Interview processing method and device and storage medium

Abstract

The embodiment of the invention provides an interview processing method and device and a storage medium. In the embodiment of the invention, a target interview video generated by interview between an applicant and a digital person can be collected. Evaluating question and answer performance of the applicant in the target interview video to generate interview evaluation information of the applicant; calculating dimension scores of the applicant under different interview dimensions, and calculating a comprehensive score of the applicant based on the dimension scores; and generating an initial interview report for the applicant based on the interview evaluation information of the applicant and the comprehensive score of the applicant. The recruiter can also correct the dimension range in the initial interview report to obtain a target interview report. Therefore, quantitative analysis is performed on the target interview video to generate the interview report, so that the influence of human subjective factors on the interview evaluation result can be reduced, the recruiter can be helped to more comprehensively know the condition of the applicant, the recruiter has a basis during decision making, and the recruitment accuracy and effectiveness are improved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to an interview processing method, device and storage medium. Background Art

[0002] AI interview refers to an interview conducted using artificial intelligence technology, usually conducted by a virtual AI interviewer.

[0003] In existing AI interview solutions, recruiters usually need to spend a lot of time and energy to evaluate the performance of candidates. This evaluation method is not only inefficient, but also causes the evaluation results to be affected by subjective factors. It is difficult to conduct standardized evaluations of candidates, resulting in recruiters lacking effective data support when making decisions, which in turn leads to poor recruitment accuracy. Summary of the invention

[0004] Multiple aspects of the present application provide an interview processing method, device and storage medium to improve the accuracy of recruitment.

[0005] The present application provides an interview processing method, the method comprising:

[0006] In the process of interviewing the applicants participating in the target recruitment information based on the digital human, collecting the target interview video generated by the recruitment questions and answers between the applicants and the digital human;

[0007] According to preset dimension evaluation rules, the question-answering performance of the candidate in the target interview video is evaluated to generate interview evaluation information of the candidate;

[0008] Based on the target interview dimension adapted to the target recruitment information, extracting a dimension range corresponding to the target interview dimension from a preset dimension library;

[0009] Calculate the dimension score of the target interview video under each target interview dimension according to the dimension range corresponding to the target interview dimension and the scoring rules pre-configured for the dimension range;

[0010] Calculate the comprehensive score of the candidate based on the dimension score of the target interview video under each target interview dimension;

[0011] Based on the interview evaluation information and the comprehensive score, generate an initial interview report for the applicant, and display the initial interview report, wherein the initial interview report includes the interview evaluation information, the comprehensive score, the dimension range and the dimension data;

[0012] In response to a modification operation initiated by the recruiter on the dimension range in the initial interview report, the initial interview report is modified to obtain a target interview report.

[0013] The embodiment of the present application also provides a computing device, including: a memory, a processor, and a communication component;

[0014] The memory is used to store one or more computer instructions;

[0015] The processor is coupled to the memory and the communication component, and is configured to execute the one or more computer instructions to perform the aforementioned interview processing method.

[0016] An embodiment of the present application also provides a computer-readable storage medium storing computer instructions. When the computer instructions are executed by one or more processors, the one or more processors are caused to execute the aforementioned interview processing method.

[0017] An embodiment of the present application also provides a computer program product, including a computer program; when the computer program is executed by a processor, the processor is caused to execute the aforementioned interview processing method.

[0018] In the embodiment of the present application, the digital human can be used to interview the applicants who participate in the target recruitment information, and the target interview video generated by the applicant during the interview process can be collected. According to the preset dimension evaluation rules, the question and answer performance of the applicant in the target interview video is evaluated to generate interview evaluation information for the applicant; based on the target interview dimension adapted to the target recruitment information, according to the scoring rules pre-configured for the target interview dimension, the dimension score of the target interview video under each target interview dimension is calculated, and based on these dimension scores, the comprehensive score of the applicant is calculated; based on the interview evaluation information of the applicant and the comprehensive score of the applicant, an initial interview report can be generated for the applicant. Based on this, in response to the correction operation initiated by the recruiter for the dimension range in the initial interview report, the initial interview report can be corrected to obtain the target interview report. Accordingly, through the automated digital human interview process, the workload of the recruiter can be reduced and the interview efficiency can be improved; the quantitative analysis of the target interview video to generate an interview report can not only reduce the impact of human subjective factors on the interview evaluation results, but also help the recruiter to understand the situation of the applicant more comprehensively, so that the recruiter has a basis to rely on when making decisions, and improve the accuracy and effectiveness of recruitment. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0020] Figure 1 A structural diagram of an interview processing system provided as an exemplary embodiment of the present application;

[0021] Figure 2 A flowchart of an interview processing method provided for an exemplary embodiment of the present application;

[0022] Figure 3 A schematic diagram of an exemplary initial interview report provided for an exemplary embodiment of the present application;

[0023] Figure 4 A schematic diagram of a report viewing interface provided by an exemplary embodiment of the present application;

[0024] Figure 5 A schematic diagram of determining a dimensional range provided for an exemplary embodiment of the present application;

[0025] Figure 6 A schematic diagram of another method for determining a dimensional range provided for an exemplary embodiment of the present application;

[0026] Figure 7 A flowchart of an exemplary application scenario provided for another exemplary embodiment of the present application;

[0027] Figure 8 A schematic diagram of the structure of a computing device is provided for yet another exemplary embodiment of the present application. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solution and advantages of the present application clearer, the technical solution of the present application will be clearly and completely described below in combination with the specific embodiments of the present application and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application.

[0029] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0030] In view of the technical problem that in the existing AI interview scheme, the recruiter evaluates the performance of the applicant, and the evaluation result is affected by subjective factors, resulting in a lack of effective data support for the recruiter when making decisions and poor recruitment accuracy, the embodiment of the present application provides a solution. The basic idea is: the applicants participating in the target recruitment information can be interviewed based on digital humans, and the target interview videos generated by the applicants during the interview process can be collected, and the target interview videos can be quantitatively analyzed to generate an interview report. This can not only reduce the impact of human subjective factors on the interview evaluation results, but also help the recruiter to understand the applicant's situation more comprehensively, so that the recruiter has a basis to rely on when making decisions, thereby improving the accuracy and effectiveness of recruitment.

[0031] The technical solutions provided by various embodiments of the present application are described in detail below in conjunction with the accompanying drawings.

[0032] Figure 1 The following is a schematic diagram of the structure of an interview processing system provided by an exemplary embodiment of the present application. Figure 1 As shown, the interview processing system includes a digital human server 101, a recruiter client 102 and an applicant client 103. The digital human server 101 and the recruiter client 102 can be connected to each other by communication, and the recruiter client 102 and the applicant client 103 can be connected to each other by communication. The communication connection can be a wireless communication connection or a wired communication connection. For the wireless communication connection mode, the communication connection can be realized through a mobile network. Accordingly, the network standard of the mobile network can be 2G (GSM), 2.5G (GPRS), 3G (WCDMA, TD-SCDMA, CDMA2000, UTMS), 4G (LTE), 4G+ (LTE+), 5G, WiMax or any of the new network standards that will appear in the future. In addition, when the two parties of the communication connection are located in the same local area network, the wireless communication connection can also be realized through Bluetooth, WiFi, infrared, zigbee or NFC.

[0033] In the embodiment of the present application, the implementation form of the digital human server 101 is not limited. The digital human server 101 can be a physical machine server, or a cloud server or server cluster, etc. The recruiter client 102 can be a terminal device used by any recruiter, and the applicant client 103 can be a terminal device used by any applicant. The embodiment of the present application does not limit the implementation form of the recruiter client 102 and the applicant client 103. For example, the recruiter client 102 or the applicant client 103 can be a smart handheld device, such as a smart phone, a tablet computer, a laptop or a desktop computer, etc.; for another example, the recruiter client 102 or the applicant client 103 can also be a smart wearable device, such as a smart watch, a smart bracelet, etc.; for another example, the recruiter client 102 or the applicant client 103 can also be various smart home appliances with display screens, such as smart TVs, smart large screens or smart robots, etc. In addition, various applications can be installed on the recruiter client 102 or the applicant client 103. The application can be an independent APP or a small program that depends on the independent APP to run. This embodiment does not limit this. The embodiment of the present application does not limit the type of application. For example, the application can be a shopping application, a taxi application, a video playback application, a recruitment application, and a digital human online interview application, etc.

[0034] It should be noted that this embodiment focuses on an application that can provide an interview room, through which applicants and digital humans can conduct online interviews based on digital humans for the positions being recruited. In addition, an application can provide one or more interview rooms, and one interview room can interview multiple different positions, or one interview room can only interview one position, which is not limited in this embodiment of the application.

[0035] Figure 2 A flowchart of an interview processing method provided by an exemplary embodiment of the present application. The method can be executed by the digital human server in the above-mentioned interview processing system, and the digital human server can control the digital human in the interview room to interview the applicant. Figure 2 , the method comprising:

[0036] Step 201, in the process of interviewing the applicants participating in the target recruitment information based on the digital human, collecting the target interview video generated by the recruitment questions and answers between the applicants and the digital human;

[0037] Step 202, evaluating the candidate's question-answering performance in the target interview video according to the preset dimension evaluation rules to generate interview evaluation information for the candidate;

[0038] Step 203, based on the target interview dimension adapted to the target recruitment information, extracting the dimension range corresponding to the target interview dimension from a preset dimension library;

[0039] Step 204, calculating the dimension score of the target interview video under each target interview dimension according to the dimension range corresponding to the target interview dimension and the scoring rule pre-configured for the dimension range;

[0040] Step 205, calculating the applicant's corresponding comprehensive score based on the dimension scores of the target interview video under each target interview dimension;

[0041] Step 206, based on the interview evaluation information and the comprehensive score, generate an initial interview report for the applicant, and display the initial interview report, wherein the initial interview report includes the interview evaluation information, the comprehensive score, the dimension range and the dimension data;

[0042] Step 207 , in response to the recruiter initiating a modification operation on the dimension range in the initial interview report, modify the initial interview report to obtain a target interview report.

[0043] The interview processing method provided in the embodiment of the present application can be applied to online recruitment scenarios, for example, an employer recruits employees through online recruitment, etc. This embodiment does not limit this.

[0044] In step 201, during the process of interviewing the applicants participating in the target recruitment information based on the digital human, the target interview video generated by the recruitment questions and answers between the applicants and the digital human can be collected. The digital human is a digital human image close to the human image created by digital technology. In this embodiment, the digital human is a virtual interviewer in the interview room. The digital human server can control the digital human in the interview room to interview the applicant. The target recruitment information is the recruitment information posted by the recruiter on the application, usually displayed in the form of recruitment posts and recruitment cards. The target recruitment information includes but is not limited to recruitment positions, job responsibilities, qualifications, etc.

[0045] After the recruiter publishes the target recruitment information on the recruiter client, the applicant can browse the target recruitment information on the applicant client and participate in the online interview of the target recruitment information. This online interview is to use the digital human as an interviewer to conduct recruitment questions and answers with the applicant in the interview room. In the process of interviewing the applicant participating in the target recruitment information based on the digital human, the digital human server can collect the interview data generated by the recruitment questions and answers between the applicant and the digital human in real time, and generate a target interview video for the applicant after the applicant completes the interview.

[0046] Based on this, the target interview video can be processed to generate a target interview report for the applicant. The target interview report can be used to characterize the applicant's personal abilities, reflect the degree of compatibility between the applicant and the recruitment position in the target recruitment information, and help the recruiter understand the applicant's situation more comprehensively.

[0047] In step 202, the candidate's question-and-answer performance in the target interview video may be evaluated according to the preset dimension evaluation rules to generate interview evaluation information for the candidate. In this embodiment, corresponding dimension evaluation rules are pre-configured for each interview dimension, and the interview dimensions may include but are not limited to age, gender, education, work experience, etc. The dimension evaluation rules corresponding to the interview dimensions can be used to measure the candidate's work ability under the interview dimensions. These rules are usually based on certain logic, data or experience, and are intended to provide objective and consistent evaluation results.

[0048] In an interview, the digital person representing the recruiter usually asks the candidate multiple interview questions. Each interview question corresponds to one or more interview dimensions, and each interview dimension is configured with a corresponding dimension evaluation rule. Therefore, for each interview question, the interview answer given by the candidate to the interview question can be evaluated based on the interview dimension corresponding to the interview question and using the dimension evaluation rule corresponding to the interview dimension. For example, for the interview question "How many years of work experience do you have?", the corresponding dimension evaluation rule "If the work experience included in the interview answer is more than 3 years, the candidate is evaluated to have rich work experience, otherwise the candidate is evaluated to have less work experience" is used to evaluate the interview answer given to the interview question and obtain the evaluation result under the interview dimension. On this basis, the corresponding evaluation results are obtained for each of the multiple interview answers in an interview, and the evaluation results for multiple interview questions can be summarized into a sentence as the interview evaluation information of the candidate. For example, the interview evaluation information can be "The basic conditions of the candidate meet the requirements, he has rich work experience, is warm and generous, and is recommended for an interview."

[0049] In step 203, based on the target interview dimension adapted to the target recruitment information, the dimension range corresponding to the target interview dimension can be extracted from the preset dimension library. The preset dimension library contains multiple interview dimensions and the dimension ranges corresponding to the interview dimensions. As mentioned above, the target recruitment information includes the recruitment positions required by the recruiter. The recruitment positions are job roles set up to complete specific tasks. The recruitment positions can be waiters, drivers, nannies, etc. Different recruitment positions have different requirements for the incumbents, and different recruitment positions also have different focuses on applicants. Therefore, the interview dimensions corresponding to different recruitment positions are not exactly the same.

[0050] In this embodiment, the corresponding interview dimensions can be pre-configured for different recruitment positions according to the recruiter's focus on the recruitment position, and the corresponding dimension range can be configured for each interview dimension. The interview dimension can be used to characterize the evaluation factors of the applicant during the interview process. The interview dimension may include but is not limited to age, gender, education, work experience, fastest time to join the job, the distance between the workplace and the place of residence, etc.; the dimension range is the recruitment requirements set by the recruiter for the interview dimension for the recruitment position. The dimension range can be: the applicant's age is limited to between 18 and 40 years old, the applicant's education is a bachelor's degree or above, etc. By configuring different interview dimensions and dimension ranges for different recruitment positions, the specific recruitment needs of different recruitment positions are clarified, which helps recruiters to screen more suitable applicants.

[0051] After obtaining the target interview video generated for the applicant, the recruitment position that the applicant is competing for can be determined from the target recruitment information that the applicant participates in. Based on the interview dimensions pre-configured for different recruitment positions, the target interview dimensions corresponding to the recruitment position and the scoring rules pre-configured for these target dimension ranges can be determined.

[0052] On this basis, in step 204, the dimension score of the target interview video under each target interview dimension can be calculated according to the dimension range corresponding to the target interview dimension and the scoring rule pre-configured for the dimension range. The dimension score can be used to describe the degree of fit between the dimension data of the applicant under the interview dimension and the recruitment requirements of the recruiter under the interview dimension. Each interview dimension has a corresponding scoring rule, which refers to the calculation method to be followed in calculating the dimension score based on the dimension range under the interview dimension.

[0053] For example, if the interview dimension is "age" and the dimension range is between 18-45 years old, the scoring rule can be: if the age contained in the dimension data is between 18-25 years old, the dimension score is 90; if the age contained in the dimension data is between 26-30 years old, the dimension score is 100; if the age contained in the dimension data is between 31-35 years old, the dimension score is 80; if the age contained in the dimension data is between 35-40 years old, the dimension score is 70; if the age contained in the dimension data is between 41-45 years old, the dimension score is 60; if the age contained in the dimension data is greater than 45 years old or less than 18 years old, the dimension score is 50.

[0054] In step 205, the comprehensive score corresponding to the applicant can be calculated based on the dimension score of the target interview video under each target interview dimension. For multiple target interview dimensions that are adapted to the target recruitment information, a dimension score is calculated for the applicant under each target interview dimension. After that, the average of the multiple dimension scores can be calculated as the dimension score of the target interview video under each target interview dimension. Of course, it is also possible to perform a weighted calculation on the multiple dimension scores, and use the result of the weighted calculation as the dimension score of the target interview video under each target interview dimension. This embodiment does not limit the calculation method of the comprehensive score.

[0055] In step 206, an initial interview report may be generated for the applicant based on the interview evaluation information and the comprehensive score, and the initial interview report may be presented to the recruiter. After the digital human server generates the initial interview report for the applicant, it may determine the recruiter who publishes the target recruitment information based on the target recruitment information in which the applicant participates, and issue a report viewing notification to the recruiter to remind the recruiter to view the applicant's initial interview report. The recruiter may view the applicant's initial interview report through the recruiter client. The recruiter client may play the report viewing process to the recruiter when the recruiter views the initial interview report for the first time, so as to guide the recruiter to view the initial interview report; if it is not the first time to view, the recruiter is directly guided to view the initial interview report.

[0056] Figure 3 A schematic diagram of an exemplary initial interview report provided for an exemplary embodiment of the present application. Figure 3 The initial interview report may include but is not limited to interview evaluation information, comprehensive scores, dimension ranges and dimension data. Among them, the interview evaluation information is a summary of the candidate's interview performance. The comprehensive score represents the candidate's competence for the recruitment position in the target recruitment information, which is convenient for recruiters to quickly understand the overall situation of the candidate; the dimension range represents the recruitment requirements set by the recruiter for the interview dimension, and the dimension data represents the actual situation of the candidate under the interview dimension. For example, the dimension range is "age between 20-40 years old", and the dimension data is 30 years old. Through the dimension range and dimension data, it is convenient for recruiters to accurately understand the specific situation of the candidate under different dimensions.

[0057] When reviewing the initial interview report, the recruiter can revise the initial interview report according to the actual recruitment needs, and timely change the outdated and incorrect dimension ranges in the initial interview report.

[0058] Continue to refer Figure 2In step 207, in response to the correction operation initiated by the recruiter for the dimension range in the initial interview report, the initial interview report can be corrected to obtain the target interview report. As mentioned above, the dimension range is the recruitment requirement set by the recruiter for the interview dimension for the recruitment position. However, the recruitment requirements of the recruiter are not static. If the applicant is always evaluated according to the outdated recruitment requirements, the interview report will provide an erroneous reference for the recruiter's decision-making, thereby affecting the accuracy of the recruitment. To this end, the recruiter can correct the dimension range in the initial interview report when browsing the initial interview report to obtain an accurate interview report; if the initial interview report does not need to be corrected, the initial interview report can be used as the target interview report. Accordingly, after the recruiter modifies the dimension range, the corresponding dimension range in the dimension library can be updated to ensure the accuracy of the dimension score calculated based on the dimension range, thereby ensuring the accuracy of the target interview report.

[0059] Accordingly, in this embodiment, the digital human can be used to interview the applicants who participate in the target recruitment information, and the target interview video generated by the applicant during the interview process can be collected. According to the preset dimension evaluation rules, the question and answer performance of the applicant in the target interview video is evaluated to generate interview evaluation information for the applicant; based on the target interview dimension adapted to the target recruitment information, according to the scoring rules pre-configured for the target interview dimension, the dimension score of the target interview video under each target interview dimension is calculated, and based on these dimension scores, the comprehensive score of the applicant is calculated; based on the interview evaluation information of the applicant and the comprehensive score of the applicant, an initial interview report can be generated for the applicant. Based on this, in response to the correction operation initiated by the recruiter for the dimension range in the initial interview report, the initial interview report can be corrected to obtain the target interview report. Through the automated digital human interview process, the workload of the recruiter can be reduced and the interview efficiency can be improved; the quantitative analysis of the target interview video to generate an interview report can not only reduce the impact of human subjective factors on the interview evaluation results, but also help the recruiter to understand the situation of the applicant more comprehensively, so that the recruiter has a basis to rely on when making decisions, and improve the accuracy and effectiveness of recruitment.

[0060] Optionally, this embodiment may also mark the target interview report. Figure 4 A schematic diagram of a report viewing interface provided by an exemplary embodiment of the present application is shown as follows: Figure 4 As shown, the report viewing interface includes a target interview report and a mark button.

[0061] After the target interview report is generated, a report viewing interface can be displayed to the recruiter, and the recruiter can view the target interview report on the report viewing interface. While displaying the target interview report, in response to a marking operation initiated by the recruiter on the target interview report, multiple marking tags corresponding to the marking operation can be displayed; in response to the recruiter's selection operation on the target marking tag, a target marking tag is added to the target interview report to mark the target interview report, and the target marking tag is any one of the multiple marking tags.

[0062] The marking operation is a selection operation performed on the marking button on the report viewing interface, and the selection operation includes but is not limited to a click operation, a double-click operation, a slide operation, etc. The marking buttons on the report viewing interface include a "suitable" marking button and an "unsuitable" marking button. Clicking different marking buttons will display different marking labels. The marking labels corresponding to the "suitable" marking button include but are not limited to rich experience, suitable time of employment, etc., and the marking labels corresponding to the "unsuitable" marking button include but are not limited to older age, lack of experience, unsuitable time of employment, etc.

[0063] It should be noted that each marking tag is associated with an interview dimension. After marking the target interview report, the marking result can be sent back to the dimension library to update the dimension range corresponding to the interview dimension in the dimension library based on the interview dimension associated with the marking result. Based on this, the updated dimension range can be used to generate a more accurate target interview report for subsequent candidates.

[0064] In actual applications, the simple marking process and targeted marking feedback greatly improve the enthusiasm of recruiters to participate in feedback, thereby increasing the marking rate; the marking operation can achieve effective communication between recruiters and digital people, avoiding information asymmetry caused by lack of communication, and can timely adjust the recruitment needs of recruiters according to the feedback of recruiters, so that digital people can screen out candidates who are more in line with the actual recruitment needs for recruiters. In addition, the whole process is simple and intuitive, and high-quality feedback submission can be completed without complicated operations, which enhances user satisfaction and system usability.

[0065] In the above or following embodiments, the dimension library may include interview dimensions, dimension ranges corresponding to the interview dimensions, etc. Among them, the interview dimensions are evaluation factors for applicants during the interview process, and the interview dimensions can be mined from data sources such as interview chat records, interview call records, applicant resumes, recruitment information, etc., and the mined interview dimensions are stored in a preset dimension library.

[0066] Optionally, the interview dimensions stored in the dimension library can be divided into general concern dimensions and core concern dimensions. The core concern dimensions are usually the interview dimensions that most recruiters generally need to know, including but not limited to age, gender, whether full-time or part-time, education, work experience, current employment status, etc. The general concern dimensions are usually the interview dimensions that some recruiters need to know, including but not limited to the fastest time to join the job, the distance between the workplace and the residence, and the acceptable working time period.

[0067] In this embodiment, a variety of implementation methods may be used to determine the dimension range in the dimension library.

[0068] In an optional implementation, a conversational question-and-answer method may be used to ask the recruiter about the dimension range that he needs to set for the interview dimension. Figure 5 A schematic diagram of determining a dimensional range is provided for an exemplary embodiment of the present application, such as Figure 5 As shown, the dimension range configuration interface includes a voice initiation button, and the voice initiation button may be a “voice call”. After the recruiter selects the “voice call” button, a voice configuration request may be initiated.

[0069] When a dimension range configuration request initiated by a recruiter is received, a dimension range configuration interface corresponding to the target recruitment information can be displayed to the recruiter. In response to the voice configuration request initiated by the recruiter on the dimension range configuration interface, the range guidance question corresponding to the interview dimension is broadcast to the recruiter, and the range guidance question is used to guide the recruiter to provide the dimension range set for the interview dimension. That is, in response to the recruiter's selection operation on the voice initiation button on the dimension range configuration interface, a voice call can be established with the recruiter, and the range guidance question is broadcast to the recruiter during the voice call. The recruiter only needs to reply to the range guidance question in the usual way of calling. During the voice call, the reply content provided by the recruiter to the range guidance question can be obtained, the dimension range corresponding to the interview dimension can be extracted from the reply content, and the dimension range corresponding to the interview dimension can be stored in a preset dimension library. Among them, the reply content obtained is a voice reply, and the voice reply can be converted into a text reply first, and then the dimension range corresponding to the interview dimension can be extracted from the text reply.

[0070] Figure 6 Another schematic diagram of determining a dimensional range is provided for an exemplary embodiment of the present application. Figure 6As shown, the voice initiation button on the dimension range configuration interface is "long press to speak". After the dimension configuration interface is displayed to the recruiter, the dimension guidance questions can be automatically displayed to the recruiter on the dimension range configuration interface. The recruiter can respond to the dimension guidance questions in the form of voice input by long pressing the voice initiation button. The digital human service end can convert the recruiter's response from voice to text, and display the content of the recruiter's response in text form on the dimension range configuration interface. After obtaining the content of the recruiter's response, the dimension range corresponding to the interview dimension can be extracted from the content of the recruiter's response, and the dimension range corresponding to the interview dimension can be stored in the preset dimension library.

[0071] It is worth mentioning that the above solution of determining the dimension range through dialogue question and answer is not only applicable to adding dimension ranges to the dimension library, but also applicable to updating dimension ranges in the dimension library. When the recruiter proposes a new dimension range for the interview dimension, the new dimension range can be used to overwrite the old dimension range added by the recruiter in the dimension library. In addition, the dimension range in the initial interview report is consistent with the dimension range in the dimension library. Therefore, when the dimension range in the dimension library is updated through dialogue question and answer, the dimension range in the initial interview report will also be updated accordingly.

[0072] In summary, in this embodiment, through voice input, recruiters can quickly update recruitment requirements, reduce manual input errors, and improve the efficiency of setting dimension ranges for interview dimensions; through question-and-answer questions and guiding words, recruiters can be gradually guided to further refine their recruitment requirements, thereby improving the accuracy of setting dimension ranges for interview dimensions.

[0073] Of course, other methods may be used to construct the dimension range. For example, the recruiter may manually input the dimension range corresponding to the interview dimension. This embodiment does not limit the method for constructing the dimension range.

[0074] In the above or below embodiments, a target interview video may be generated in a variety of implementations.

[0075] In an optional implementation, a to-be-processed interview video generated by a recruitment question-and-answer session between a job applicant and a digital human can be collected. The to-be-processed video is a video obtained by directly recording the interview process, and the target interview video is a video obtained after processing the to-be-processed video. The target interview video can be used to generate an initial interview report for the job applicant.

[0076] After obtaining the video to be processed, the interview questions and corresponding interview answers can be extracted from the interview video to be processed. Interview questions are questions that the digital human asks the applicant during the interview process, and interview answers are answers given by the applicant to the interview questions. Each question has an interview answer. Based on the correlation between the interview questions and the interview dimensions contained in the preset question library, the interview dimensions associated with each interview question can be determined.

[0077] For any interview question, the dimension range corresponding to the interview dimension associated with the interview question can be extracted from the preset dimension library; the interview answer corresponding to the interview question is semantically understood to extract dimension data from the interview answer as the dimension data corresponding to the interview question. Then, the dimension range corresponding to the interview question and the dimension data corresponding to the interview question can be combined and displayed in the interview video to generate a target interview video for the applicant. The combination of the dimension range and the dimension data can be added to the interview video to be processed in a hovering form to generate a target interview video. The target interview video includes the interview record in video form and the applicant information in text form. The applicant information includes dimension data, dimension range, etc.

[0078] Among them, the interview answers can be semantically understood to understand the semantic information represented by the interview answers, and the information that matches the interview questions can be screened out from the semantic information as the dimensional data corresponding to the interview questions. For example, if the interview question asks the applicant's age, the interview answer "I am 25 years old" can be semantically understood, and the information "25 years old" that matches the interview question can be extracted from the understood voice information as the dimensional data corresponding to the interview question. The dimension range corresponding to the interview dimension of age in the dimension library is "age between 18-40 years old". Based on this, the dimension range and dimension data can be combined, and the combination "age between 18-40 years old, 25 years old" can be displayed in the target interview video in a hovering form.

[0079] In practical applications, processing the interview video to be processed and generating a target interview video containing dimensional range and dimensional data can help recruiters quickly understand the compatibility of applicants with recruitment requirements, so as to screen out suitable applicants from a large number of applicants and improve the accuracy of recruitment.

[0080] Optionally, after the video to be processed is collected, the interview video to be processed may be reviewed first. If the interview video to be processed meets the preset review conditions, it is determined to have passed the review, and the interview video to be processed may continue to be processed to generate a target interview video. If the interview video to be processed does not meet the preset review conditions, it is determined to have failed the review, and a video fault prompt may be sent to the applicant's client to remind the applicant that the interview video has not passed the review, and the applicant is advised to re-interview. Among them, the preset review conditions include but are not limited to: the interview video to be processed failed to be generated, the applicant's face in the interview video to be processed appears for less than a preset threshold, there is noise in the interview video to be processed, and the interview video to be processed violates laws and regulations or public order and good customs. In the case of multiple review conditions, as long as any one of the review conditions is not met, it is determined that the interview video to be processed generated for the applicant has not passed the review.

[0081] In this optional implementation, the process of generating the target interview video based on the interview video to be processed occurs after the interview is completed, while the process of generating the interview video to be processed occurs during the interview. The process of generating the interview video to be processed is explained below.

[0082] During the interview process, in response to the interview request initiated by the applicant for the target recruitment information, the target interview dimension adapted to the target recruitment information can be obtained. Interview questions corresponding to the target interview dimension are extracted from the preset question library to form an interview question list. The number of target interview dimensions is usually multiple, and accordingly, the number of interview questions included in the interview question list is usually multiple. There is a questioning sequence between the multiple interview questions in the interview question list, and the interview questions can be broadcast to the applicant according to the questioning sequence. Then, the interview answers given by the applicant to the interview questions can be obtained. When it is determined that the interview answers are adapted to the interview questions and there are no unanswered interview questions in the interview question list, a pending interview video is generated for the applicant.

[0083] For any interview question, after obtaining the interview answer given by the candidate to the interview question, it can be determined whether the interview answer is compatible with the interview question. If it is not compatible, the preset transition words or introductory words will be sent to the candidate, and a new interview question will be regenerated for the interview dimension corresponding to the interview question and sent to the candidate; if it is compatible, the next interview question will be selected from the interview question list and sent to the candidate. If there is no next interview question at this time, the closing remarks will be sent to the candidate to end the interview. The order of questions in the interview question list is not fixed, and the order of questions in the interview question list can be flexibly adjusted according to the candidate's answers.

[0084] It should be noted that transition words or introductory words are used to connect interview questions, and closing words are used to indicate the end of the interview. Transition words, introductory words or closing words can be pre-configured in the question library or generated in real time by a digital human.

[0085] Optionally, in an exemplary implementation scheme for determining whether an interview answer is compatible with an interview question, slot identification may be performed on the interview answer to identify dimension data related to the interview dimension in the interview answer. The answer requirements under the interview dimension corresponding to the interview question are obtained from a preset question library, and when the dimension data meets the answer requirements, it is determined that the interview answer is compatible with the interview question.

[0086] Among them, a slot is an information unit with specific attributes in text information. In this embodiment, the specific attribute represented by the slot is associated with the interview dimension. For example, a slot can be a date, age, name, etc. The answer requirement refers to the conditions that need to be met to determine whether the dimensional data in the interview answer is compatible with the interview dimension corresponding to the interview question. For example, if the interview dimension is age, the answer requirement can be "the interview answer contains numbers, and the number range is between 0-100".

[0087] In summary, the automated digital interview process can reduce the workload of recruiters and improve interview efficiency. In addition, during the online interview process, the semantic understanding of the interview answers of the applicants can be performed, and the subsequent interview questions can be flexibly adjusted based on the semantic understanding results, avoiding repeated questions to the applicants and improving the interview experience of the applicants.

[0088] In this embodiment, a preset question bank is pre-constructed before the interview. The question bank contains interview questions corresponding to the interview dimensions. In this embodiment, a variety of implementation methods can be used to generate interview questions.

[0089] In an exemplary optional implementation, the recruiter may manually define the interview questions, and the digital human server may collect the interview questions configured by the recruiter for each interview dimension, and store the collected interview questions in a preset question library.

[0090] In another exemplary optional implementation, interview questions can be automatically generated by a question generation model based on artificial intelligence. The digital human server can extract multiple interview dimensions from the dimension library, cluster the multiple interview dimensions, and obtain an interview dimension group; the interview dimension and the interview dimension group are respectively input into the question generation model based on artificial intelligence, so that the question generation model generates interview questions corresponding to the interview dimension and interview questions corresponding to the interview dimension group based on the pre-learned question generation knowledge. Then, the interview questions corresponding to the interview dimension and the interview questions corresponding to the interview dimension group can be stored in a preset question library.

[0091] In this optional implementation, the question generation knowledge used by the question generation model is learned during the model training phase. In this embodiment, a large number of interview dimension samples and interview question samples pre-configured for the interview dimension samples can be used as training samples to train the question generation model. During the training process, the question generation model will learn and store the mapping relationship between the interview dimension and the interview question as question generation knowledge, so that after the question generation model receives the interview dimension in the future, it can output the interview question corresponding to the received interview dimension based on the learned mapping relationship.

[0092] Optionally, during the model training process, the interview dimension samples can be vectorized and extracted to obtain dimensional features of the interview dimension samples, which may contain one or more interview dimensions; the interview question samples are converted into embedded representations, and the embedded representations are encoded to generate question features of the interview question samples; based on the association between the interview dimension samples and the interview question samples, a mapping relationship between dimensional features and question features is constructed as the mapping relationship between the interview dimensions and the interview questions, that is, question generation knowledge, and the constructed mapping relationship between dimensional features and question features is stored in the knowledge base associated with the question generation model.

[0093] On this basis, in the process of generating interview questions for the interview dimension using the trained question generation model, after the interview dimension is input into the question generation model, the question generation model can extract features of the interview dimension to obtain the dimensional features of the interview dimension; calculate the correlation between the dimensional features and the dimensional features contained in the knowledge base, and use the question features corresponding to the dimensional features whose correlation exceeds a specified threshold as the question features corresponding to the interview dimension; and decode the question features corresponding to the interview dimension to generate interview questions corresponding to the interview dimension.

[0094] Based on this, by using the question generation model to generate interview questions for the interview dimension, a large number of interview questions can be automatically generated according to job requirements in a short period of time, saving the time and effort required for the recruiter to manually input. In addition, using the question generation model to generate interview questions for the interview dimension can ensure the accuracy and pertinence of the generated interview questions, reducing errors and omissions caused by manual input. This helps recruiters evaluate applicants more efficiently and accurately, and improves recruitment quality and efficiency.

[0095] In the above or below embodiments, a variety of implementation methods may be used to calculate the dimension scores and the comprehensive scores.

[0096] In an optional implementation, the dimension ranges corresponding to the target recruitment information of multiple target interview dimensions that are adapted to the target recruitment information can be obtained; and the dimension data corresponding to each target interview dimension can be extracted from the interview answers contained in the target interview video. For any target interview dimension, the dimension range corresponding to the target interview dimension and the dimension data corresponding to the target interview dimension are input into the scoring model based on artificial intelligence, so that the scoring model calculates the dimension score under the target interview dimension according to the pre-learned scoring rules.

[0097] In this optional implementation, the scoring rules used by the scoring model are learned during the model training phase. In this embodiment, a large number of training samples can be used to train the scoring model, and the training samples include dimension data, dimension ranges, and dimension scores with associated relationships. During the training process, the question model can learn the scoring rules under the interview dimension based on the dimension scores corresponding to the dimension data under the dimension range. This allows the subsequent scoring model to output the dimension score under the interview dimension based on the scoring rules under the interview dimension corresponding to the dimension data after receiving the dimension range and dimension data.

[0098] Optionally, during the model training process, the dimension data and the dimension range can be vectorized and extracted respectively to obtain a first vector corresponding to the dimension data and a second vector corresponding to the dimension range; the correlation coefficient between the first vector and the second vector is calculated, and the correlation coefficient between the first vector and the second vector can be a Pearson correlation coefficient, a Spearman rank correlation coefficient, etc. This correlation coefficient is intended to measure the closeness of the correlation between two vectors, to help people understand the interaction between vectors and how changes in one vector affect another vector. Then, based on the dimension scores associated with the dimension data, the correlation relationship between the correlation coefficient and the dimension score under the interview dimension can be established, and the correlation relationship between the correlation coefficient and the dimension score under multiple interview dimensions can be stored in the knowledge base associated with the scoring model.

[0099] Based on this, in the process of using the trained scoring model to calculate the dimensional scores of the target interview video under each target interview dimension, after the dimensional range and dimensional data corresponding to the target interview dimension are input into the scoring model, the scoring model can respectively perform vectorized extraction on the dimensional data and dimensional range corresponding to the target interview dimension, obtain a first vector corresponding to the dimensional data and a second vector corresponding to the dimensional range, and calculate the target correlation coefficient between the first vector and the second vector; based on the correlation relationship between the correlation coefficients and dimensional scores under multiple interview dimensions stored in the knowledge base, infer the dimensional score corresponding to the target correlation coefficient as the dimensional score of the applicant under the target interview dimension.

[0100] In this implementation, a variety of implementation schemes can be used to calculate the comprehensive score corresponding to the applicant. For example, the dimension weights corresponding to the multiple target interview dimensions under the target recruitment information can be determined, and based on the dimension weights corresponding to each target interview dimension, the dimension scores under the multiple target interview dimensions are weighted and calculated to obtain the comprehensive score corresponding to the applicant.

[0101] Among them, the dimension weight corresponding to the target interview dimension can be pre-defined by the recruiter, or it can be configured by the digital human server based on the recruiter's attention to different target interview dimensions. Different target recruitment information has different recruitment positions and different requirements for applicants. Therefore, different target interview dimensions can be configured for different target recruitment information, and different dimension weights can be configured for different target interview dimensions. Weighted calculation of dimension scores under multiple target interview dimensions means assigning different weights to multiple dimension scores, and then summing or averaging these weighted values. The importance of each dimension score can be adjusted through weighted calculation to reflect their relative contribution in the overall calculation.

[0102] For example, if the target interview dimensions that match the target recruitment information include age, gender, education level and work experience, the dimension scores corresponding to these four target interview dimensions are 100, 60, 90 and 90 respectively, and the dimension weights corresponding to the four target interview dimensions are 20%, 10%, 60% and 10% respectively. Then, the dimension scores under these target dimensions are weighted calculated as (100*20%+60*10%+90*60%+90*10%) / 4=89, and the corresponding comprehensive score of the applicant is 89 points.

[0103] Of course, other implementation schemes can also be used in this embodiment to calculate the comprehensive score of the applicant. For example, the total score or average score of all dimension scores can be directly used as the total score of the applicant. In addition, other implementation methods can also be used to calculate the dimension score, such as the previous example, pre-configure the scoring rules for the dimension range, and calculate the dimension score of the target interview video under each target interview dimension according to the mapping relationship between the dimension data and the dimension score contained in the scoring rules. This embodiment does not limit the calculation method of the dimension score and the comprehensive score.

[0104] In summary, in this embodiment, the interview performance of the applicants during the interview is scored using a pre-trained scoring model, which can achieve an objective analysis of the interview process and reduce the impact of human subjective factors on the evaluation results. For different recruitment information, different dimension weights are configured for the interview dimensions involved in the recruitment information, so that the comprehensive score calculated for the applicant is more in line with the recruitment needs of the recruiter, which can improve the accuracy of the recruiter's screening of applicants based on the comprehensive score, thereby improving the accuracy of recruitment.

[0105] Figure 7 This is a flow chart of an exemplary application scenario provided by another exemplary embodiment of the present application. Figure 7 As shown, this application scenario can explain the interview process based on digital humans from three stages: "before the interview", "during the interview" and "after the interview".

[0106] 1) “Pre-interview” stage:

[0107] 711. Mining interview dimensions from data sources such as chat records, call records, and recruitment information, and storing the mined interview dimensions in a preset dimension library;

[0108] 712. Collect the dimension ranges set by the recruiter for the interview dimensions through dialogue question and answer, and store the dimension ranges in a preset dimension library;

[0109] 713. Cluster the interview dimensions in the dimension library to obtain interview dimension groups;

[0110] 714, input the interview dimension and the interview dimension group into the question generation model based on artificial intelligence, so as to generate interview questions corresponding to the interview dimension and the interview questions corresponding to the interview dimension group through the question generation model, and store the generated interview questions in a preset question library.

[0111] 2) “Interviewing” stage:

[0112] 721. After monitoring the applicant entering the interview room, the target recruitment information of the applicant can be obtained, and a list of interview questions for the applicant can be extracted from the question library based on the target recruitment information;

[0113] 722. Determine the interview questions that need to be asked based on the questioning order of multiple interview questions in the interview question list;

[0114] 723. Convert the interview questions that need to be asked from text to voice, and broadcast the interview questions in voice to the applicants;

[0115] 724. Receive the interview answers given by the applicant to the interview questions, and perform slot identification on the received interview answers to determine whether the interview answers given by the applicant are suitable for the interview questions;

[0116] 725. If the interview question and answer do not match the interview question, a preset transitional phrase or introductory phrase is sent to the applicant, and a new interview question is generated for the interview dimension corresponding to the interview question, and step 723 is executed;

[0117] 726. If the interview question and answer are compatible with the interview question, determine the next interview question in the interview question list that needs to be asked based on the question order in the interview question list and the current candidate's answer;

[0118] 727. When all the interview questions in the interview question list have been asked, it is determined that the applicant has completed the interview and a pending interview video is generated for the applicant.

[0119] 3) “Post-interview” stage:

[0120] 731, review the generated interview videos to be processed;

[0121] 732, if the review fails, a video failure prompt will be sent to the applicant to remind the applicant that the interview video has not passed the review and suggest the applicant to re-interview;

[0122] 733, if the review is passed, formatting the interview video to be processed, so as to extract the dimension range set by the recruiter and the dimension data answered by the applicant from the interview video to be processed, and adding the dimension range and the dimension data combination to the interview video to be processed to generate a target interview video;

[0123] 734, according to the preset dimension evaluation rules, evaluate the candidate's question-answering performance in the target interview video to generate interview evaluation information for the candidate; for the target interview dimension adapted to the target recruitment information, according to the dimension range corresponding to the target interview dimension and the scoring rules pre-configured for the dimension range, calculate the dimension score of the target interview video under each target interview dimension; based on the dimension score of the target interview video under each target interview dimension, calculate the corresponding comprehensive score of the candidate;

[0124] 735, based on the interview evaluation information and the comprehensive score, generate an initial interview report for the applicant, the initial interview report including the interview evaluation information, the comprehensive score, the dimension range and the dimension data; send the initial interview report and the report viewing notification to the recruiter to notify the recruiter to view the initial interview report;

[0125] 736, the recruiter client determines whether the recruiter is viewing the initial interview report for the first time. If it is the first time, the report viewing process is played to the recruiter and the recruiter is guided to view the initial interview report. If it is not the first time, the recruiter is directly guided to view the initial interview report.

[0126] 737, in response to the modification operation initiated by the recruiter on the dimension range in the initial interview report, modify the initial interview report to obtain a target interview report; update the corresponding dimension range in the dimension library, so that the updated interview dimension can be used in the subsequent interview process to more accurately evaluate the applicant;

[0127] 738. In response to a marking operation initiated by the recruiter for a target interview report, multiple marking tags corresponding to the marking operation are displayed; in response to a selection operation performed by the recruiter on a target marking tag, a target marking tag is added to the target interview report to mark the interview report, where the target marking tag is any one of the multiple marking tags; the marking result is transmitted back to the dimension library to update the dimension range corresponding to the interview dimension in the dimension library based on the interview dimension associated with the marking result, so that the updated dimension range can be used later to generate a more accurate target interview report for subsequent applicants.

[0128] Based on this, in this embodiment, a complete online interview process is designed from the aspects of dimension construction, question generation, online interview, generation and marking of interview reports, etc. The generated interview report contains a comprehensive score obtained by objectively evaluating the interview process of the applicant. The comprehensive score can help the recruiter quickly and accurately understand the applicant's competence for the recruitment position, help the recruiter screen out more suitable applicants, and improve the efficiency and accuracy of recruitment.

[0129] It should be noted that the execution subject of each step of the method provided in the above embodiment can be the same device, or the method can be executed by different devices. For example, the execution subject of steps 201 to 207 can be device A; for another example, the execution subject of steps 201 and 202 can be device A, and the execution subject of step 203 can be device B; and so on.

[0130] In addition, in some of the processes described in the above embodiments and the accompanying drawings, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this article or executed in parallel, and the sequence numbers of the operations, such as 201, 202, etc., are only used to distinguish between different operations, and the sequence numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent the order of precedence, and do not limit the "first" and "second" to be different types.

[0131] Figure 8 A schematic diagram of a computing platform provided as another exemplary embodiment of the present application. Figure 8 As shown, the computing platform includes: a memory 80 , a processor 81 and a communication component 82 .

[0132] The processor 81 is coupled to the memory 80 and is configured to execute the computer program in the memory 80 to:

[0133] In the process of interviewing the applicants participating in the target recruitment information based on the digital human, the target interview video generated by the recruitment question and answer session between the applicant and the digital human is collected through the communication component 82;

[0134] According to the preset dimension evaluation rules, the candidate's question-answering performance in the target interview video is evaluated to generate interview evaluation information for the candidate;

[0135] Based on the target interview dimension adapted to the target recruitment information, extract the dimension range corresponding to the target interview dimension from the preset dimension library;

[0136] Calculate the dimension score of the target interview video under each target interview dimension according to the dimension range corresponding to the target interview dimension and the scoring rules pre-configured for the dimension range;

[0137] Calculate the candidate's corresponding comprehensive score based on the dimension scores of the target interview video under each target interview dimension;

[0138] Based on the interview evaluation information and comprehensive score, generate an initial interview report for the applicant and display the initial interview report, which includes the interview evaluation information, comprehensive score, dimension range and dimension data;

[0139] In response to a correction operation initiated by the recruiter on the dimension range in the initial interview report, the initial interview report is corrected to obtain a target interview report.

[0140] In an optional embodiment, the processor 81 is further used to:

[0141] Collect the interview videos to be processed generated by the recruitment questions and answers between the applicant and the digital human;

[0142] Extract interview questions and corresponding interview answers from the interview videos to be processed;

[0143] Based on the correlation between the interview questions and the interview dimensions contained in the preset question bank, determine the interview dimensions associated with each interview question in the interview video;

[0144] For any interview question, extract the dimension range corresponding to the interview dimension associated with the interview question from the dimension library;

[0145] Perform semantic understanding on the interview answers corresponding to the questions to extract dimension data from the interview answers as the dimension data corresponding to the interview questions;

[0146] The dimension range corresponding to the interview question and the dimension data corresponding to the interview question are combined and displayed in the interview video to generate a target interview video for the applicant.

[0147] In an optional embodiment, the processor, in the process of generating the interview video to be processed, is further configured to:

[0148] In response to an interview request initiated by an applicant for a target recruitment information, obtaining a target interview dimension adapted to the target recruitment information;

[0149] Extract interview questions corresponding to the target interview dimensions from the question bank to form an interview question list;

[0150] Announce the interview questions to the candidate in the order in which they are included in the interview question list;

[0151] Obtain the interview answers given by the candidates to the interview questions;

[0152] When it is determined that the interview answers are suitable for the interview questions and there are no unanswered interview questions in the interview question list, a pending interview video is generated for the applicant.

[0153] In an optional embodiment, in the process of determining whether the interview answer is adapted to the interview question, the processor 81 is further configured to:

[0154] Perform slot recognition on interview answers to identify dimension data related to the interview dimensions in the interview answers;

[0155] From the preset question library, obtain the answer requirements under the interview dimension corresponding to the interview questions;

[0156] When the dimensional data meets the answer requirements, make sure that the interview answers are compatible with the interview questions.

[0157] In an optional embodiment, in the process of calculating the dimension score of the target interview video under each target interview dimension according to the dimension range corresponding to the target interview dimension and the scoring rule pre-configured for the dimension range, the processor 81 is further used to:

[0158] Get the dimension range corresponding to the target interview dimension under the target recruitment information;

[0159] Extracting dimension data under the target interview dimension from the interview answers contained in the target interview video;

[0160] The dimension range and dimension data are input into the AI-based scoring model so that the scoring model can calculate the dimension score under each target interview dimension according to the pre-learned scoring rules.

[0161] In an optional embodiment, in the process of calculating the comprehensive score corresponding to the applicant based on the dimension score of the target interview video under each target interview dimension, the processor 81 is further configured to:

[0162] Determine the dimension weights corresponding to multiple target interview dimensions under the target recruitment information;

[0163] Based on the dimension weights corresponding to each target interview dimension, the dimension scores under multiple target interview dimensions are weightedly calculated to obtain the corresponding comprehensive scores of the applicants.

[0164] In an optional embodiment, the processor 81 is further configured to:

[0165] A report viewing interface is shown to the recruiter, and a target interview report is shown on the report viewing interface;

[0166] In response to a marking operation initiated by a recruiter on a target interview report, displaying a plurality of marking tags corresponding to the marking operation;

[0167] In response to the recruiter's selection operation on the target marking tag, the target marking tag is added to the target interview report to mark the interview report, where the target marking tag is any one of the multiple marking tags.

[0168] In an optional embodiment, the interview dimension in the dimension library corresponds to a dimension range, and the processor 81 is further configured to:

[0169] Display the dimension range configuration interface corresponding to the target recruitment information;

[0170] In response to a voice configuration request initiated by the recruiter on the dimension range configuration interface, the dimension guiding questions corresponding to the interview dimension are broadcast to the recruiter;

[0171] Obtain the responses provided by recruiters to dimension-guided questions;

[0172] The dimension range corresponding to the interview dimension is extracted from the response content, and the dimension range corresponding to the interview dimension is stored in the dimension library.

[0173] In an optional embodiment, the processor 81 is further configured to:

[0174] Extract multiple interview dimensions from the dimension library;

[0175] Cluster multiple interview dimensions to obtain interview dimension groups;

[0176] Inputting the interview dimensions and the interview dimension groups into the question generation model based on artificial intelligence, so that the question generation model generates interview questions corresponding to the interview dimensions and interview questions corresponding to the interview dimension groups based on the pre-learned question generation knowledge;

[0177] The interview questions corresponding to the interview dimensions and the interview questions corresponding to the interview dimension groups are stored in the question library.

[0178] Further, if Figure 8As shown, the computing platform also includes: a display 83, a power component 84, an audio component 85 and other components. Figure 8 Only some components are shown schematically, which does not mean that the computing platform only includes Figure 8 Components shown.

[0179] It is worth noting that the technical details in the above-mentioned embodiments of the computing device can refer to the relevant description of the actions of the computing device in the above-mentioned embodiments of the interface comparison test method. In order to save space, they will not be repeated here, but this should not cause a loss in the scope of protection of this application.

[0180] Accordingly, an embodiment of the present application further provides a computer-readable storage medium storing a computer program, which, when executed, can implement the steps that can be executed by the computing platform in the above method embodiment.

[0181] Accordingly, an embodiment of the present application also provides a computer program product, which can implement the steps in the above method embodiment when the computer program included in the product is executed.

[0182] Above Figure 8 The memory in the computer is used to store computer programs and can be configured to store various other data to support operations on the computing platform. Examples of such data include instructions for any application or method operating on the computing platform, contact data, phone book data, messages, pictures, videos, etc. The memory can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0183] Above Figure 8 The communication component in is configured to facilitate wired or wireless communication between the device where the communication component is located and other devices. The device where the communication component is located can access a wireless network based on a communication standard, such as WiFi, 2G, 3G, 4G / LTE, 5G and other mobile communication networks, or a combination thereof. In an exemplary embodiment, the communication component receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.

[0184] Above Figure 8The display includes a screen, and the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from a user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundary of a touch or slide action, but also detect the duration and pressure associated with the touch or slide operation.

[0185] Above Figure 8 The power supply component in the device provides power to various components of the device where the power supply component is located. The power supply component may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to the device where the power supply component is located.

[0186] Above Figure 8 The audio component in can be configured to output and / or input audio signals. For example, the audio component includes a microphone (MIC), and when the device where the audio component is located is in an operating mode, such as a call mode, a recording mode, and a speech recognition mode, the microphone is configured to receive an external audio signal. The received audio signal can be further stored in a memory or sent via a communication component. In some embodiments, the audio component also includes a speaker for outputting an audio signal.

[0187] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0188] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0189] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0190] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0191] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0192] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0193] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0194] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0195] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0196] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. An interview processing method, characterized in that: The method comprises: In the process of interviewing the applicants participating in the target recruitment information based on the digital human, collecting the target interview video generated by the recruitment questions and answers between the applicants and the digital human; According to the preset dimension evaluation rules, the question-answering performance of the candidate in the target interview video is evaluated to generate interview evaluation information of the candidate; Based on the target interview dimension adapted to the target recruitment information, extracting a dimension range corresponding to the target interview dimension from a preset dimension library; Calculate the dimension score of the target interview video under each target interview dimension according to the dimension range corresponding to the target interview dimension and the scoring rules pre-configured for the dimension range; Calculate the comprehensive score of the candidate based on the dimension score of the target interview video under each target interview dimension; Based on the interview evaluation information and the comprehensive score, generate an initial interview report for the applicant, and display the initial interview report, wherein the initial interview report includes the interview evaluation information, the comprehensive score, the dimension range and the dimension data; In response to a modification operation initiated by the recruiter on the dimension range in the initial interview report, the initial interview report is modified to obtain a target interview report.

2. The method according to claim 1, characterized in that The process of generating the target interview video includes: Collecting the interview video to be processed generated by the recruitment question and answer session between the applicant and the digital human; Extracting interview questions and corresponding interview answers from the interview video to be processed; Based on the correlation between the interview questions and the interview dimensions contained in the preset question library, determine the interview dimensions associated with each interview question in the interview video; For any interview question, extracting a dimension range corresponding to the interview dimension associated with the interview question from the dimension library; Performing semantic understanding on the interview answer corresponding to the question to extract dimension data from the interview answer as the dimension data corresponding to the interview question; The dimensional range corresponding to the interview question and the dimensional data corresponding to the interview question are combined and displayed in the interview video to generate a target interview video for the applicant.

3. The method according to claim 2, characterized in that The generation process of the interview video to be processed includes: In response to an interview request initiated by the applicant for the target recruitment information, obtaining a target interview dimension adapted to the target recruitment information; Extracting interview questions corresponding to the target interview dimension from the question library to form an interview question list; Reporting the interview questions to the applicant in the order of questions included in the interview question list; Obtaining the interview answers given by the applicant to the interview questions; When it is determined that the interview answer is adapted to the interview question and there is no unanswered interview question in the interview question list, a to-be-processed interview video is generated for the applicant.

4. The method according to claim 3, characterized in that Determining that the interview answer is suitable for the interview question includes: Performing slot identification on the interview answer to identify dimension data related to the interview dimension in the interview answer; Obtaining answer requirements under the interview dimension corresponding to the interview question from the preset question library; When the dimensional data meets the answer requirement, it is determined that the interview answer is adapted to the interview question.

5. The method according to claim 1, characterized in that According to the dimension range corresponding to the target interview dimension and the scoring rules pre-configured for the dimension range, the dimension score of the target interview video under each target interview dimension is calculated, including: Obtaining dimension ranges corresponding to each of a plurality of target interview dimensions adapted to the target recruitment information under the target recruitment information; Extracting the dimensional data corresponding to each target interview dimension from the interview answers contained in the target interview video; For any target interview dimension, the dimension range corresponding to the target interview dimension and the dimension data corresponding to the target interview dimension are input into the artificial intelligence-based scoring model, so that the scoring model can calculate the dimension score under the target interview dimension according to the pre-learned scoring rules.

6. The method according to claim 5, characterized in that Based on the dimension scores of the target interview video under each target interview dimension, the corresponding comprehensive score of the applicant is calculated, including: Determine the dimension weights corresponding to each of the multiple target interview dimensions under the target recruitment information; Based on the dimension weights corresponding to each target interview dimension, the dimension scores under the target interview dimension are weightedly calculated to obtain the corresponding comprehensive score of the applicant.

7. The method according to claim 1, characterized in that Also includes: Displaying a report viewing interface to the recruiter, wherein the target interview report is displayed on the report viewing interface; In response to a marking operation initiated by the recruiter on the target interview report, displaying a plurality of marking tags corresponding to the marking operation; In response to the recruiter's selection operation on a target marking tag, the target marking tag is added to the target interview report to mark the interview report, wherein the target marking tag is any one of the multiple marking tags.

8. The method according to claim 1, characterized in that The interview dimension in the dimension library corresponds to a dimension range, and the method further includes: Display the dimension range configuration interface corresponding to the target recruitment information; In response to a voice configuration request initiated by the recruiter on the dimension range configuration interface, broadcasting dimension guidance questions corresponding to the interview dimension to the recruiter; Obtaining the response content provided by the recruiter to the dimension guiding question; The dimension range corresponding to the interview dimension is extracted from the reply content, and the dimension range corresponding to the interview dimension is stored in the dimension library.

9. The method according to claim 1, characterized in that: The method further comprises: extracting a plurality of interview dimensions from the dimension library; Clustering the multiple interview dimensions to obtain an interview dimension group; Inputting the interview dimension and the interview dimension group into an artificial intelligence-based question generation model, so that the question generation model generates interview questions corresponding to the interview dimension and interview questions corresponding to the interview dimension group based on pre-learned question generation knowledge; The interview questions corresponding to the interview dimension and the interview questions corresponding to the interview dimension group are stored in the question library.

10. A computing device, characterized in that Includes memory, processor, and communication components; The memory is used to store one or more computer instructions; The processor is coupled to the memory and the communication component, and is used to execute the one or more computer instructions to execute the interview processing method according to any one of claims 1-9.

11. A computer-readable storage medium storing computer instructions, characterized in that: When the computer instructions are executed by one or more processors, the one or more processors are caused to execute the interview processing method according to any one of claims 1 to 9.

Images