Artificial intelligence gamification interview training real-time feedback system

Abstract

The invention discloses an artificial intelligence gamification interview training real-time feedback system, which relates to the technical field of feedback systems, analyzes a feedback curve of a job seeker after each round of interview through a difficulty dynamic adjustment module, dynamically adjusts the difficulty of an interview problem by adopting a Bayesian optimization algorithm, and determines the interview problem of the job seeker according to the performance of the job seeker in different difficulty interview problems. The personalized module outputs comprehensive performance scores for the job seekers, the personalized module dynamically generates personalized interview scenes by adopting a programmed content generation method according to the comprehensive performance scores, and records interview conditions of the job seekers in different interview scenes, and the feedback module combines the interview conditions of the job seekers in different interview scenes and different interview difficulties to generate personalized interview data. And feeding back the overall interview performance of the job seeker and generating a visual report. The feedback system adopts Bayesian optimization to dynamically adjust the difficulty of the interview problem, and optimizes the difficulty in real time according to the feedback curve of the job seeker, so that the challenge is moderate, and the stable improvement of the learning curve is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of feedback systems, and particularly to a real-time feedback system for gamified interview training of artificial intelligence. Background Art

[0002] In the current environment of increasingly fierce competition in the job market, the interview performance of job seekers directly affects their chances of obtaining an ideal position. The feedback system, based on artificial intelligence technology and combined with gamified interaction design, creates a highly immersive, personalized, and efficient interview training environment, which is mainly applied to college students seeking jobs, career changers, and working professionals who hope to improve their interview skills.

[0003] The existing technologies have the following defects:

[0004] 1. Existing interview training systems usually adopt a fixed question set or a simple difficulty grading (such as beginner, intermediate, advanced), and cannot dynamically adjust according to the actual performance of job seekers, resulting in job seekers either feeling too easy and unchallenging during the training process, or facing excessive difficulty, leading to increased frustration and anxiety.

[0005] 2. Existing AI interview training systems only provide a limited number of interview questions, and most of the questions lack industry pertinence, resulting in job seekers being difficult to obtain effective training during the actual job search process. Moreover, since virtual interviewers are usually static, different job seekers face the same AI interviewer, lacking real situational changes (such as high-pressure interviews, friendly interviews, behavioral interviews, etc.), and unable to simulate the questioning methods of different roles such as corporate HR, technical supervisors, and CEOs, with poor adaptability.

[0006] Based on this, the present invention proposes a real-time feedback system for gamified interview training of artificial intelligence, which uses Bayesian optimization to dynamically adjust the difficulty of interview questions, and optimizes the difficulty in real time according to the feedback curve of job seekers, making the challenge moderate and ensuring a stable improvement of the learning curve. Summary of the Invention

[0007] The purpose of the present invention is to provide a real-time feedback system for gamified interview training of artificial intelligence to solve the deficiencies in the background art.

[0008] To achieve the above purpose, the present invention provides the following technical solution: A real-time feedback system for gamified interview training of artificial intelligence, including an initialization module, a difficulty dynamic adjustment module, a personalized module, and a feedback module;

[0009] Initialization module: used to record job seeker information, obtain the initial interview performance of job seekers, and set the initial interview difficulty;

[0010] Difficulty Dynamic Adjustment Module: After each round of interviews, analyze the feedback curve of the job seeker, dynamically adjust the difficulty of interview questions using the Bayesian optimization algorithm, and output a comprehensive performance score for the job seeker based on the job seeker's performance on interview questions of different difficulties;

[0011] Personalization Module: Dynamically generate a personalized interview scenario using a programmatic content generation method based on the comprehensive performance score, and record the interview status of the job seeker in different interview scenarios;

[0012] Feedback Module: Combine the interview status of the job seeker in different interview scenarios and different interview difficulties, provide feedback on the overall interview performance of the job seeker and generate a visual report.

[0013] Preferably, after each round of interviews, the Difficulty Dynamic Adjustment Module collects and analyzes multi-dimensional data of the job seeker to draw a feedback curve;

[0014] Input the feedback data of the job seeker, generate a performance score for the job seeker based on the feedback data, compare the performance score of the job seeker with a preset performance threshold. If the performance score of the job seeker is less than the performance threshold, it is determined that the comprehensive performance of the job seeker is poor and there is no need to increase the interview difficulty. If the performance score of the job seeker is greater than or equal to the performance threshold, it is determined that the comprehensive performance of the job seeker is good, increase the next-level interview difficulty, calculate the performance score of the job seeker in each interview difficulty until, in any interview difficulty, the performance score is less than the performance threshold or the interview difficulty is the maximum difficulty;

[0015] After obtaining the performance scores of the job seeker at different interview difficulties, obtain the comprehensive performance score of the job seeker through weighted calculation.

[0016] Preferably, after the Difficulty Dynamic Adjustment Module obtains the performance scores of the job seeker at different interview difficulties, calculate the comprehensive performance score of the job seeker, and the expression is: , where is the comprehensive performance score, is the number of interview difficulties, is the performance score at the i-th interview difficulty, is the weight of the i-th interview difficulty.

[0017] Preferably, the Difficulty Dynamic Adjustment Module generates a performance score for the job seeker based on the feedback data. The calculation logic of the performance score is: obtain the answer correct rate, language fluency, and emotional fluctuation index, perform normalization processing on the answer correct rate, language fluency, and emotional fluctuation index to map the value ranges of the answer correct rate, language fluency, and emotional fluctuation index to [0,1], and obtain the performance score by adding the normalized answer correct rate, adding the language fluency, and subtracting the emotional fluctuation index.

[0018] Preferably, the personalized module uses programmatic content generation to dynamically generate personalized interview scenarios and dynamically adjusts the interview content based on the comprehensive performance score of the job seeker;

[0019] After obtaining the comprehensive performance score of the job seeker, a personalized interview scenario is generated based on the comprehensive performance score and the personalized interview scenario table.

[0020] Preferably, the feedback module obtains the comprehensive performance score of the job seeker under the adjustment of the interview difficulty and obtains the comprehensive performance score of the job seeker under the adjustment of the interview scenario;

[0021] The comprehensive performance score of the job seeker under the adjustment of the interview difficulty and the comprehensive performance score of the job seeker under the adjustment of the interview scenario are weighted and calculated to obtain the overall performance index of the job seeker.

[0022] Preferably, the calculation expression of the overall performance index is: , where is the overall performance index, is the comprehensive performance score under the adjustment of the interview difficulty, is the comprehensive performance score under the adjustment of the interview scenario, , are weights, and .

[0023] Preferably, the calculation expression of the language fluency is: L = 1 / KG, where L is the language fluency and KG is the number of answer pauses;

[0024] The calculation logic of the emotional fluctuation index is: dividing the number of emotional changes of the user by the monitoring duration to obtain the emotional fluctuation index;

[0025] The calculation logic of the answer correct rate is: dividing the number of correct answers by the total number of answers to obtain the answer correct rate.

[0026] Preferably, when the job seeker first enters the system, the initialization module records the information related to the job seeker's interview, including but not limited to: career direction, interview experience, target position, and target industry;

[0027] The initial interview performance of the job seeker is obtained through multi-dimensional evaluation by AI, and based on the initial performance of the job seeker, the initial interview difficulty is determined:

[0028] If the number of interviews of the job seeker is less than the first number threshold and the interview experience is judged to be less, the primary difficulty is set;

[0029] If the number of interviews of the job seeker is greater than or equal to the first number threshold and less than the second number threshold, and the interview experience is judged to be average, the intermediate difficulty is set;

[0030] If the number of interviews of a job applicant is greater than or equal to the second threshold, it is determined that the applicant has rich interview experience, and the advanced difficulty level is set.

[0031] In the above technical solution, the technical effects and advantages provided by the present invention are as follows:

[0032] After each round of interviews, the present invention analyzes the feedback curve of the job applicant through the difficulty dynamic adjustment module, dynamically adjusts the difficulty of interview questions by using the Bayesian optimization algorithm, and outputs a comprehensive performance score for the job applicant based on the performance of the applicant in interview questions of different difficulties. The personalized module dynamically generates a personalized interview scenario by using the programmed content generation method based on the comprehensive performance score, and records the interview status of the applicant in different interview scenarios. The feedback module combines the interview status of the applicant in different interview scenarios and different interview difficulties, feedbacks the overall interview performance of the applicant and generates a visual report. The feedback system dynamically adjusts the difficulty of interview questions by using Bayesian optimization, and optimizes the difficulty in real time according to the feedback curve of the job applicant, making the challenge moderate and ensuring the stable improvement of the learning curve. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required in the embodiments. Obviously, the drawings in the following description are only some embodiments recorded in the present invention, and those of ordinary skill in the art can also obtain other drawings based on these drawings.

[0034] Figure 1 It is the system architecture diagram of the present invention.

[0035] Figure 2 It is the mind map of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0036] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0037] Embodiment 1: Please refer to Figure 1 and Figure 2 as shown, which is described in this embodiment.

[0038] After each round of interviews, the present application uses a difficulty dynamic adjustment module to analyze the feedback curve of the job seeker, dynamically adjusts the difficulty of the interview questions using the Bayesian optimization algorithm, and outputs a comprehensive performance score for the job seeker based on the job seeker's performance on interview questions of different difficulties. The personalization module dynamically generates a personalized interview scenario using a procedural content generation method based on the comprehensive performance score, and records the interview status of the job seeker in different interview scenarios. The feedback module combines the interview status of the job seeker in different interview scenarios and different interview difficulties, provides feedback on the overall interview performance of the job seeker, and generates a visual report. The feedback system dynamically adjusts the difficulty of the interview questions using Bayesian optimization, and optimizes the difficulty in real time according to the feedback curve of the job seeker to make the challenge moderate and ensure a stable improvement of the learning curve.

[0039] The specific working process of the feedback system is as follows:

[0040] Record the basic information of the job seeker (such as career direction, interview experience, target position, etc.), obtain the initial interview performance of the job seeker, including indicators such as voice, language expression, facial expression, logical thinking, etc., set the initial interview difficulty through Bayesian optimization to ensure moderate challenge. After each round of interviews, analyze the feedback curve of the job seeker, dynamically adjust the difficulty of the interview questions using the Bayesian optimization algorithm, and output a comprehensive performance score for the job seeker based on the job seeker's performance on interview questions of different difficulties. Dynamically generate a personalized interview scenario using a procedural content generation method (PCG) based on the comprehensive performance score: including changes in virtual interviewers (such as serious, high-pressure, friendly HR). Changes in question types (such as open-ended questions, behavioral interviews, situational interviews, stress tests). Industry relevance (adjust questions to be more in line with the target industry of the job seeker). For example: In the primary stage: Provide standardized questions and a friendly AI interviewer. In the intermediate stage: Introduce more challenging conversations, such as rhetorical questions, in-depth probing, etc. In the advanced stage: Simulate a real corporate interview, such as the questioning method of a technical supervisor or CEO. Combine the interview status of the job seeker in different interview scenarios and different interview difficulties, provide feedback on the overall interview performance of the job seeker, and generate a visual report.

[0041] Example 2: The initialization module records the basic information of the job seeker (such as career direction, interview experience, target position, etc.), obtains the initial interview performance of the job seeker, including indicators such as voice, language expression, facial expression, logical thinking, etc., sets the initial interview difficulty to ensure moderate challenge, and sends the initial interview difficulty to the difficulty dynamic adjustment module.

[0042] When a job seeker first enters the system, the initialization module records their personal interview-related information, including but not limited to: career direction (such as software development, marketing, product management, etc.), interview experience (no experience / little experience / rich experience (judged by the comparison result of the number of interviews and the corresponding threshold)), target position (such as data analyst, front-end engineer, HR manager, etc.), target industry (such as Internet, finance, manufacturing, etc.).

[0043] Obtain the initial interview performance of the job seeker and conduct multi-dimensional evaluations through AI, including but not limited to: speech expression: evaluate speaking speed, clarity, fluency, number of pauses, etc.; language organization ability: analyze grammar errors, logical coherence, keyword usage; facial expression: detect eye contact, micro-expressions, body language, etc.; logical thinking ability: test the coherence of answers, information completeness, argumentation ability; stress response: judge the emotional changes of the job seeker when facing high-pressure questions.

[0044] Based on the initial performance of the job seeker, determine the most suitable initial interview difficulty to ensure that the training is neither boring nor overly difficult:

[0045] If the number of interviews of the job seeker is less than the first number threshold, it is judged that the interview experience is less. If the performance is relatively nervous, set the primary difficulty and provide a more friendly AI interviewer and basic questions.

[0046] If the number of interviews of the job seeker is greater than or equal to the first number threshold and less than the second number threshold, it is judged that they have certain interview experience. If their logical thinking is strong, set the intermediate difficulty and introduce more challenging in-depth questions.

[0047] If the number of interviews of the job seeker is greater than or equal to the second number threshold, it is judged that they have rich interview experience, then set the advanced difficulty, simulate a real enterprise interview environment, and introduce high-pressure situations.

[0048] Once the initial interview difficulty is determined, this information will be transmitted to the difficulty dynamic adjustment module. During the subsequent training process, this module will continuously optimize the interview difficulty according to the feedback curve of the job seeker, making the training path more intelligent and personalized.

[0049] This process ensures that the interview training system can customize a training plan for each job seeker, enabling them to effectively improve at the optimal difficulty.

[0050] After each round of interviews, the difficulty dynamic adjustment module analyzes the feedback curve of the job seeker, dynamically adjusts the difficulty of interview questions using the Bayesian optimization algorithm, and outputs a comprehensive performance score for the job seeker based on their performance in interview questions of different difficulties. The comprehensive performance score is sent to the personalized module and the feedback module.

[0051] After each round of interviews, the difficulty dynamic adjustment module collects and analyzes multi-dimensional data of job seekers to draw a feedback curve. The feedback curve consists of multiple key indicators, including but not limited to:

[0052] Answer accuracy rate (such as whether technical or situational questions can be answered accurately);

[0053] Language fluency (number of pauses, whether the speech rate is balanced);

[0054] Logical coherence (whether the answer is hierarchical and whether the arguments and evidence are sufficient);

[0055] Facial expressions and emotional changes (tension, confidence, eye contact);

[0056] Stress adaptation ability (performance under high-pressure questions, such as whether there is incoherence or obvious avoidance);

[0057] These data are used to draw the growth curve of job seekers, and to judge their progress trend, bottleneck points, and the most suitable difficulty range during the training process.

[0058] Input the feedback data of job seekers (such as answer accuracy rate, language fluency, emotional fluctuation index), and generate the performance score of job seekers based on the feedback data. The calculation logic of the performance score is: obtain the answer accuracy rate, language fluency, and emotional fluctuation index, normalize the answer accuracy rate, language fluency, and emotional fluctuation index, map the value range of the answer accuracy rate, language fluency, and emotional fluctuation index to [0,1], add the normalized answer accuracy rate, language fluency, and subtract the emotional fluctuation index to obtain the performance score. The larger the performance score, the better the comprehensive performance of the job seeker.

[0059] The calculation expression of language fluency is: L = 1 / KG, where L is language fluency and KG is the number of answer pauses.

[0060] The calculation logic of the emotional fluctuation index is: divide the number of emotional changes of the user by the monitoring duration to obtain the emotional fluctuation index.

[0061] The calculation logic of the answer accuracy rate is: divide the number of correct answers by the total number of answers to obtain the answer accuracy rate.

[0062] Compare the performance score of the job seeker with the preset performance threshold. If the performance score of the job seeker is less than the performance threshold, it is judged that the comprehensive performance of the job seeker is poor and there is no need to increase the interview difficulty. If the performance score of the job seeker is greater than or equal to the performance threshold, it is judged that the comprehensive performance of the job seeker is good, increase to the next-level interview difficulty, calculate the performance score of the job seeker in each interview difficulty until in any interview difficulty, the performance score is less than the performance threshold or the interview difficulty is the maximum difficulty.

[0063] After the difficulty dynamic adjustment module obtains the performance scores of the job seeker under different interview difficulties, it calculates the comprehensive performance score of the job seeker. The expression is as follows: , where is the comprehensive performance score, is the number of interview difficulties, is the performance score under the i-th interview difficulty, is the weight of the i-th interview difficulty. Generally speaking, the greater the interview difficulty, the greater the weight. For example, there are interview difficulties A, B, and C, and the interview difficulty A < B < C, then the weights are A = 0.2, B = 0.35, C = 0.45, and so on.

[0064] The personalized module dynamically generates a personalized interview scenario based on the comprehensive performance score using the procedural content generation method (PCG): including changes in virtual interviewers (such as serious, high-pressure, friendly HR). Changes in question types (such as open-ended questions, behavioral interviews, situational interviews, stress tests). Industry relevance (adjusting questions to be more in line with the target industry of the job seeker). For example: In the primary stage: Provide standardized questions and a friendly AI interviewer. In the intermediate stage: Introduce more challenging conversations, such as rhetorical questions, in-depth probing, etc. In the advanced stage: Simulate a real enterprise interview, such as the questioning method of a technical supervisor or CEO level, record the interview status of the job seeker in different interview scenarios, and send the interview status of the job seeker in different interview scenarios to the feedback module.

[0065] The core goal of this module is to dynamically adjust the interview scenario based on the comprehensive performance score of the job seeker using the procedural content generation (PCG) technology to ensure that the training experience of the job seeker is more real and targeted. At the same time, record the performance of the job seeker in different interview scenarios to provide data support for subsequent feedback.

[0066] The comprehensive performance score is calculated by the difficulty dynamic adjustment module, and this score determines the current interview ability level of the job seeker. The system selects a suitable training mode based on the score and adjusts elements such as the type of interviewer, question type, and industry matching degree.

[0067] The personalized module uses procedural content generation to dynamically generate a personalized interview scenario, and dynamically adjusts the interview content based on the comprehensive performance score of the job seeker. The specific adjustment dimensions are as follows:

[0068] After obtaining the comprehensive performance score of the job seeker, a personalized interview scenario is generated based on the comprehensive performance score and the personalized interview scenario table. This table divides the job seeker's comprehensive performance score into different intervals based on the score and matches a corresponding personalized interview scenario for each interval to measure the overall interview ability of the job seeker. The higher the performance score, the more complex and stressful the interview scenario. The personalized interview scenario table is shown in Table 1:

[0069] Personalized Interview Scenario Table Table 1

[0070] Comprehensive performance score Performance score range Personalized interview scenario Virtual interviewer type Question type Industry relevance Interview characteristics 0.00-0.20 Needs significant improvement Basic communication training scenario Friendly HR Standardized questions Low Focus on guiding the job seeker to overcome nervousness and improve basic expression ability 0.20-0.40 Basic interview ability Basic interview simulation Approachable HR Common behavioral interview questions Medium Provide structured guidance to examine logical thinking and confidence 0.40-0.60 Have certain abilities Advanced interview challenges Ordinary HR Industry common questions, situational questions High Appropriately follow up to examine adaptability and problem-solving ability 0.60-0.80 Strong interview ability High-pressure interview simulation Serious supervisor Counter-question, in-depth follow-up, industry challenges High Simulate the real workplace to examine stress resistance and professional qualities 0.80-0.95 Excellent interviewee Professional-level interview confrontation Department head In-depth business, case analysis Very high Emphasize industry knowledge, innovation ability, and test leadership 0.95-1.00 Top job seeker Final interview at the CEO level CEO / Founder Strategic perspective, industry forward-looking questions Highest Examine the job seeker's overall thinking and leadership potential, extremely high difficulty

[0071] Virtual interviewer type:

[0072] Friendly HR: Suitable for beginners, relieves anxiety and boosts confidence.

[0073] Ordinary HR: A common interviewer, provides industry-standard questions.

[0074] Serious supervisor: More challenging, tests the on-the-spot reaction of job seekers.

[0075] CEO / Founder: The highest difficulty, examines strategic thinking and leadership.

[0076] Question type:

[0077] Standardized questions: Basic questions to help job seekers get familiar with the interview process.

[0078] Behavioral interview questions: Examine the past experiences and coping abilities of job seekers.

[0079] Industry challenge questions: Focus on the target industry of job seekers and examine professionalism.

[0080] CEO-level questions: Require job seekers to think from the perspective of the overall enterprise.

[0081] This personalized interview scenario form ensures that job seekers can continuously be challenged while adapting to their own abilities, thus improving their interview performance.

[0082] The feedback module combines the interview situations of job seekers under different interview scenarios and different interview difficulties, and gives feedback on the overall interview performance of job seekers and generates a visual report.

[0083] The core goal of this module is to combine the performance of job seekers under different interview scenarios and different interview difficulties, provide all-round feedback, and generate a visual report to help job seekers accurately identify their own advantages and improvement points, and improve the interview success rate.

[0084] The feedback module obtains the comprehensive performance score of job seekers under the adjustment of interview difficulty, and obtains the comprehensive performance score of job seekers under the adjustment of interview scenario (the calculation method is the same as the comprehensive performance score under the adjustment of interview difficulty, which will not be elaborated in this application);

[0085] The comprehensive performance score of job seekers under the adjustment of interview difficulty and the comprehensive performance score of job seekers under the adjustment of interview scenario are weighted and calculated to obtain the overall performance index of job seekers. The expression is:

[0086] , where is the overall performance index, The comprehensive performance score under the adjusted interview difficulty The comprehensive performance score under the adjusted interview scenario 、 are weights, and 。

[0087] The larger the overall performance index, the better the overall interview performance of the job seeker. The following is the visualization report generated based on the overall performance index:

[0088] Before generating the visualization report, it is first necessary to obtain the basic information of the job seeker, including the job seeker's career direction, target position, interview experience, etc. This information helps to personalize the subsequent report and make it more in line with the actual needs of the job seeker.

[0089] During the interview process, the system will calculate the comprehensive performance score under the adjusted interview difficulty and the comprehensive performance score under the adjusted interview scenario respectively, and obtain the overall performance index. Among them, the comprehensive performance score under the adjusted interview difficulty reflects the performance of the job seeker in the interview questions of different difficulties, and the comprehensive performance score under the adjusted interview scenario reflects the adaptability of the job seeker in different interview environments. The system will calculate the overall performance index by combining the weights of the two scores, and this index is used to measure the overall interview ability of the job seeker.

[0090] The system evaluates the performance of the job seeker based on the overall performance index of the job seeker and in combination with the score range. For example, if the overall performance index is close to the full score, it means that the job seeker already has strong interview ability and can be competent for high-difficulty interviews; if the overall performance index is low, it means that the job seeker still needs further training in some aspects. The evaluation content mainly covers the following aspects:

[0091] Speech and language expression ability, including speech rate, speech clarity, expression fluency, pausing situation, etc. Logical thinking ability, including the logic of expression, argumentation ability, structural integrity of answers, etc. Facial expressions and body language, including whether the expressions are natural, whether the eye contact is good, whether the gestures are appropriate, etc. Ability to handle industry-related questions, including the mastery of professional knowledge, case analysis ability, adaptability, etc.

[0092] After evaluating the overall performance of job seekers, the system provides specific optimization suggestions for different score ranges. For example: If the overall performance index of a job seeker is low, the system may suggest that they strengthen language expression practice, improve logical thinking ability, and provide some basic interview training, such as simulating the answering of standardized questions. If the overall performance index of a job seeker is at a medium level, the system will suggest that they try more challenging interviews, such as facing in-depth questioning or a high-pressure interview environment, to improve their on-the-spot response ability. If the overall performance index of a job seeker is high, the system will suggest that they conduct more advanced simulation training, such as in-depth interviews at the technical supervisor or CEO level, to further enhance their competitiveness.

[0093] To enable job seekers to more intuitively understand their progress, the system generates a visual trend chart based on the scores of different interview sessions. For example: A line chart is used to show the changing trend of the comprehensive performance index of a job seeker in multiple interview stages, helping the job seeker understand their progress. A radar chart is used to show the performance of a job seeker in different ability dimensions (such as language expression, logical thinking, body language, industry knowledge, etc.), enabling the job seeker to clearly see their strengths and weaknesses.

[0094] Based on the results of the visual report, the system recommends the next training strategy for the job seeker. For example: If the language expression ability of a job seeker is weak, the system will recommend relevant language training courses to improve the job seeker's fluency and expression confidence. If the logical thinking ability of a job seeker needs to be improved, the system will recommend behavioral interview skill training, such as optimizing the answering logic using the STAR method (Situation, Task, Action, Result). If the industry professionalism of a job seeker is insufficient, the system will recommend mock interviews for the target industry to improve the job seeker's professional knowledge reserve and industry response ability.

[0095] The visual report is not only a summary of the job seeker's current interview performance but also a guiding basis for their future interview training. The system dynamically adjusts the interview training plan based on the job seeker's performance. For example: If the performance index of a job seeker has increased rapidly in recent interviews, the system may increase the difficulty of the interview questions to enhance the challenge. If a job seeker continues to perform weakly in a certain interview session, the system may provide special training for that session, such as in-depth practice for stress interviews, behavioral interviews, or technical interviews.

[0096] Through the above steps, the system can comprehensively evaluate the interview ability of job seekers and provide them with targeted optimization solutions, ultimately helping job seekers achieve better interview performance in the actual job search process.

[0097] The above formulas are all dimensionless and take their numerical values for calculation. The formula is obtained by collecting a large amount of data for software simulation to get a formula closest to the actual situation. The preset parameters in the formula are set by those skilled in the art according to the actual situation.

[0098] In the description of this specification, the descriptions referring to terms such as "one embodiment", "example", "specific example", etc. mean that the specific features, structures, materials or characteristics described in connection with that embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in a suitable manner in any one or more embodiments or examples.

[0099] The preferred embodiments of the present invention disclosed above are only used to help illustrate the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the present invention to only the specific embodiments. Obviously, many modifications and variations can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art in the relevant technical field can well understand and utilize the present invention. The present invention is only limited by the claims and their full scope and equivalents.

Claims

1. A real-time feedback system for gamified interview training of artificial intelligence, characterized in that: It includes an initialization module, a difficulty dynamic adjustment module, a personalization module, and a feedback module; Initialization module: Used to record job seeker information, obtain the initial interview performance of the job seeker, and set the initial interview difficulty; Difficulty dynamic adjustment module: After each round of interview, analyze the feedback curve of the job seeker, dynamically adjust the difficulty of interview questions using the Bayesian optimization algorithm, and output a comprehensive performance score for the job seeker based on the performance of the job seeker in interview questions of different difficulties; Personalization module: Dynamically generate personalized interview scenarios using a programmatic content generation method based on the comprehensive performance score, and record the interview status of the job seeker in different interview scenarios; Feedback module: Combine the interview status of the job seeker in different interview scenarios and different interview difficulties, feedback the overall interview performance of the job seeker and generate a visual report.

2. The real-time feedback system for gamified interview training of artificial intelligence according to claim 1, characterized in that: After each round of interview ends, the difficulty dynamic adjustment module collects and analyzes multi-dimensional data of the job seeker and draws a feedback curve; Input the feedback data of the job seeker, generate the performance score of the job seeker based on the feedback data, compare the performance score of the job seeker with a preset performance threshold. If the performance score of the job seeker is less than the performance threshold, it is judged that the comprehensive performance of the job seeker is poor and there is no need to increase the interview difficulty. If the performance score of the job seeker is greater than or equal to the performance threshold, it is judged that the comprehensive performance of the job seeker is good, increase the next-level interview difficulty, calculate the performance score of the job seeker in each interview difficulty until in any interview difficulty, the performance score is less than the performance threshold or the interview difficulty is the maximum difficulty; After obtaining the performance scores of the job seeker in different interview difficulties, obtain the comprehensive performance score of the job seeker through weighted calculation.

3. The real-time feedback system for gamified interview training of artificial intelligence according to claim 2, characterized in that: After the difficulty dynamic adjustment module obtains the performance scores of the job seeker under different interview difficulties, it calculates the comprehensive performance score of the job seeker. The expression is as follows: , where is the comprehensive performance score, is the number of interview difficulties, is the performance score under the i-th interview difficulty, is the weight of the i-th interview difficulty.

4. The real-time feedback system for gamified interview training of artificial intelligence according to claim 3, characterized in that: The difficulty dynamic adjustment module generates the performance score of the job seeker based on the feedback data. The calculation logic of the performance score is: obtain the answer correct rate, language fluency, and emotional fluctuation index, normalize the answer correct rate, language fluency, and emotional fluctuation index so that the value ranges of the answer correct rate, language fluency, and emotional fluctuation index are mapped to [0,1], and obtain the performance score by adding the normalized answer correct rate, adding the language fluency, and subtracting the emotional fluctuation index.

5. The real-time feedback system for gamified interview training of artificial intelligence according to claim 4, characterized in that: The personalization module uses programmatic content generation to dynamically generate personalized interview scenarios and dynamically adjusts the interview content based on the comprehensive performance score of the job seeker; After obtaining the comprehensive performance score of the job seeker, generate a personalized interview scenario based on the comprehensive performance score and the personalized interview scenario table.

6. The real-time feedback system for gamified interview training of artificial intelligence according to claim 5, wherein: The feedback module obtains the comprehensive performance score of the job seeker under the interview difficulty adjustment and obtains the comprehensive performance score of the job seeker under the interview scenario adjustment; Weightedly calculate the comprehensive performance score of the job seeker under the interview difficulty adjustment and the comprehensive performance score of the job seeker under the interview scenario adjustment to obtain the overall performance index of the job seeker.

7. The real-time feedback system for gamified interview training of artificial intelligence according to claim 6, characterized in that: The calculation expression of the overall performance index is as follows: , where is the overall performance index, is the comprehensive performance score under the adjustment of interview difficulty, is the comprehensive performance score under the adjustment of interview scenario, , are weights, and .

8. The real-time feedback system for gamified interview training of artificial intelligence according to claim 4, wherein: The calculation expression of the language fluency is: L = 1 / KG, where L is the language fluency and KG is the number of answer pauses; The calculation logic of the emotional fluctuation index is: divide the number of emotional changes of the user by the monitoring duration to obtain the emotional fluctuation index; The calculation logic of the answer correct rate is: divide the number of correct answers by the total number of answers to obtain the answer correct rate.

9. The real-time feedback system for gamified interview training of artificial intelligence according to claim 8, characterized in that: When a job seeker first enters the system, the initialization module records information related to the job seeker's interview, including but not limited to: career direction, interview experience, target position, and target industry; Obtain the initial interview performance of the job seeker through multi-dimensional evaluation by AI, and determine the initial interview difficulty based on the initial performance of the job seeker: If the number of interviews of the job seeker is less than the first number threshold and it is judged that the interview experience is less, set the primary difficulty; If the number of interviews of the job seeker is greater than or equal to the first number threshold and less than the second number threshold, and it is judged that the interview experience is average, set the intermediate difficulty; If the number of interviews of the job seeker is greater than or equal to the second number threshold and it is judged that the interview experience is rich, set the advanced difficulty.

Images