Interview answering state monitoring method and system

Abstract

The invention relates to the technical field of data processing, and discloses an interview answering state monitoring method and system, and the method comprises the steps: carrying out the analysis and arrangement of an interview request which is sent by an interview request end and aims at target recruitment information, and generating an interview task list containing the interview reservation window time and interview post correlation information; and obtaining idle time periods of a plurality of GPU servers, considering interview reservation window time in the interview task list and interview GPU demand information and interview monitoring demand information of links, solving an interview agreed time period corresponding to the GPU server allocated to each interview request end, and driving each GPU server to perform interview scheduling according to a received interview answer video stream. And executing interview answering state monitoring. Therefore, the GPU resources with adaptive computing power and up-to-standard delay are matched for each interview request, the contradiction between delay caused by computing power overmatching and resource waste caused by low computing power matching is solved, and the benefit of a third-party service mechanism is improved as much as possible while the user experience is ensured.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a method and system for monitoring the status of interview answers. Background Art

[0002] Interview status monitoring is a technical process based on multimodal perception technology and artificial intelligence analysis, which collects and dynamically analyzes the interviewee's behavior, language, environment, and progress during the interview process in real time. Its core is to obtain multi-dimensional data through the interview terminal, including video data, voice data, behavioral data, and answer data from the video stream. Combined with customized AI monitoring models for each position (such as code operation analysis models for technical positions and communication and appeal analysis models for sales positions), it can determine the compliance (e.g., cheating, focus) and effectiveness (e.g., logical thinking ability, professional competence matching) of the answer status in real time, and simultaneously output the status analysis results to provide objective evaluations for the interviewer.

[0003] Since recruiting companies typically lack dedicated GPU server resources for interview status monitoring, third-party service providers in the current industry face the challenges of facilitating interviews across multiple recruiting companies. In scenarios with high-volume interviews (such as regular interview schedules for large state-owned enterprises and conglomerates), recruiting companies need to complete numerous interviews in a short period of time. When providing interview status monitoring services, third-party service providers must consider each recruiting company's job requirements, recruitment timeframe, and number of positions required. This results in existing technologies facing challenges such as difficulty dynamically matching GPU computing power requirements, user experience degradation caused by over-allocation of computing power, resource cost waste caused by under-allocation of computing power, and high allocation complexity under dynamic demand. This impacts user experience (for example, the high latency of the analysis results returned by the interview status monitoring model prevents interviewers from providing real-time analysis and evaluation), and also impacts the profitability of third-party service providers (under-allocation of computing power reduces the number of recruiting companies they can serve, given limited GPU server resources).

[0004] Therefore, how to improve the intelligent allocation of GPU server resources of third-party service agencies when facing the auxiliary interview needs of multiple recruiting companies, and maximize the efficiency of third-party service agencies while ensuring user experience, is a technical problem that needs to be solved urgently. Summary of the Invention

[0005] The present invention provides a method and system for monitoring the status of interview answers, aiming to solve at least one of the above-mentioned technical problems.

[0006] To achieve the above object, the present invention provides a method for monitoring the status of interview questions, the method comprising the following steps: Receive interview requests for target recruitment information sent by the interview request end, and build an interview task list based on the interview appointment window time and interview position association information in the interview request; Generate interview GPU requirement information based on the interview answering status monitoring requirements corresponding to the interview position association information, and link it to the interview task list as the first additional requirement information; Generate interview monitoring requirement information based on the content of several interview questions corresponding to the interview position-related information, and connect it to the interview task list as second additional requirement information; Obtain the interview schedule for the target job posting, query the idle time slots of several GPU servers within the interview schedule, consider the interview appointment window time and the linked first and second additional requirements in the interview task list, and solve the interview request allocation strategy; Based on the interview appointment time period corresponding to the GPU server allocated to each interview request end in the interview request allocation strategy, the GPU server is driven to perform interview answering status monitoring on the interview request end.

[0007] Optionally, before receiving an interview request for a target recruitment information from an interview request terminal and building an interview task list based on the interview appointment window time and interview position association information in the interview request, the following steps may also be included: Obtain target recruitment information sent by the recruiting company and extract the recruitment interview requirements in the target recruitment information; wherein the recruitment interview requirements include the recruitment interview positions, the number of vacancies for each recruitment interview position, and the interview request delivery time period; Based on the recruitment interview positions and the number of recruitments for each recruitment interview position in the recruitment interview requirements, the first generation judgment condition of the interview task list is determined, and based on the interview request delivery time period in the recruitment interview requirements, the second generation judgment condition of the interview task list is determined.

[0008] Optionally, receive an interview request for the target recruitment information sent by the interview request end, and build an interview task list according to the interview appointment window time and interview position association information in the interview request, specifically including: Receive interview requests for target recruitment information sent by multiple interview request terminals, and extract the request submission time, interview appointment window time and interview position association information of the interview requests; Based on the first generation judgment condition determined by the interview position-related information of the interview request and the recruitment interview position and the number of vacancies in the recruitment interview requirements, it is judged whether the number of interview requests received for the recruitment interview position that matches the interview position-related information reaches the corresponding number of vacancies; if so, it is determined that the interview request meets the first generation judgment condition; determining, based on a second generation judgment condition determined by the request submission time of the interview request and the interview request delivery period in the recruitment interview requirements, whether the request submission time of the interview request falls within the interview request delivery period; and if so, determining that the interview request meets the second generation judgment condition; The interview appointment window time and interview position association information in the received interview request that meets both the first generation judgment condition and the second generation judgment condition are added to the constructed interview task list.

[0009] Optionally, based on the interview answering status monitoring requirements corresponding to the interview position-related information, generate interview GPU requirement information and link it to the interview task list as the first additional requirement information, specifically including: Determine the interview question answering status monitoring requirements based on the interview position type and the interview position ability requirement set in the interview position association information; wherein the interview question answering status monitoring requirements include a plurality of interview question answering status monitoring items; Based on the video analysis test data of several GPU servers used for interview answer video analysis, determine the candidate set of GPU servers whose GPU computing power quantization requirements and video analysis result feedback delay meet the interview experience requirements when executing the same interview answer status monitoring project; The candidate set of GPU servers includes several GPU servers whose video analysis result feedback delay parameter in the video analysis test data is lower than a preset delay threshold required by the interview experience; The GPU server candidate set and the GPU computing power quantification requirement parameters are aggregated to generate interview GPU requirement information, and the interview GPU requirement information is linked to the interview task list as the first additional requirement information corresponding to the interview request.

[0010] Optionally, based on the interview position type and the interview position ability requirement set in the interview position association information, determining the interview answer status monitoring requirement steps specifically include: Based on the interview position type and the interview position ability requirement set in the interview position association information, query the interview question answering status monitoring model database for an interview question answering status monitoring model that meets the monitoring requirements corresponding to the interview position type and the interview position ability requirement set; Query several interview question answering status monitoring items covered in the interview question answering status monitoring model, and use them as the interview question answering status monitoring requirements corresponding to the interview request.

[0011] Optionally, based on the content of several interview questions corresponding to the interview position-related information, interview monitoring requirement information is generated and connected to the interview task list step as second additional requirement information, specifically including: Extracting job type and job requirement keywords from the interview job association information and the interview job competency requirement set, and matching several interview question contents with the highest correlation with the job type requirement keywords in the corresponding interview question content set; According to the interview answering time requirement recorded in each interview question content, the estimated total interview answering time for each interview request is estimated, and the estimated total interview answering time is converted into interview monitoring demand information as the second additional demand information corresponding to the interview request and linked to the interview task list.

[0012] Optionally, the interview schedule for the target recruitment information is obtained, the idle time periods of several GPU servers within the interview schedule are searched, and the interview appointment window time and the linked first additional requirement information and second additional requirement information in the interview task list are considered to solve the interview request allocation strategy, specifically including: Obtain the interview schedule of the target recruitment information, query the idle time periods of several GPU servers within the interview schedule, and use the idle time periods and GPU computing power quantization parameters of each GPU server as the interview answering status monitoring resource information of the GPU server; Based on the interview answering status monitoring resource information of each GPU server, considering the interview appointment window time and the first additional demand information and the second additional demand information of the link in the interview task list, with the time window limit and GPU computing power limit as the constraints, and the highest uniformity of GPU computing power under the premise of minimizing GPU server resource usage as the optimization goal, an optimization algorithm is used to solve the GPU server assigned to each interview request end corresponding to the interview appointment time period; Generate an interview request allocation strategy for the target recruitment information based on the interview appointment time period corresponding to the GPU server assigned to each interview request end.

[0013] Optionally, based on the interview answering status monitoring resource information of each GPU server, the interview appointment window time and the first additional requirement information and the second additional requirement information of the link in the interview task list are considered, with the time window limit and the GPU computing power limit as constraints, and the highest uniformity of GPU computing power under the premise of minimizing GPU server resource usage as the optimization goal, an optimization algorithm is used to solve the steps of the GPU server assigned to each interview request end corresponding to the interview appointment time period, specifically including: Based on the GPU computing power quantization parameters of each GPU server and the idle time period in the interview answering status monitoring resource information, the interview appointment window time, GPU server candidate set, GPU computing power quantization requirement parameters and the estimated total interview answering time of each interview request are considered; After the interview request of each interview request end is assigned to the corresponding GPU server corresponding to the interview appointment time period, the first constraint condition is that the interview request of each interview request end is assigned to the GPU server within the candidate set of GPU servers corresponding to the interview request; the second constraint condition is that the interview appointment time period assigned to each interview request falls within the intersection of the interview appointment window time corresponding to the interview request and the idle time period of the corresponding GPU server; the third constraint condition is that the sum of the GPU computing power quantization requirement parameters of all interview requests assigned to each GPU server at each moment is not higher than the GPU computing power quantization parameters of the GPU server; the optimization goal is to minimize the variance of the GPU computing power requirement ratio of each GPU server assigned to the interview request under the premise of minimizing the number of GPU servers assigned to the interview request; A heuristic optimization algorithm is used to solve the interview schedule corresponding to the GPU server assigned to each interview request end.

[0014] Optionally, based on the interview scheduled time period corresponding to the GPU server allocated to each interview request terminal in the interview request allocation strategy, driving the GPU server to perform an interview answering status monitoring step on the interview request terminal, specifically including: Based on the interview appointment time period corresponding to the GPU server allocated to each interview request terminal in the interview request allocation strategy, establish an interview answering video stream communication link between each interview request terminal and the corresponding GPU server within the interview appointment time period; Drive each GPU server to use the corresponding interview answering status monitoring model based on the received interview answering video stream to execute the interview answering status monitoring project to collect interview answering status features and generate interview evaluation information in real time.

[0015] In addition, in order to achieve the above-mentioned purpose, the present invention also provides an interview answering status monitoring system, comprising: The interview request receiving module is used to receive interview requests for target recruitment information sent by the interview request end, and build an interview task list based on the interview appointment window time and interview position association information in the interview request; A first generating module is configured to generate interview GPU requirement information according to the interview answering status monitoring requirement corresponding to the interview position association information, and link the information to the interview task list as the first additional requirement information; The second generating module is used to generate interview monitoring requirement information according to the content of the multiple interview questions corresponding to the interview position related information, and connect it to the interview task list as the second additional requirement information; A strategy solving module is used to obtain the interview schedule of the target recruitment information, query the idle time periods of several GPU servers within the interview schedule, consider the interview appointment window time and the linked first additional requirement information and second additional requirement information in the interview task list, and solve the interview request allocation strategy; The monitoring execution module is used to drive the GPU server to perform interview answering status monitoring on the interview request terminal based on the interview agreed time period corresponding to the GPU server allocated to each interview request terminal in the interview request allocation strategy.

[0016] The beneficial effects of the present invention are as follows: a method and system for monitoring the state of interview answering questions is proposed, which analyzes and organizes the interview request for the target recruitment information sent by the interview request end, generates an interview task list containing the interview appointment window time and the interview position association information, and then obtains the idle time periods of several GPU servers, considers the interview appointment window time and the linked interview GPU demand information and interview monitoring demand information in the interview task list, solves the interview appointment time period corresponding to the GPU server allocated to each interview request end, and drives each GPU server to perform interview answering question status monitoring according to the received interview answering question video stream. Thus, the present invention completely solves the contradiction between delay caused by over-allocation of computing power and waste of resources caused by under-allocation of computing power by matching GPU resources with computing power adaptation and delay standards for each interview request, reduces the number of GPU servers involved in the allocation through a heuristic algorithm, and at the same time improves the uniformity of GPU computing power, reduces the hardware operating cost of the service agency, and provides recruiting companies with efficient, accurate and low-cost interview monitoring servers, while ensuring the user experience, and improving the benefits of third-party service agencies as much as possible. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 Schematic diagram of the process of monitoring the interview answering status according to an embodiment of the present invention; Figure 2 This is a structural diagram of the interview answering status monitoring system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present 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 only used to explain the present invention and are not intended to limit the present invention.

[0019] The embodiment of the present invention provides a method for monitoring the status of interview questions, referring to Figure 1 , Figure 1 Schematic diagram of the process of monitoring the interview answering status according to an embodiment of the present invention.

[0020] In this embodiment, a method for monitoring the status of interview questions is provided, the method comprising the following steps: S1: Receive an interview request for the target recruitment information sent by the interview request end, and build an interview task list based on the interview appointment window time and interview position association information in the interview request; S2: Generate interview GPU requirement information based on the interview answering status monitoring requirements corresponding to the interview position association information, and link it to the interview task list as the first additional requirement information; S3: Generate interview monitoring requirement information based on the content of the multiple interview questions corresponding to the interview position-related information, and connect it to the interview task list as second additional requirement information; S4: Obtain the interview schedule of the target recruitment information, query the idle time periods of several GPU servers within the interview schedule, consider the interview appointment window time in the interview task list and the linked first additional requirement information and second additional requirement information, and solve the interview request allocation strategy; S5: Based on the interview appointment time period corresponding to the GPU server allocated to each interview request terminal in the interview request allocation strategy, drive the GPU server to perform interview answering status monitoring on the interview request terminal.

[0021] It should be noted that in scenarios with large-scale recruitment interview needs (such as regular recruitment interview plans of large state-owned enterprises and group companies), recruiting companies need to complete a large number of interview tasks in a relatively short period of time. Third-party service agencies need to consider the recruitment position requirements, recruitment period requirements and recruitment quantity requirements of each recruiting company when providing interview answering status monitoring services. As a result, existing technologies have problems such as difficulty in dynamically matching GPU computing power requirements, over-allocation of computing power leading to user experience degradation, under-allocation of computing power leading to waste of resource costs, and high allocation complexity under dynamic requirements. On the one hand, this affects the user experience (for example, the monitoring and analysis results fed back by the interview answering status monitoring model have high latency and cannot provide real-time analysis and evaluation for interviewers). On the other hand, it also affects the efficiency of third-party service agencies (under-allocation of computing power leads to a reduction in the number of recruiting companies served by third-party service agencies under the premise of limited GPU server resources).

[0022] In order to solve the above problems, this embodiment analyzes and organizes the interview requests for the target recruitment information sent by the interview request end, generates an interview task list containing the interview appointment window time and the interview position association information, and then obtains the idle time periods of several GPU servers, considers the interview appointment window time and the linked interview GPU demand information and interview monitoring demand information in the interview task list, solves the interview appointment time period corresponding to the GPU server assigned to each interview request end, and drives each GPU server to perform interview answering status monitoring according to the received interview answering video stream. Therefore, the present invention solves the contradiction between delay caused by over-allocation of computing power and waste of resources caused by under-allocation of computing power by matching each interview request with GPU resources with adapted computing power and delay that meet the standards, ensuring user experience while improving the benefits of third-party service agencies as much as possible.

[0023] In a preferred embodiment, before the step of receiving an interview request for a target job posting sent by an interview request terminal and constructing an interview task list based on the interview appointment window time and interview position association information in the interview request, the following steps are further included: S101: Obtain target recruitment information sent by the recruiting company and extract the recruitment interview requirements in the target recruitment information; wherein the recruitment interview requirements include the recruitment interview positions, the number of positions for each recruitment interview position, and the interview request delivery time period; S102: Based on the recruitment interview positions and the number of recruitments for each recruitment interview position in the recruitment interview requirements, determine the first generation judgment condition of the interview task list; based on the interview request delivery time period in the recruitment interview requirements, determine the second generation judgment condition of the interview task list.

[0024] In this embodiment, the system first obtains target recruitment information sent by the recruiting company and extracts core interview requirements, including the job title, the planned number of positions for each position, and the valid submission period for interview requests. Based on these extracted requirements, the system then sets two criteria: a first generation criterion (linking the position to the number of positions, used to filter requests that do not exceed the quota); and a second generation criterion (linking the request submission period, used to filter requests that do not time out). These criteria set the standard for the precise construction of subsequent task lists. By clearly defining the generation boundaries and screening criteria for interview task lists, invalid requests that exceed the recruitment quota or the submission period are prevented from entering the subsequent process, reducing inefficient resource consumption and improving the effectiveness and accuracy of task lists.

[0025] On this basis, the interview request for the target recruitment information sent by the interview request end is received. According to the interview appointment window time and interview position related information in the interview request, the interview task list steps are constructed, specifically including: S111: receiving interview requests for target recruitment information sent by multiple interview request terminals, and extracting the request submission time, interview appointment window time, and interview position association information of the interview requests; S112: Based on the first generation judgment condition determined by the interview position-related information of the interview request and the recruitment interview position and the number of positions in the recruitment interview requirement, it is determined whether the number of interview requests received for the recruitment interview position that matches the interview position-related information reaches the corresponding number of positions; if so, it is determined that the interview request meets the first generation judgment condition; S113: determining, based on a second generation judgment condition determined by the request submission time of the interview request and the interview request delivery period in the recruitment interview requirement, whether the request submission time of the interview request falls within the interview request delivery period; if so, determining that the interview request meets the second generation judgment condition; S114: Add the interview appointment window time and interview position association information in the received interview request that meets both the first generation judgment condition and the second generation judgment condition to the constructed interview task list.

[0026] In this embodiment, after receiving requests from multiple interview request terminals, the key information of each request is first extracted, and then the requests are double-checked according to the first and second generation judgment conditions: checking whether the number of job requests corresponding to the job-related information does not exceed the number of recruitments (meeting the first condition), and checking whether the request submission time is within the delivery period (meeting the second condition). Finally, for requests that meet both conditions, only their appointment window time and job-related information are extracted and added to the interview task list. In this way, accurate filtering of invalid requests is achieved through dual condition verification, ensuring that the interview task list only contains valid requests within the quantity limit and within the time validity period, avoiding invalid requests from occupying GPU resources and scheduling energy, and improving the efficiency of subsequent resource allocation.

[0027] In a preferred embodiment, based on the interview answering status monitoring requirements corresponding to the interview position association information, the interview GPU requirement information is generated and linked to the interview task list step as the first additional requirement information, specifically including: S21: Determine an interview question answering status monitoring requirement based on the interview position type and the interview position ability requirement set in the interview position association information; wherein the interview question answering status monitoring requirement includes a plurality of interview question answering status monitoring items; S22: Based on video analysis test data of several GPU servers used for interview answer video analysis, determine a candidate set of GPU servers whose GPU computing power quantization requirement parameters and video analysis result feedback delay meet interview experience requirements when executing the same interview answer status monitoring project; The candidate set of GPU servers includes several GPU servers whose video analysis result feedback delay parameter in the video analysis test data is lower than a preset delay threshold required by the interview experience; S23: Aggregate the GPU server candidate set and the GPU computing power quantification requirement parameters to generate interview GPU requirement information, and link the interview GPU requirement information to the interview task list as the first additional requirement information corresponding to the interview request.

[0028] On this basis, based on the interview position type and interview position ability requirement set in the interview position related information, the required steps for monitoring the interview answer status are determined, specifically including: S211: Based on the interview position type and the interview position ability requirement set in the interview position association information, query the interview question answering status monitoring model database for an interview question answering status monitoring model that meets the monitoring requirements corresponding to the interview position type and the interview position ability requirement set; S22: Query several interview question answering status monitoring items covered in the interview question answering status monitoring model, and use them as interview question answering status monitoring requirements corresponding to the interview request.

[0029] In this embodiment, based on the job type and job ability requirement set in the interview job association information, the interview answering status monitoring requirements required for the job are determined, and then based on the historical GPU server video analysis test data, two key parameters for executing these monitoring projects are determined: GPU computing power quantification requirement parameters (computing power required to complete monitoring), and a candidate set of GPU servers that meet the interview experience (for example, GPU numbers with video analysis feedback delay ≤100ms). Finally, the candidate set and computing power parameters are summarized as interview GPU requirement information, and linked to the list entry of the corresponding interview task as the first additional requirement information.

[0030] It should be noted that based on the job type and job ability requirement set in the interview job related information, the preset interview question answering status monitoring model database is queried for a monitoring model that fully matches the job type and ability requirements, and then all the specific interview question answering status monitoring items contained in the matched monitoring model are extracted. These items are used as the final monitoring requirements of the interview request to achieve accurate matching of monitoring requirements and positions.

[0031] For example, the job-related information for "back-end development position" shows that "Java code ability assessment" is required, and the corresponding monitoring requirements are "code operation identification and syntax correctness analysis". Historical test data shows that executing these two monitoring items requires 24TFLOPS computing power, and only the NVIDIA A100 model GPU (numbered G1-G15) meets the feedback delay standard. Based on this, the system generates GPU requirement information (computing power 24TFLOPS, candidate set G1-G15) and links it to the interview task item for the position.

[0032] Therefore, this embodiment accurately defines the GPU resource requirements for each interview task, thereby avoiding the problems of "delay due to insufficient computing power" or "waste due to excess computing power" in subsequent allocation, and provides a basis for accurate scheduling of GPU resources.

[0033] In a preferred embodiment, based on the content of several interview questions corresponding to the interview position-related information, interview monitoring requirement information is generated and connected to the interview task list step as the second additional requirement information, specifically including: S31: extracting job type and job requirement keywords from the job skill requirement set in the job association information, and matching a number of interview question contents with the highest correlation with the job type requirement keywords in the corresponding interview question content set; S32: Estimate the total interview answering time for each interview request based on the interview answering time requirement recorded in the content of each interview question, and convert the estimated total interview answering time into interview monitoring demand information as the second additional demand information corresponding to the interview request and link it to the interview task list.

[0034] In this embodiment, the core keywords of the position type and the position ability requirement set (such as "communication ability" for functional positions) are first extracted from the interview position related information, and then several interview questions with the highest correlation with these keywords are matched in the preset interview question content set (such as "situational communication questions" and "team collaboration questions" for functional positions). Finally, based on the answering time requirements recorded for each matching question, the estimated total interview answering time for the interview request is accumulated and estimated, and the time is converted into interview monitoring demand information, which is linked to the corresponding interview task list entry as the second additional demand information.

[0035] For example, the keywords for "human resources position" (functional position) are "communication and coordination, labor law knowledge", which are matched with "employee dispute resolution questions" (20 minutes in duration) and "labor contract interpretation questions" (15 minutes in duration). The system estimates that the total interview time is 35 minutes, and uses "monitoring time of 35 minutes" as the second additional requirement information, which is linked to the interview task entry for this position.

[0036] Therefore, this embodiment clarifies the monitoring time required for each interview task, providing a basis for subsequent querying of GPU idle periods and matching idle windows with adapted duration, thereby avoiding task interruption problems caused by the GPU idle period being shorter than the interview duration.

[0037] In a preferred embodiment, the interview schedule of the target recruitment information is obtained, the idle time periods of several GPU servers within the interview schedule are queried, the interview appointment window time and the linked first additional requirement information and second additional requirement information in the interview task list are considered, and the interview request allocation strategy is solved, specifically including: S41: Obtain the interview plan period of the target recruitment information, query the idle periods of several GPU servers within the interview plan period, and use the idle period and GPU computing power quantization parameters of each GPU server as the interview answering status monitoring resource information of the GPU server; S42: Based on the interview answering status monitoring resource information of each GPU server, considering the interview appointment window time and the first additional requirement information and the second additional requirement information of the link in the interview task list, with the time window limit and the GPU computing power limit as constraints, and with the highest uniformity of GPU computing power under the premise of minimizing GPU server resource usage as the optimization goal, an optimization algorithm is used to solve the GPU server assigned to each interview request end corresponding to the interview appointment time period; S43: Generate an interview request allocation strategy for the target recruitment information according to the interview appointment time period corresponding to the GPU server allocated to each interview request terminal.

[0038] On this basis, based on the interview answering status monitoring resource information of each GPU server, considering the interview appointment window time and the first additional requirement information and the second additional requirement information of the link in the interview task list, with the time window limit and GPU computing power limit as constraints, and the highest uniformity of GPU computing power under the premise of minimizing GPU server resource usage as the optimization goal, an optimization algorithm is used to solve the steps for the GPU server assigned to each interview request end corresponding to the interview appointment time period, specifically including: S421: Based on the GPU computing power quantization parameters of each GPU server and the idle time period in the interview answering status monitoring resource information, consider the interview appointment window time, GPU server candidate set, GPU computing power quantization requirement parameters and the estimated total interview answering time of each interview request; S422: After the interview request of each interview request end is assigned to the corresponding interview appointment time period of the corresponding GPU server, the first constraint condition is that the interview request of each interview request end is assigned to a GPU server within the candidate set of GPU servers corresponding to the interview request; the second constraint condition is that the interview appointment time period assigned to each interview request falls within the intersection of the interview appointment window time corresponding to the interview request and the idle time period of the corresponding GPU server; the third constraint condition is that the sum of the GPU computing power quantization requirement parameters of all interview requests assigned to each GPU server at each moment is not higher than the GPU computing power quantization parameter of the GPU server; the optimization goal is to minimize the variance of the GPU computing power requirement ratio of each GPU server assigned to the interview request under the premise of minimizing the number of GPU servers assigned to the interview request; S423: A heuristic optimization algorithm is used to determine the GPU server assigned to each interview requester and the corresponding interview scheduled time period.

[0039] In this embodiment, the overall interview plan period of the target recruitment information is first obtained, the idle periods of all GPU servers within the period are queried, and the idle periods and GPU computing power quantization parameters are integrated into the interview monitoring resource information of the GPU server. Based on these resource information, combined with the appointment window time, the first additional requirement information (GPU requirement), and the second additional requirement information (duration requirement) in the interview task list, two conditions are used as constraints: time window limit (the allocation period must be within the intersection of the appointment window and the GPU idle period), GPU computing power limit (the total computing power of the allocated tasks ≤ GPU upper limit), with the minimum GPU server resource occupancy (fewer servers) and the highest computing power uniformity (avoiding single GPU overload) as the optimization goal, an optimization algorithm is used to solve the GPU server and the corresponding interview appointment period that should be assigned to each interview request, and finally a complete interview request allocation strategy is generated based on the solution results.

[0040] Specifically, considering the interview appointment window time and the linked interview GPU demand information and interview monitoring demand information in the interview task list, with the time window limit and GPU computing power limit as constraints, and the highest uniformity of GPU computing power under the premise of minimum GPU server resource occupancy, the optimization goal is to solve the GPU server assigned to each interview request end corresponding to the interview agreed time period, drive each GPU server according to the received interview answer video stream, and use the corresponding interview answer status monitoring model to execute the interview answer status monitoring project to collect interview answer status features and generate interview evaluation information in real time.

[0041] Therefore, this embodiment achieves optimized scheduling of GPU resources while meeting time and computing power constraints: it reduces the number of required GPU servers (reducing costs) while ensuring that the computing power of each GPU is evenly used (avoiding overload or idleness), thereby improving resource utilization efficiency and scheduling rationality.

[0042] In a preferred embodiment, based on the interview appointment time period corresponding to the GPU server assigned to each interview request terminal in the interview request allocation strategy, the GPU server is driven to perform the interview answering status monitoring step on the interview request terminal, specifically including: S51: Based on the GPU server corresponding to the interview appointment time period allocated to each interview request terminal in the interview request allocation strategy, establish an interview answering video stream communication link between each interview request terminal and the corresponding GPU server within the interview appointment time period; S52: Drive each GPU server to perform interview answering status feature collection and interview evaluation information real-time generation of the interview answering status monitoring project based on the received interview answering video stream and using the corresponding interview answering status monitoring model.

[0043] In this embodiment, interview monitoring and evaluation are automated, real-time, and job-based, and status collection and evaluation generation can be completed without human intervention, which not only improves the efficiency of interview monitoring, but also provides interviewers with objective data-based evaluation basis, while ensuring stable video stream transmission and avoiding monitoring interruptions.

[0044] Therefore, the present invention completely solves the contradiction between delay caused by over-allocation of computing power and waste of resources caused by under-allocation of computing power by matching each interview request with GPU resources with adapted computing power and delay that meet the standards. It reduces the number of GPU servers involved in allocation through heuristic algorithms, and at the same time improves the uniformity of GPU computing power, reduces the hardware operating costs of service agencies, and provides recruiting companies with efficient, accurate, and low-cost interview monitoring servers. On the basis of ensuring user experience, it maximizes the benefits of third-party service agencies.

[0045] Reference Figure 2 , Figure 2 This is a structural diagram of the interview answering status monitoring system according to an embodiment of the present invention.

[0046] like Figure 2 As shown, the interview answering status monitoring system proposed in the embodiment of the present invention includes: An interview request receiving module 10 is configured to receive an interview request for a target job posting sent by an interview request terminal, and to construct an interview task list based on the interview appointment window time and interview position association information in the interview request; The first generating module 20 is used to generate interview GPU requirement information according to the interview answering status monitoring requirements corresponding to the interview position association information, and link it to the interview task list as the first additional requirement information; The second generating module 30 is used to generate interview monitoring requirement information according to the content of the multiple interview questions corresponding to the interview position related information, and connect it to the interview task list as the second additional requirement information; A strategy solving module 40 is configured to obtain a planned interview period for a target job posting, query idle periods of a plurality of GPU servers within the planned interview period, and solve an interview request allocation strategy by considering the interview appointment window time and the linked first additional requirement information and the second additional requirement information in the interview task list; The monitoring execution module 50 is used to drive the GPU server to perform interview answering status monitoring on the interview request terminal based on the interview agreed time period corresponding to the GPU server allocated to each interview request terminal in the interview request allocation strategy.

[0047] Other embodiments or specific implementation methods of the interview answering status monitoring system of the present invention can refer to the above-mentioned method embodiments and will not be repeated here.

[0048] It should be understood that, in the description of this specification, reference to terms such as "one embodiment," "another embodiment," "other embodiments," or "first to Nth embodiments" means that the specific features, structures, materials, or characteristics described in conjunction with that embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any appropriate manner in any one or more embodiments or examples.

[0049] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or system comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or system. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or system comprising the element.

[0050] The above are only preferred embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention description and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for monitoring the status of interview questions, characterized in that: The method comprises the following steps: Receive interview requests for target recruitment information sent by the interview request end, and build an interview task list based on the interview appointment window time and interview position association information in the interview request; Generate interview GPU requirement information based on the interview answering status monitoring requirements corresponding to the interview position association information, and link it to the interview task list as the first additional requirement information; Generate interview monitoring requirement information based on the content of several interview questions corresponding to the interview position-related information, and connect it to the interview task list as second additional requirement information; Obtain the interview schedule for the target job posting, query the idle time slots of several GPU servers within the interview schedule, consider the interview appointment window time and the linked first and second additional requirements in the interview task list, and solve the interview request allocation strategy; Based on the interview appointment time period corresponding to the GPU server allocated to each interview request end in the interview request allocation strategy, the GPU server is driven to perform interview answering status monitoring on the interview request end.

2. The method for monitoring the interview answering status according to claim 1, wherein: Receiving an interview request for the target recruitment information sent by the interview request end, and building the interview task list according to the interview appointment window time and interview position association information in the interview request, the following steps are also included before the step: Obtain target recruitment information sent by the recruiting company and extract the recruitment interview requirements in the target recruitment information; wherein the recruitment interview requirements include the recruitment interview positions, the number of vacancies for each recruitment interview position, and the interview request delivery time period; Based on the recruitment interview positions and the number of recruitments for each recruitment interview position in the recruitment interview requirements, the first generation judgment condition of the interview task list is determined, and based on the interview request delivery time period in the recruitment interview requirements, the second generation judgment condition of the interview task list is determined.

3. The method for monitoring the interview answering status according to claim 2, wherein: Receive an interview request for the target job posting from the interview requester, and build an interview task list based on the interview appointment window time and interview position association information in the interview request. The steps include: Receive interview requests for target recruitment information sent by multiple interview request terminals, and extract the request submission time, interview appointment window time and interview position association information of the interview requests; Based on the first generation judgment condition determined by the interview position-related information of the interview request and the recruitment interview position and the number of vacancies in the recruitment interview requirements, it is judged whether the number of interview requests received for the recruitment interview position that matches the interview position-related information reaches the corresponding number of vacancies; if so, it is determined that the interview request meets the first generation judgment condition; determining, based on a second generation judgment condition determined by the request submission time of the interview request and the interview request delivery period in the recruitment interview requirements, whether the request submission time of the interview request falls within the interview request delivery period; and if so, determining that the interview request meets the second generation judgment condition; The interview appointment window time and interview position association information in the received interview request that meets both the first generation judgment condition and the second generation judgment condition are added to the constructed interview task list.

4. The method for monitoring the interview answering status according to claim 1, wherein: Generate interview GPU requirement information based on the interview answer status monitoring requirements corresponding to the interview position association information, and link it to the interview task list step as the first additional requirement information, specifically including: Determine the interview question answering status monitoring requirements based on the interview position type and the interview position ability requirement set in the interview position association information; wherein the interview question answering status monitoring requirements include a plurality of interview question answering status monitoring items; Based on the video analysis test data of several GPU servers used for interview answer video analysis, determine the candidate set of GPU servers whose GPU computing power quantization requirements and video analysis result feedback delay meet the interview experience requirements when executing the same interview answer status monitoring project; The candidate set of GPU servers includes several GPU servers whose video analysis result feedback delay parameter in the video analysis test data is lower than a preset delay threshold required by the interview experience; The GPU server candidate set and the GPU computing power quantification requirement parameters are aggregated to generate interview GPU requirement information, and the interview GPU requirement information is linked to the interview task list as the first additional requirement information corresponding to the interview request.

5. The method for monitoring the interview answering status according to claim 4, characterized in that: Based on the interview position type and the interview position ability requirement set in the interview position association information, determine the required steps for monitoring the interview answer status, specifically including: Based on the interview position type and the interview position ability requirement set in the interview position association information, query the interview question answering status monitoring model database for an interview question answering status monitoring model that meets the monitoring requirements corresponding to the interview position type and the interview position ability requirement set; Query several interview question answering status monitoring items covered in the interview question answering status monitoring model, and use them as the interview question answering status monitoring requirements corresponding to the interview request.

6. The method for monitoring the interview answering status according to claim 4, characterized in that: Generate interview monitoring requirement information based on the content of several interview questions corresponding to the interview position-related information, and connect it to the interview task list step as the second additional requirement information, specifically including: Extracting job type and job requirement keywords from the interview job association information and the interview job competency requirement set, and matching several interview question contents with the highest correlation with the job type requirement keywords in the corresponding interview question content set; According to the interview answering time requirement recorded in each interview question content, the estimated total interview answering time for each interview request is estimated, and the estimated total interview answering time is converted into interview monitoring demand information as the second additional demand information corresponding to the interview request and linked to the interview task list.

7. The method for monitoring the interview answering status according to claim 6, characterized in that: Obtain the interview schedule for the target job posting, query the idle time slots of several GPU servers within the interview schedule, consider the interview appointment window time and the linked first and second additional requirements in the interview task list, and solve the interview request allocation strategy, specifically including: Obtain the interview schedule of the target recruitment information, query the idle time periods of several GPU servers within the interview schedule, and use the idle time periods and GPU computing power quantization parameters of each GPU server as the interview answering status monitoring resource information of the GPU server; Based on the interview answering status monitoring resource information of each GPU server, considering the interview appointment window time and the first additional demand information and the second additional demand information of the link in the interview task list, with the time window limit and GPU computing power limit as the constraints, and the highest uniformity of GPU computing power under the premise of minimizing GPU server resource usage as the optimization goal, an optimization algorithm is used to solve the GPU server assigned to each interview request end corresponding to the interview appointment time period; Generate an interview request allocation strategy for the target recruitment information based on the interview appointment time period corresponding to the GPU server assigned to each interview request end.

8. The method for monitoring the interview answering status according to claim 7, wherein: Based on the interview answering status monitoring resource information of each GPU server, considering the interview appointment window time and the first additional requirement information and the second additional requirement information of the link in the interview task list, with the time window limit and GPU computing power limit as the constraints, and the highest uniformity of GPU computing power under the premise of minimizing GPU server resource usage as the optimization goal, an optimization algorithm is used to solve the steps for each interview request end to be assigned to the GPU server corresponding to the interview appointment time period, specifically including: Based on the GPU computing power quantization parameters of each GPU server and the idle time period in the interview answering status monitoring resource information, the interview appointment window time, GPU server candidate set, GPU computing power quantization requirement parameters and the estimated total interview answering time of each interview request are considered; After the interview request of each interview request end is assigned to the corresponding GPU server corresponding to the interview appointment time period, the first constraint condition is that the interview request of each interview request end is assigned to the GPU server within the candidate set of GPU servers corresponding to the interview request; the second constraint condition is that the interview appointment time period assigned to each interview request falls within the intersection of the interview appointment window time corresponding to the interview request and the idle time period of the corresponding GPU server; the third constraint condition is that the sum of the GPU computing power quantization requirement parameters of all interview requests assigned to each GPU server at each moment is not higher than the GPU computing power quantization parameters of the GPU server; the optimization goal is to minimize the variance of the GPU computing power requirement ratio of each GPU server assigned to the interview request under the premise of minimizing the number of GPU servers assigned to the interview request; A heuristic optimization algorithm is used to solve the interview schedule corresponding to the GPU server assigned to each interview request end.

9. The method for monitoring the interview answering status according to claim 1, wherein: Based on the interview scheduled time period corresponding to the GPU server assigned to each interview request terminal in the interview request allocation strategy, the GPU server is driven to perform the interview answer status monitoring step for the interview request terminal, specifically including: Based on the interview appointment time period corresponding to the GPU server allocated to each interview request terminal in the interview request allocation strategy, establish an interview answering video stream communication link between each interview request terminal and the corresponding GPU server within the interview appointment time period; Drive each GPU server to use the corresponding interview answering status monitoring model based on the received interview answering video stream to execute the interview answering status monitoring project to collect interview answering status features and generate interview evaluation information in real time.

10. An interview answering status monitoring system, characterized in that: include: The interview request receiving module is used to receive interview requests for target recruitment information sent by the interview request end, and build an interview task list based on the interview appointment window time and interview position association information in the interview request; A first generating module is configured to generate interview GPU requirement information according to the interview answering status monitoring requirement corresponding to the interview position association information, and link the information to the interview task list as the first additional requirement information; The second generating module is used to generate interview monitoring requirement information according to the content of the multiple interview questions corresponding to the interview position related information, and connect it to the interview task list as the second additional requirement information; A strategy solving module is used to obtain the interview schedule of the target recruitment information, query the idle time periods of several GPU servers within the interview schedule, consider the interview appointment window time and the linked first additional requirement information and second additional requirement information in the interview task list, and solve the interview request allocation strategy; The monitoring execution module is used to drive the GPU server to perform interview answering status monitoring on the interview request terminal based on the interview agreed time period corresponding to the GPU server allocated to each interview request terminal in the interview request allocation strategy.

Images