPipe, OpenCV, and TensorFlow to implement facial recognition verification for system login and real-time facial monitoring during the exam. Before the exam, in addition to a username and password, ArcSoft AI's ArcFace technology compares the examinee's facial information in the facial database with the logged-in facial information; only after a successful facial match can the examinee enter the examination system. During the exam, facial detection for cheating is mainly based on two types. Specifically, 1) Real-time facial recognition based on ArcFace technology. The camera is on throughout the exam, periodically collecting and recognizing the examinee's facial information. If a face is obscured, moves out of the camera's area, or the camera captures an image of a non-examinee's face for more than 2 seconds, the system determines that the examinee has potential cheating behavior and issues a warning. 2) Binocular depth measurement based on the Mediapipe iris tracking model detects the distance between the examinee's eyes and the exam screen. This data is used to determine if the examinee is looking around or moving away from (or closer to) the camera. Under normal exam conditions, the distance between the examinee's eyes and the exam screen is within a certain range. If the examinee looks around or moves their face away from the exam screen during the exam, and this continues for more than two days, the depth values ​​'s' of the left and right irises are used to determine if the examinee is potentially cheating and a warning is issued. Simultaneously, the exam machine's intelligent monitoring during the exam includes: initially preventing screen switching by setting the exam paper to full screen; mouse monitoring to ensure the mouse moves within the valid area of ​​the exam screen to prevent cheating through a second screen; and keyboard monitoring to check for the use of prohibited shortcut keys, such as screen switching, copy / paste (which is disabled during the exam), and zooming. If the examinee engages in any of these invalid exam behaviors, a warning will be issued, and the number of warnings will accumulate. When the number of warnings reaches a set threshold, the examinee will be forced to submit their exam.

[0037] The training and examination system has the following functions:

[0038] (I) General functions: 1. User registration and login; 2. System main page design; 3. Announcement and notification posting; 4. Communication circle function to facilitate nurses to exchange experiences and knowledge.

[0039] (II) Training Center: 1. Training introduction page; 2. Support for training courses in multiple formats (MP4, Word, Excel, PPT, PDF); 3. Viewing time statistics and interactive quizzes; 4. Training test papers and examination system; 5. Automatic issuance of qualification certificates.

[0040] (III) Test Paper Center: 1. Simulation practice function; 2. Official examination system, including anti-cheating measures and facial recognition technology; 3. Make-up examination mechanism; 4. Examination record management;

[0041] (iv) Personal Center: 1. Personal Information Management; 2. My KPI Tracking and Display; 3. OKR Completion Status.

[0042] The training and examination system's service functions include: system deployment and use: providing system installation, configuration, and initial deployment services; user training and system usage guidance; standard operation and maintenance services: regular system maintenance; technical support and fault response; security updates and backup management; personalized services: customizing functions according to the organization's needs; advanced customer support; and data analysis and reporting.

[0043] This invention is goal-oriented, ensuring that both training and examination content aim to achieve specific educational and practical objectives. Its quantitative assessment quantifies training effectiveness through key results from examinations and assessments. The transparency of its OKRs ensures all participants understand the training and assessment goals. Continuous improvement through regular review of OKRs allows the system to be continuously adjusted and optimized to enhance training quality. This invention offers high participation, allowing nurses to directly participate in setting their own individual OKRs, thereby increasing engagement and motivation. This invention provides a structured and systematic training platform to enhance nurses' professional capabilities. Through precise training, this invention addresses weaknesses in nurses' clinical practice. This invention improves the overall quality of nursing services, providing patients with a superior nursing experience.

[0044] In this invention, nurses can effectively master clinical theoretical knowledge and skills through this system to improve the quality of nursing services, reduce medical errors, and increase nurses' job satisfaction and career development. The implementation of this system will greatly promote the modernization of nursing education. The OKR-based nurse training and examination system aims to ensure that nurses' continuing education and professional development align with the goals of medical institutions, while simultaneously improving nursing quality and patient outcomes. By integrating OKR into the training and assessment process, progress can be better tracked, helping nurses improve their professional skills and ultimately enhancing the quality of medical services for the benefit of patients. This system creates a customized, high-efficiency nurse training system, contributing to the development of the nursing industry.

[0045] This invention incorporates a nursing education and assessment tool based on the OKR (Objectives and Key Results) framework. OKR is a popular management strategy used to set and track goals and their outcomes, commonly employed in businesses and organizations to improve efficiency and measurability of results. Applying OKR to a nurse training and examination system can help nursing education institutions and healthcare facilities ensure that nurses' training and development are aligned with organizational goals. By providing comprehensive online educational resources and functionalities, it strengthens nurses' clinical theoretical knowledge and practical skills. The system will utilize advanced information technology, combined with actual clinical nursing needs, to provide nurses with a personalized and precise learning experience to improve the quality of nursing services.

[0046] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A nurse training and examination system based on OKR, characterized in that: The OKR-based nurse training and examination system includes: General function modules; Training Center Module; Examination Center; Personal Center Module; The examination center is equipped with teacher-side and student-side modules.

2. The OKR-based nurse training and examination system as described in claim 1, characterized in that: The general function module includes the following internal features: user registration and login, system main page design, notification and announcement publishing, and a communication circle function to facilitate nurses to exchange experiences and knowledge.

3. The OKR-based nurse training and examination system as described in claim 1, characterized in that: The training center module includes:

1. Training introduction page; 2. Support for multi-format training courses; 3. Viewing time statistics and interactive quizzes; 4. Training test papers and examination system; 5. Automatic issuance of qualification certificates.

4. The OKR-based nurse training and examination system as described in claim 1, characterized in that: The examination center module includes a simulation practice function, a formal examination system, anti-cheating measures and facial recognition technology, a make-up exam module, and examination record management.

5. The OKR-based nurse training and examination system as described in claim 1, characterized in that: The personal center module includes: personal information management, my KPI tracking and display, and OKR completion status.

6. The OKR-based nurse training and examination system as described in claim 1, characterized in that: The training center module includes functions such as training management framework, system management, personnel management, and training quality management, with training quality management being the key focus of the system construction. Intelligent goal management and process management are the basic elements of the system's training quality management. Goal management is achieved through goal quality control and task progress management. The nursing department presets training goals at the individual and nursing unit levels. The system automatically compares the correctness of the teaching goals submitted by nursing units, dynamically displays the completion status of nursing unit teaching tasks and individual training tasks, and can remind users to complete tasks. Closed-loop management of teaching activities and full-process, full-element analysis provide support for training process management. Both teachers and students synchronize training and assessment information, connecting teaching management data from the nursing department, general departments, specialties, and nursing units. It will also automatically generate vertical and horizontal comparison results to achieve continuous improvement of training quality through "evaluation-analysis-improvement".

7. The OKR-based nurse training and examination system as described in claim 1, characterized in that: The training center module features internal resource sharing: the system integrates multi-disciplinary teaching resources and enables resource sharing; resource management is achieved through classification, quality, and access control; the system covers operational skill standards and videos, industry standards, courseware resources, question banks, case studies, evaluation forms, and other resources, and resource quality management is achieved through resource review and user feedback; users can browse and download teaching resources within their authorized scope.

8. The OKR-based nurse training and examination system as described in claim 1, characterized in that: The teacher-side module comprises three aspects: efficient question bank management, diverse test paper generation, and diverse marking methods. Question bank management includes diverse question types such as multiple choice, fill-in-the-blank, true / false, short answer, essay, and analytical questions. Teachers can manage the question bank online, including adding, editing, and deleting questions. New questions can be entered manually one by one or automatically imported in batches using templates. When adding questions, teachers can set the difficulty, score, and answer key, with difficulty levels ranging from 1 to 5. Diverse test paper generation includes both manual and intelligent methods. Manual test paper generation is primarily performed by teachers who manually select questions from the question bank according to exam requirements. Manual test paper generation involves selecting questions and assembling them into a suitable test paper. Manual test paper generation is inefficient, susceptible to human error, and prone to leaks. Intelligent test paper generation allows teachers to set the test paper's composition, question difficulty, generation logic, and exam duration. The system automatically selects and combines questions from the question bank according to these requirements. Intelligent test paper generation is highly efficient and offers greater randomness in question selection, reducing the risk of leaks to some extent. Clicking the "Customize" button in intelligent test paper generation displays the following customization options, allowing users to select question difficulty, question type, number of questions, and question score according to their needs. Clicking "Generate" will generate the test questions based on the customized requirements.

9. The OKR-based nurse training and examination system as described in claim 1, characterized in that: The teacher-side module offers both automatic and manual grading modes. Teachers can choose manual grading to independently grade papers, or use automatic grading for one-click grading. The automatic grading principle varies depending on the type of question. Exam questions mainly include objective and subjective questions: multiple choice, true / false, fill-in-the-blank, and short answer. Objective questions are graded by directly matching the answers to the correct answers; a perfect match earns points, otherwise no points are awarded. Subjective questions are graded using TF feature vectors and the simhash algorithm to calculate the similarity of Chinese text, thus assigning corresponding scores based on the similarity. When comparing the similarity of two feature vectors using the simhash algorithm, Hamming distance is employed. Intelligent grading and performance analysis are implemented by directly comparing answers for multiple choice and true / false questions, thus achieving automatic grading. Furthermore, the teacher-side module can provide comprehensive analysis of student exam results: exam analysis, exam situation analysis, and student analysis.

10. The OKR-based nurse training and examination system as described in claim 1, characterized in that: The student-side module employs two methods to prevent cheating and achieve intelligent proctoring. One method is to monitor the examination machine, which uses front-end JavaScript to prevent cheating during web examinations by disabling keyboard copy and paste, screen switching, and enabling full-screen mode during the examination. Secondly, candidate monitoring is implemented through methods such as Mediapipe, OpenCV, and TensorFlow to achieve facial recognition verification for login to the examination system and real-time facial monitoring during the examination. Before the examination, not only a username and password are required, but also ArcFace technology provided by ArcSoft AI is used to compare the candidate's facial information in the facial database with the candidate's login facial information. Only after a successful facial match can the candidate enter the examination system. During the examination, the criteria for judging cheating based on facial detection are mainly classified into two types. Specifically, 1) Real-time facial recognition based on ArcFace technology: The camera is turned on throughout the examination, and the candidate's facial information is collected and recognized at regular intervals. If a face is obscured, a face moves out of the camera area, or the camera captures a face image of a non-candidate for more than 2 seconds during the examination, the system judges that the candidate has potential cheating behavior and issues a reminder. 2) Binocular depth measurement based on the Mediapipe iris tracking model monitors the distance between the examinee's eyes and the exam screen. This data is used to determine if the examinee is looking around or moving away from the camera. Under normal exam conditions, the distance between the examinee's eyes and the exam screen is within a certain range. If the examinee looks around or moves their face away from the exam screen during the exam, and this continues for more than two days, the depth values ​​(s) of the left and right irises are used to determine if the examinee is potentially cheating and a warning is issued. Simultaneously, the intelligent monitoring of the exam machine includes: initially preventing screen switching by setting the exam paper to full screen; mouse monitoring to prevent cheating through a second screen by monitoring whether the mouse moves within the effective area of ​​the exam screen; and keyboard monitoring to monitor whether prohibited shortcut keys such as screen switching, copy / paste, and zoom are used during the exam. If a candidate engages in any of the aforementioned invalid examination behaviors during the examination, a reminder will be issued, and the number of reminders will be accumulated. When the number of reminders reaches a set threshold, the candidate will be forced to submit their paper.

Images