Interview questioning method and device, storage medium and computer program product

Abstract

The embodiment of the invention provides an interview questioning method and device, a storage medium and a computer program product. In the embodiment of the invention, after the answer content provided by the interviewer for the interview question is obtained, the answer content can be detected, and if the answer content is detected to be lack of any key information required by the interview question, the interview question and the answer content are provided for the first large model; and sending the first prompt information to the first big model, so that the first big model generates a question asking question for the interview question according to the first prompt information, and sending the question asking question to the interviewer to guide the interviewer to provide key information required by question asking. Wherein when the answer content provided by the interviewer is not enough to answer the interview intention of the interview question, a new interview question is regenerated for the interview intention, and the new interview question is adopted to ask the interviewer, so that the interviewer can supplement information in time; the recruiter can quickly and accurately obtain more effective information for the interviewer, so that the recruitment experience of the recruiter is improved.

Description

Technical Field

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

[0002] As companies continue to pursue recruitment efficiency and quality, and interviewees pay more and more attention to the interview experience, AI interviews have emerged as an innovative recruitment method. AI interviews simulate real interview scenarios and conduct intelligent assessments of interviewees, thereby providing companies with more efficient and accurate recruitment decision support.

[0003] At present, AI interviews usually present preset interview questions to the interviewee in sequence, and evaluate the interviewee's ability based on the job seeker's answers. However, the interviewee's comprehension ability varies, and the answers given to the interview questions may not contain the effective information required by the employer, making it impossible for the employer to evaluate the interviewee's ability, which in turn leads to a poor recruitment experience for the employer. Summary of the invention

[0004] Multiple aspects of the present application provide an interview follow-up method, device, storage medium and computer program product to improve the recruitment experience of the recruiter during the interview process.

[0005] The present application embodiment provides an interview questioning method, including:

[0006] After obtaining the answer content provided by the interviewee to any interview question, extracting key information required for the interview question from the answer content;

[0007] If it is determined that the answer content lacks any key information required by the interview question, constructing the interview question and the answer content as input data of the first model;

[0008] Obtaining preset first prompt information, wherein the first prompt information is used to instruct the first large model to ask follow-up questions based on the interview questions and answers in the input data, wherein the first prompt information also includes interview question examples, answer content examples, and corresponding follow-up question examples;

[0009] providing the input data and the first prompt information to the first large model, so as to generate follow-up questions for the interview question using the first large model;

[0010] The follow-up questions are issued to the interviewer to guide the interviewer to provide key information required by the follow-up questions.

[0011] The embodiment of the present application also provides a computing device, including: a memory, a processor, and a communication component;

[0012] The memory is used to store one or more computer instructions;

[0013] The processor is coupled to the memory and the communication component, and is configured to execute the one or more computer instructions to execute the aforementioned interview follow-up method.

[0014] 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 follow-up method.

[0015] An embodiment of the present application also provides a computer program product, including a computer program, wherein when the computer program is executed by a processor, the processing area is caused to execute the aforementioned interview follow-up method.

[0016] In an embodiment of the present application, after obtaining the answer content provided by the interviewee to the interview question, the answer content can be detected. If it is detected that the answer content lacks any key information required for the interview question, the interview question and the answer content are provided to the first large model, so that the first large model can generate follow-up questions for the interview question according to the first prompt information, and send the follow-up questions to the interviewee to guide the interviewee to provide the key information required for the follow-up questions. Among them, when the answer content provided by the interviewee is not sufficient to answer the interview intention of the interview question, a new interview question is regenerated for the interview intention, and the new interview question is used to ask the interviewee, so that the interviewee can supplement the information in time, so that the employer can quickly and accurately obtain more effective information for the interviewee, thereby improving the employer's recruitment experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] 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:

[0018] Figure 1 A system architecture diagram is provided for an exemplary embodiment of the present application;

[0019] Figure 2 A flowchart of an interview follow-up method provided for an exemplary embodiment of the present application;

[0020] Figure 3 A schematic diagram of a process for extracting key information provided by another exemplary embodiment of the present application;

[0021] Figure 4 A schematic diagram of a flow chart for generating interview questions provided by another exemplary embodiment of the present application;

[0022] Figure 5 A logic diagram of an interview in an exemplary application scenario provided for another exemplary embodiment of the present application;

[0023] Figure 6 A schematic diagram of the structure of a computing device is provided for yet another exemplary embodiment of the present application. DETAILED DESCRIPTION

[0024] 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.

[0025] At present, only fixed preset interview questions can be shown to the interviewee during the interview process. The answers given by the interviewee to the interview questions may not contain the effective information required by the employer, making it impossible for the employer to evaluate the interviewee's ability, which leads to a poor recruitment experience for the employer. In order to improve this technical problem, the embodiment of the present application provides a solution. The basic idea is: after obtaining the answer content provided by the interviewee to the interview question, the answer content is detected. If it is detected that any key information required for the interview question is missing in the answer content, the interview question and the answer content are provided to the first large model, so that the first large model generates follow-up questions for the interview question according to the first prompt information, and sends follow-up questions to the interviewee to guide the interviewee to provide the key information required for the follow-up questions. Among them, when the answer content provided by the interviewee is not enough to answer the interview intention of the interview question, a new interview question is regenerated for the interview intention, and the new interview question is used to ask the interviewee, so that the interviewee can supplement the information in time, so that the employer can quickly and accurately obtain more effective information for the interviewee, thereby improving the recruitment experience of the employer.

[0026] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. 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 those skilled in the art without creative work are within the scope of protection of this application.

[0027] Figure 1A system architecture diagram is provided for an exemplary embodiment of the present application. The system architecture may include a client 101 and a server 102. A communication connection is established between the client 101 and the server 102. The communication connection method includes but is not limited to wired, wireless communication links or optical fiber cables.

[0028] Among them, the client 101 can be an APP (Application), a web application, a light application (also known as a mini-program, a lightweight application) or a cloud application, etc. The client 101 can be deployed in an electronic device and needs to rely on the electronic device to run. The electronic device may include but is not limited to mobile phones, tablet computers, personal computers, desktop computers, smart speakers, smart watches, etc. The server 102 may include servers that provide various services, such as a server for background training that supports the model used on the client 101, and a server that processes the interactive information sent by the client, etc. The server 102 can be implemented as a single server, a distributed server cluster consisting of multiple servers, a server or cloud server combined with blockchain, etc.

[0029] It should be noted that the interview questioning method provided in the embodiment of the present application is generally executed by the server 102, and the corresponding data processing device is generally arranged in the server 102. However, in other embodiments of the present application, the client 101 may also have similar functions as the server 102, so as to execute the interview questioning method provided in the embodiment of the present application. In other embodiments, the interview questioning method provided in the embodiment of the present application may also be jointly executed by the client 101 and the server 102.

[0030] The implementation details of the technical solution of the embodiment of the present application are elaborated in detail below.

[0031] Figure 2 A flowchart of an interview follow-up method provided by an exemplary embodiment of the present application. The method can be executed by a data processing device, which can be implemented as a combination of software and / or hardware, and the data processing device can be integrated in a computing device. Figure 2 , the method comprising:

[0032] Step 200, after obtaining the answer content provided by the interviewee to any interview question, extracting key information required for the interview question from the answer content;

[0033] Step 201, if it is determined that the answer content lacks any key information required for the interview question, the interview question and the answer content are constructed as input data of the first model;

[0034] Step 202, obtaining preset first prompt information, the first prompt information is used to instruct the first large model to ask follow-up questions according to the interview questions and answers in the input data;

[0035] The first prompt information also includes interview question examples, answer content examples, and corresponding follow-up question examples;

[0036] Step 203, providing the input data and the first prompt information to the first large model, so as to generate follow-up questions for the interview questions using the first large model;

[0037] Step 204, asking follow-up questions to the interviewee to guide the interviewee to provide key information required for the follow-up questions.

[0038] The interview questioning method provided in the embodiment of the present application can be applied to various scenarios where intelligent interviews are required. For example, scenarios of recruiting job seekers, scenarios of providing psychological counseling to visitors, scenarios of conducting admission interviews for students, etc. Of course, these are only exemplary, and the present embodiment is not limited thereto.

[0039] Prior to step 200, natural language processing and machine learning technologies may be used to quickly browse and screen a large number of resumes, screen out interviewees that meet the job requirements, and send interview invitations to the interviewees to interview them using preset interview questions or interview questions generated for the interviewees. Alternatively, in response to the interview invitation initiated by the interviewee, preset interview questions or interview questions generated for the interviewee may be sent to the interviewee to interview the interviewee. The technical details related to generating interview questions for the interviewee will be described in detail below.

[0040] After the interview questions are sent to the interviewee, the interviewee's answer status to the interview questions can be monitored, and different operation actions can be performed based on the monitored answer status. The monitored answer status includes but is not limited to answered status, unanswered status, etc.

[0041] When an unanswered state is detected, the duration of the unanswered state is recorded. If the duration of the unanswered state is detected to exceed the preset waiting threshold, the interview question may be resent to the interviewee, or a follow-up question may be constructed for the interviewee to guide the interviewee to provide key information required for the follow-up question. Optionally, if the duration of the unanswered state of the answer exceeds the preset waiting threshold, the interview question may be resent to the interviewee first. If the number of times the interview question is repeatedly initiated exceeds the preset repetition threshold, a follow-up question may be constructed for the interviewee to guide the interviewee to provide key information required for the follow-up question.

[0042] When the answer status is detected as answered, the answer content provided by the interviewee to the interview question can be obtained, and the key information required for the interview question can be extracted from the answer content. Based on the extracted key information, it can be determined whether the answer content contains all the key information required for the interview question.

[0043] If it is determined that the answer contains all the key information required for the interview question, then determine whether the current interview question is the last interview question. If the current interview question is not the last interview question, send a transition statement to the interviewee to prompt him to move on to a new interview question; if the current interview question is the last interview question, provide a closing statement to the interviewee to end the interview.

[0044] If it is determined that the answer lacks any key information required for the interview question, the interview question and the answer content will be constructed as the input data of the first model. The first model is an artificial intelligence large model used to generate follow-up questions. Artificial intelligence large models refer to machine learning models with ultra-large-scale parameters and complex computing structures. These models can process large-scale data sets and complete various complex tasks, such as natural language processing, image recognition, speech recognition, etc. through deep learning and neural network technology.

[0045] Continue to refer Figure 2 In step 202, a preset first prompt information may be obtained, and the first prompt information is used to instruct the first large model to ask follow-up questions based on the interview questions and answers in the input data. The first prompt information also includes interview question examples, answer content examples, and corresponding follow-up question examples. The first prompt information is a prompt template (promt template) preset for the first large model. The prompt template is a text template used to guide the large model to generate specific outputs or perform specific tasks. The prompt template can guide the model to generate specific outputs that meet the requirements by providing detailed context, instructions or questions to the model.

[0046] For example, the first prompt information preset for the first large model may be:

[0047] "You are an AI interviewer. After the user has answered, you need to ask follow-up questions for unclear or illogical responses. When the user clearly answers the question, output an empty string."

[0048] "Where the question in input is your question, and the answer is the user's answer."

[0049] “Requirements: 1. The questions you output must be reliable, compliant, and objective. Content that violates laws and regulations is not allowed.

[0050] 2. Output of other irrelevant results is not allowed. "

[0051] Please output the corresponding questions according to the following case examples and requirements:

[0052] “case1:'input:{"question":"When can I start work at the earliest?","answer":"I haven't. What are your requirements?"}'

[0053] 'output: Our job requirements are detailed in the recruitment post. When can I start work at the earliest? '"

[0054] "case2:'input:{"question":"What type of truck have you driven before?","answer":"4.2 meters."}'

[0055] 'output: For this question, you can answer flatbed trucks and box trucks. What type of trucks have you driven before? '"

[0056] "case3:'input:{"question":"Have you ever worked in this field before? How many years did you work there?","answer":"Hotel service industry, right?"}'

[0057] 'output: May I ask how many years you have worked in this industry? '"

[0058] "Please complete your inquiry according to the above case examples and requirements. No other irrelevant results are allowed."

[0059] On this basis, the first prompt information and the input data can be provided to the first large model, so as to use the first large model to generate follow-up questions for the interview questions, and the follow-up questions are a new interview question. The follow-up questions are sent to the interviewee to guide the interviewee to provide key information required for the new interview question.

[0060] Accordingly, in this embodiment, after obtaining the answers provided by the interviewee to the interview questions, the answers can be detected. If it is detected that the answers lack any key information required for the interview questions, the interview questions and answers are provided to the first large model, so that the first large model can generate follow-up questions for the interview questions according to the first prompt information, and send follow-up questions to the interviewee to guide the interviewee to provide the key information required for the follow-up questions. Among them, when the answers provided by the interviewee are not sufficient to answer the interview intention of the interview question, new interview questions are regenerated for the interview intention, and the new interview questions are used to ask the interviewee, so that the interviewee can supplement the information in time, so that the employer can quickly and accurately obtain more effective information for the interviewee, thereby improving the employer's recruitment experience.

[0061] In the above or following embodiments, multiple implementations may be used to generate follow-up questions for interview questions. In an optional implementation, the step of generating follow-up questions for interview questions by the first large model may include: performing a semantic extraction operation on the interview question to obtain the semantic information of the interview question, and searching for a target inspection dimension that matches the semantic information of the interview question from the inspection dimension associated with the type of position applied for by the interviewee. Performing a semantic extraction operation on the answer content to obtain the semantic information of the answer content, and searching for the target semantic information that matches the target inspection dimension from the semantic information of the answer content. For the target inspection dimension for which there is no target semantic information that matches it, follow-up questions are generated for the interview question according to the follow-up question generation knowledge learned from the first prompt information. In this way, only generating follow-up questions for interview questions that have not been answered in the interview questions can not only reduce the workload of generating follow-up questions, but also improve the interview efficiency. It can also avoid asking repeated questions to the interviewee, and improve the interview experience of the interviewee.

[0062] The same method can be used to perform semantic extraction on interview questions / answers. Semantic extraction can include the following steps: obtaining text data corresponding to interview questions / answers, performing word segmentation on the text data, and obtaining multiple specific entities contained in the text data as semantic information of the interview questions / answers. Specific entities are meaningful, interview-related concepts and relationships in text data. Specific entities include but are not limited to names, positions, place names, contact information, etc.

[0063] In this embodiment, at least one assessment dimension is pre-configured for each position type. According to the position type applied by the interviewee, the assessment dimension associated with the position type applied by the interviewee can be obtained, and the target assessment dimension adapted to the semantic information of the interview question can be found from the assessment dimensions associated with the position type; the target semantic information adapted to the target assessment dimension can be found from the semantic information of the answer content. If there is any target assessment dimension, the target semantic information adapted to it cannot be found from the semantic information of the answer content, then a follow-up question is generated for the interview question with respect to the target assessment dimension.

[0064] For example, the interview question is "What type of truck have you driven before", and the answer is "4.2 meters", the semantic information extracted from the interview question is "what type of truck", and the semantic information extracted from the answer is "4.2 meters". The assessment dimensions associated with the job type applied for by the interviewee include age, education, work experience, working hours, etc. Based on this, the target assessment dimension found from the assessment dimensions associated with the job type that is adapted to the semantic information of the interview question is "work experience". The semantic information of the answer content is not compatible with the target assessment dimension. Therefore, a follow-up question can be generated for the interview question based on the assessment dimension of "work experience". The generated follow-up question can be "For this question, the answers can be flatbed trucks and box trucks. What type of truck have you driven before?"

[0065] Of course, other implementations may be used to generate follow-up questions. For example, at least one follow-up question may be generated for an interview question by manual configuration, which is not limited in this embodiment.

[0066] After asking the interviewee a follow-up question, the answer content submitted by the interviewee to the follow-up question can be obtained, and key information can be extracted from the answer content corresponding to the follow-up question to detect the answer content corresponding to the follow-up question. If it is determined based on the extracted key information that the answer content corresponding to the follow-up question lacks any key information required for the follow-up question, the follow-up question is reconstructed for the interview question until the answer content to the follow-up question contains all the key information required for the interview question or the number of times the follow-up question is constructed exceeds the preset question threshold.

[0067] Based on this, when all the key information required for the interview question is obtained based on the interview question and / or one or more follow-up questions constructed for the interview question, a new interview question can be displayed to the interviewee or the interview can be ended. In addition, when all the key information required for the interview question is not obtained after the number of follow-up questions constructed for the interview question exceeds the preset follow-up threshold, a new interview question can also be displayed to the interviewee or the interview can be ended. Before displaying a new interview question or ending the interview, it can be determined whether the current interview question is the last interview question. If the current interview question is not the last interview question, a transition statement is sent to the interviewee to prompt the interviewee to move on to a new interview question; if the current interview question is the last interview question, a closing statement is provided to the interviewee to end the interview.

[0068] As mentioned above, for any target assessment dimension corresponding to the interview question, if the target semantic information that matches the target assessment dimension cannot be found from the semantic information of the answer content, a follow-up question is generated for the interview question for the target assessment dimension. That is, when generating follow-up questions for the interview question, only follow-up questions are asked for the unanswered target assessment dimensions, and there is no need to ask follow-up questions for all target assessment dimensions corresponding to the interview question.

[0069] Accordingly, when regenerating follow-up questions for interview questions, the interview questions sent to the interviewee and the target assessment dimensions that have not received a response after the previous follow-up questions are obtained, and only these target assessment dimensions that have not received a response need to be followed up. This can not only reduce the burden of generating follow-up questions and improve interview efficiency, but also avoid repeatedly asking the same questions to the interviewee, thereby avoiding causing the interviewee's disgust and effectively improving the interview experience of the interviewee.

[0070] Optionally, during the interview process, the number of follow-up questions constructed for any interview question can be recorded in real time. For any interview question, after determining the number of follow-up questions constructed for the interview question, the interviewee's evaluation level can be determined based on the mapping relationship between the preset number of follow-up questions and the evaluation level, and the interviewee's evaluation level can be saved. The evaluation level corresponding to the interview question can be used to reflect the interviewee's assessment results on the assessment dimension corresponding to the interview question, which can provide an important reference for the employer to make recruitment decisions, save the employer's decision-making time, and thus improve the efficiency of the interview.

[0071] Among them, the mapping relationship between the number of follow-up questions and the evaluation level is relative to the interview questions. Each interview question is associated with a corresponding mapping relationship. Each interview question corresponds to multiple evaluation levels. The evaluation levels corresponding to different interview questions can be used to characterize the interview results of the interviewer in different examination dimensions. Therefore, the evaluation level for the interview question and the key information extracted for the interview question can be saved in the same location. For example, the evaluation level and key information can be saved as the interview data corresponding to the interviewer in the interview database. The interview database stores the data generated during the interview according to the job type. Each job type contains the interview data corresponding to multiple interviewers. The interview data includes but is not limited to interview questions, answer content, key information, evaluation level and resume data.

[0072] Based on this, if the interviewee's evaluation level is lower than the specified level, the interviewee is provided with an information collection form corresponding to the interview question, and the information collection form is used to collect the key information required for the interview question that was not obtained. Extract key information from the information collection form and save the extracted key information. For any interview question, if the interviewee's evaluation level is determined to be lower than the specified level, it means that the interviewee is very likely to not understand the interview question or the intention of asking follow-up questions. At this time, an information collection form for the interview question can be sent to the interviewer. The information collection form may contain unanswered target assessment dimensions and the input boxes corresponding to each target assessment dimension, so that the interviewee can understand the intention of the question more intuitively and guide the interviewee to fill in the key information required for the interview question in the information collection form.

[0073] In this embodiment, the concept of an information collection form is introduced as a fallback strategy in the interaction process with the interviewee. The form shows the interviewee the information that the employer wants to know in a simpler and more intuitive way, which helps the interviewee understand the employer's intentions and make accurate responses, thereby further improving the effectiveness of the interview.

[0074] In addition, in this embodiment, the first large model can also be fine-tuned in real time: historical interview records under any job type can be obtained, and for any interview question in the historical interview record, the answer content to the interview question and the follow-up question constructed for the interview question are obtained. The interview question, the answer content to the interview question, the follow-up question constructed for the interview question, and the first prompt information are constructed as a data sample, and the data sample is input into the first large model, and the first large model is trained to achieve fine-tuning of the first large model. In the process of fine-tuning the first large model, the first prompt information can be flexibly adjusted according to the performance of the first large model, data feedback, etc., so as to achieve iterative optimization of the first large model, which can effectively ensure the accuracy of the first large model, thereby effectively improving the accuracy of the generated follow-up questions.

[0075] In the above or below embodiments, multiple implementations may be used to extract key information from the answer content.

[0076] Figure 3 A schematic diagram of a process for extracting key information provided by another exemplary embodiment of the present application. Figure 3 , the answers provided by the interviewee to the interview questions can be obtained, the type of position applied for by the interviewee can be determined, and the interview questions, the position type and the answers can be constructed as input data for the second largest model. The preset second prompt information is obtained, and the second prompt information and the input data are provided to the second largest model, so as to use the second largest model to extract key information from the answers provided by the interviewee.

[0077] Among them, the second prompt information is used to instruct the second largest model to extract key information from the answer content according to the interview questions and job types in the input data, and the second prompt information also includes interview question examples, job type examples, answer content examples and corresponding key information examples. The second largest model is an artificial intelligence large model for extracting key information from the answer content. The second prompt information is a prompt template (promt template) preset for the second largest model, which is used to instruct the second largest model to extract key information from the answer content according to the interview questions, job types and answer content.

[0078] Of course, other implementation methods may also be used to extract key information from the answer content. For example, the key information may be manually extracted by the staff of the recruiting party, and this embodiment does not limit this.

[0079] In this embodiment, by detecting the content of the answer, it is possible to determine in real time whether the key information required for the interview questions is obtained, and if the required key information is not obtained, timely follow-up questions can be asked to the interviewee to effectively ensure the accuracy and completeness of the collected key information, facilitate rapid screening of the interviewees, and eliminate the need to contact the interviewees again, thereby effectively improving the efficiency of the interview.

[0080] In the above or below embodiments, multiple implementations may be used to generate interview questions for the interviewer.

[0081] Figure 4 A flowchart of generating interview questions is provided for another exemplary embodiment of the present application. Figure 4 Before any interview question is sent to the interviewee, the interview request initiated by the interviewee may be responded to to determine the type of position applied for by the interviewee. The assessment dimensions required for the position type are obtained, the resume data and historical interview data of the interviewee are obtained, and the assessment dimensions, resume data and historical interview data are constructed as input data of the third model. The preset third prompt information is obtained, and the input data and the third prompt information are provided to the third model, so as to generate interview questions for the interview request using the third model.

[0082] Among them, the third prompt information is used to instruct the third model to generate interview questions adapted to the inspection dimension based on the resume data and historical interview data in the input data. The third prompt information also includes resume data examples, historical interview data examples, inspection dimension examples, and corresponding interview question examples. The third model is an artificial intelligence large model for generating interview questions for interviewers. The third prompt information is a preset prompt template (promt template) for the third model, which is used to instruct the third model to generate interview questions adapted to the inspection dimension based on resume data and historical interview data.

[0083] Of course, other implementations may also be used to generate interview questions. For example, interview questions may be generated for the interviewer by manual configuration, which is not limited in this embodiment.

[0084] In this embodiment, the assessment dimensions that need to be paid attention to under the job type are obtained according to the job type applied for by the interviewee, and the obtained assessment dimensions are the assessment dimensions for the interviewee. Based on the assessment dimensions for the interviewee, the interviewee's resume data, and the interviewee's historical interview data, at least one interview question adapted to the assessment dimensions is generated for the interviewee in real time, so as to guide the interviewee to provide key information under each assessment dimension corresponding to the job type through the interview questions. In this way, according to the actual situation of each interviewee, corresponding interview questions are customized for each interviewee, which can effectively reduce the occurrence of mismatches between interview questions and the interviewee's actual experience or skills, so that the interview questions provided to the interviewee are more in line with the interviewee's actual situation, thereby effectively improving the interview experience of the interviewee.

[0085] It should be noted that, in this embodiment, the form of the interview with the interviewee includes but is not limited to text form and voice form. If the interview process is in voice form, the interview questions, follow-up questions, and follow-up words can be converted from text form to voice form after the interview questions and follow-up questions are generated, or before the follow-up words are sent, and the converted voice is used to interact with the interviewee. Among them, the text form can be converted into voice form by using text-to-speech (TTS) technology. The form conversion of interview questions and follow-up questions is real-time conversion, and the interview questions and follow-up questions are converted into voice in real time, and sent directly to the interviewee after conversion; the form conversion of the follow-up words is cache conversion, and the follow-up words can be converted into voice in advance and continue to be cached so that they can be directly called when needed.

[0086] Figure 5 This is a schematic diagram of an interview logic in an exemplary application scenario provided by another exemplary embodiment of the present application. Figure 5 As shown, the interview logic in this exemplary application scenario includes the following steps:

[0087] Step 501, displaying multiple recruitment information to the interviewer;

[0088] Step 502, in response to an interview request initiated by the interviewer for the target recruitment information, the position type associated with the target recruitment information may be used as the position type applied for by the interviewer;

[0089] Step 503, obtaining the assessment dimensions required for the position type, the interviewee's resume data, and the interviewee's historical interview data;

[0090] Step 504, inputting the inspection dimension, the resume data and the historical interview data into the third model, so that the third model generates interview questions adapted to the inspection dimension according to the resume data and the historical interview data and the third prompt information;

[0091] Step 505, sending the interview question to the interviewee, and obtaining the answer provided by the interviewee to the interview question; wherein, if the interview question is the first interview question sent to the interviewee, the opening remarks of this interview may be shown to the interviewee before sending the first interview question;

[0092] Step 506, inputting the interview questions, answer content and job type into the second largest model, so that the second largest model extracts key information from the answer content according to the interview questions and job type and the second prompt information;

[0093] Step 507, judging whether the answer lacks any key information required for the interview question based on the extracted key information;

[0094] If it is determined in step 507 that the answer contains all the key information required for the interview question, then jump to step 508 to determine whether the current interview question is the last interview question;

[0095] If the current interview question is not the last interview question, then jump to step 509 and send the target follow-up words to the interviewee to prompt him to enter a new interview question;

[0096] If the current interview question is the last interview question, then jump to step 510 to provide the interviewee with a closing statement to end the interview;

[0097] If it is determined in step 507 that the answer lacks any key information required for the interview question, then jump to step 511 to determine whether the key information extraction time exceeds a preset time threshold;

[0098] If the preset time threshold is exceeded, the process jumps back to step 508 .

[0099] If the preset time threshold is not exceeded, jump to step 512 and input the interview questions and answers into the first large model so that the first large model can generate follow-up questions for the interview questions according to the interview questions and answers and the first prompt information.

[0100] Step 513, sending follow-up questions to the interviewee, and obtaining answers provided by the interviewee to the follow-up questions;

[0101] Step 514, determining whether the answer provided for the follow-up question lacks any key information required for the follow-up question;

[0102] If the answer provided for the follow-up question contains all the key information required for the follow-up question, then jump to step 508;

[0103] If the answer provided for the follow-up question lacks any key information required for the follow-up question, then jump to step 515 to determine whether the number of follow-up questions constructed for the interview question exceeds a preset follow-up threshold;

[0104] If it exceeds the preset inquiry threshold, jump to step 508;

[0105] If the preset inquiry threshold is not exceeded, the process jumps to step 512 .

[0106] Accordingly, in the embodiments of the present application, interview questions can be customized for each interviewee based on the interviewee's actual situation, and the answers returned by the interviewee can be detected in real time, so that when the answers lack any key information required for the interview questions, follow-up questions can be generated in time for the interview questions to guide the interviewee to provide the missing key information. This can provide the interviewee with a more realistic interview experience and a faster and more accurate understanding of the interviewee's abilities; and, corresponding response measures are provided for various situations that may arise during the interview process, so that the interview process can proceed smoothly, thereby effectively improving the interview experience and efficiency of both parties.

[0107] 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 200 to 203 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.

[0108] 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.

[0109] Figure 6 A schematic diagram of a computing device provided as another exemplary embodiment of the present application is shown in FIG. Figure 6 As shown, the computing device includes: a memory 60 , a processor 61 , and a communication component 62 .

[0110] The processor 61 is coupled to the memory 60 and the communication component 62, and is used to execute the computer program in the memory 60, so as to:

[0111] After obtaining the interviewee's answer to any interview question, extract the key information required for the interview question from the answer;

[0112] If it is determined that the answer content lacks any key information required by the interview question, the interview question and the answer content are constructed as the input data of the first model;

[0113] Obtaining preset first prompt information, the first prompt information is used to instruct the first large model to ask follow-up questions based on the interview questions and answers in the input data, and the first prompt information also includes interview question examples, answer content examples, and corresponding follow-up question examples;

[0114] providing the input data and the first prompt information to the first large model to generate follow-up questions for the interview question using the first large model;

[0115] Ask the interviewee follow-up questions to elicit the interviewee's key information needed for the follow-up questions.

[0116] In an optional embodiment, in the process of using the first large model to generate follow-up questions for the interview question, the processor 61 is further used to:

[0117] Use the first model to perform semantic extraction operations on interview questions to obtain semantic information of interview questions;

[0118] Using the first model, find the target assessment dimension that matches the semantic information of the interview question from the assessment dimensions associated with the position type applied for by the interviewee;

[0119] The first model is used to perform semantic extraction operations on the answer content to obtain the semantic information of the answer content;

[0120] Using the first model, search for target semantic information that matches the target inspection dimension from the semantic information of the answer content;

[0121] For target examination dimensions for which there is no target semantic information adapted thereto, the first large model is used to generate knowledge for follow-up questions learned from the first prompt information, so as to generate follow-up questions for the interview questions.

[0122] In an optional embodiment, the processor 61 is further configured to:

[0123] Get historical interview records for any job type;

[0124] For any interview question in the historical interview records, obtain the answer content and the follow-up questions constructed for the interview question;

[0125] The interview questions, answers to the interview questions, first prompt information, and follow-up questions constructed for the interview questions are constructed as data samples;

[0126] The data samples are input into the first large model to train the first large model.

[0127] In an optional embodiment, the processor 61 is further configured to:

[0128] After asking the interviewee a follow-up question, obtain the interviewee's answer to the follow-up question;

[0129] Extract key information from the answers to follow-up questions;

[0130] If it is determined based on the extracted key information that the answer to the follow-up question lacks any key information required for the follow-up question, the follow-up question will be reconstructed for the interview question until the answer to the follow-up question contains all the key information required for the interview question or the number of times the follow-up question is constructed exceeds the preset follow-up threshold.

[0131] In an optional embodiment, the processor 61 is further configured to:

[0132] After all the key information required for the interview question has been obtained based on the interview question and / or one or more follow-up questions constructed for the interview question, present a new interview question to the interviewee or end the interview; or,

[0133] If all the key information required for the interview question is not obtained after the number of follow-up questions constructed for the interview question exceeds the preset follow-up threshold, a new interview question is presented to the interviewee or the interview is ended.

[0134] In an optional embodiment, the processor 61 is further configured to:

[0135] For any interview question, determine the number of follow-up questions constructed for the interview question;

[0136] Determine the interviewee's evaluation level based on the mapping relationship between the preset number of follow-up questions and the evaluation level;

[0137] Save the interviewee's evaluation rating.

[0138] In an optional embodiment, the processor 61 is further configured to:

[0139] If the interviewee's evaluation level is lower than the specified level, the interviewee will be provided with an information collection form corresponding to the interview questions, which is used to collect key information required for the interview questions that were not obtained;

[0140] Extract key information from the information collection form;

[0141] In an optional embodiment, in the process of extracting the key information required for the interview question from the answer content provided by the interviewee, the processor 61 is further used to:

[0142] Obtain the answers provided by the interviewee to the interview questions;

[0143] Determine the type of position the interviewee is applying for;

[0144] Interview questions, job types, and answer content are constructed as input data for the second largest model;

[0145] Obtaining preset second prompt information, the second prompt information is used to instruct the second largest model to extract key information from the answer content according to the interview questions and job types in the input data, the second prompt information also including interview question examples, job type examples, answer content examples and corresponding key information examples;

[0146] The second prompt information and the input data are provided to the second large model, so as to extract key information from the answer content provided by the interviewee using the second large model. In an optional embodiment, the processor 61 is also used for:

[0147] In response to an interview request initiated by an interviewer, determining the type of position applied for by the interviewer;

[0148] Obtain the assessment dimensions that need to be focused on under the position type;

[0149] Obtaining resume data and historical interview data of the interviewee;

[0150] The inspection dimension, the resume data and the historical interview data are constructed as input data of the third model;

[0151] Obtaining preset third prompt information, the third prompt information is used to instruct the third model to generate interview questions adapted to the examination dimension according to the resume data and historical interview data in the input data, the third prompt information also including resume data examples, historical interview data examples, examination dimension examples, and corresponding interview question examples;

[0152] The input data and the third prompt information are provided to the third model to generate interview questions for the interview request using the third model.

[0153] Further, if Figure 6As shown, the computing device also includes: a display 63, a power component 64, an audio component 65 and other components. Figure 6 Only some components are shown schematically, and it does not mean that the computing device only includes Figure 6 Components shown.

[0154] It is worth noting that the technical details in the above-mentioned embodiments of the computing device can refer to the relevant descriptions of the actions of the computing device in the above-mentioned embodiments of the interview questioning method. In order to save space, they will not be repeated here, but this should not cause any loss of the protection scope of this application.

[0155] The present application also provides a computer program product, including a computer program. When the computer program is executed by a processor, the steps in the above interview follow-up method can be implemented. It should be noted that the technical solution of the computer program and the technical solution of the above interview follow-up method belong to the same concept, and the details of the technical solution of the computer program that are not described in detail can be found in the description of the technical solution of the above interview follow-up method.

[0156] 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 a computing device in the above method embodiment.

[0157] Above Figure 6 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.

[0158] Above Figure 6The 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.

[0159] Above Figure 6 The display in the embodiment includes a screen, which 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.

[0160] Above Figure 6 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.

[0161] Above Figure 6 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.

[0162] 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.

[0163] 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.

[0164] 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.

[0165] 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 A step that specifies a function in one or more boxes.

[0166] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0167] 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.

[0168] 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.

[0169] 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.

[0170] 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 questioning method, characterized in that: include: After obtaining the answer content provided by the interviewee to any interview question, extracting key information required for the interview question from the answer content; If it is determined that the answer content lacks any key information required by the interview question, constructing the interview question and the answer content as input data of the first model; Obtaining preset first prompt information, wherein the first prompt information is used to instruct the first large model to ask follow-up questions based on the interview questions and answers in the input data, wherein the first prompt information also includes interview question examples, answer content examples, and corresponding follow-up question examples; providing the input data and the first prompt information to the first large model, so as to generate follow-up questions for the interview question using the first large model; The follow-up questions are issued to the interviewer to guide the interviewer to provide key information required by the follow-up questions.

2. The method according to claim 1, characterized in that The first model is used to generate follow-up questions for the interview questions, including: Using the first large model to perform a semantic extraction operation on the interview question to obtain semantic information of the interview question; Using the first model, from the assessment dimensions associated with the position type applied for by the interviewee, a target assessment dimension that is adapted to the semantic information of the interview question is searched; Using the first large model to perform a semantic extraction operation on the answer content to obtain semantic information of the answer content; Using the first large model to search for target semantic information that matches the target inspection dimension from the semantic information of the answer content; For target examination dimensions for which there is no target semantic information adapted thereto, the first large model is used to generate knowledge for follow-up questions learned from the first prompt information, so as to generate follow-up questions for the interview questions.

3. The method according to claim 1, characterized in that Also includes: Get historical interview records for any job type; For any interview question in the historical interview record, obtaining the answer content to the interview question and the follow-up question constructed for the interview question; The interview question, the answer to the interview question, the first prompt information, and the follow-up question constructed for the interview question are constructed into a data sample; The data samples are input into the first large model to train the first large model.

4. The method according to claim 1, characterized in that Also includes: After sending the follow-up question to the interviewee, obtaining the answer content submitted by the interviewee to the follow-up question; Extract key information from the answer content corresponding to the follow-up question; If it is determined based on the extracted key information that the answer content corresponding to the follow-up question lacks any key information required for the follow-up question, then a new follow-up question is generated for the interview question until the answer content to the follow-up question contains all the key information required for the interview question or the number of times the follow-up question is constructed exceeds a preset follow-up threshold.

5. The method according to claim 1, characterized in that Also includes: When all key information required for the interview question is obtained based on the interview question and / or one or more follow-up questions constructed for the interview question, present a new interview question to the interviewee or end the interview; or, When all key information required for the interview question is still not obtained after the number of follow-up questions constructed for the interview question exceeds a preset follow-up threshold, a new interview question is presented to the interviewee or the interview is terminated.

6. The method according to claim 1, characterized in that Also includes: For any interview question, determine the number of follow-up questions constructed for the interview question; Determining the evaluation level of the interviewee based on a mapping relationship between the number of preset follow-up questions and the evaluation level; The evaluation rating of the interviewee is saved.

7. The method according to claim 6, characterized in that Also includes: If the interviewee's evaluation level is lower than the specified level, providing the interviewee with an information collection form corresponding to the interview question, wherein the information collection form is used to collect key information required for the interview question that has not been obtained; extracting key information from the information collection form; Save the extracted key information.

8. The method according to claim 1, characterized in that: Extracting key information required for the interview question from the answer provided by the interviewee includes: Obtaining answers provided by the interviewee to the interview questions; Determine the type of position for which the interviewee is applying; The interview questions, the job types and the answer contents are constructed as input data of the second largest model; Obtaining preset second prompt information, where the second prompt information is used to instruct the second large model to extract key information from the answer content according to the interview questions and job types in the input data, and the second prompt information also includes interview question examples, job type examples, answer content examples, and corresponding key information examples; The second prompt information and the input data are provided to the second large model, so as to utilize the second large model to extract key information from the answer content provided by the interviewee.

9. The method according to claim 1, characterized in that: Also includes: In response to an interview request initiated by an interviewer, determining the type of position applied for by the interviewer; Obtain the assessment dimensions that need to be focused on under the position type; Obtaining resume data and historical interview data of the interviewee; The inspection dimension, the resume data and the historical interview data are constructed as input data of the third model; Obtaining preset third prompt information, wherein the third prompt information is used to instruct the third model to generate interview questions adapted to the examination dimension according to the resume data and the historical interview data in the input data, wherein the third prompt information also includes resume data examples, historical interview data examples, examination dimension examples, and corresponding interview question examples; The input data and the third prompt information are provided to the third large model to generate interview questions for the interview request using the third large model.

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 follow-up method described in 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 questioning method described in any one of claims 1-9.

12. A computer program product, characterized in that It comprises a computer program, wherein when the computer program is executed by a processor, the processor is caused to execute the interview questioning method according to any one of claims 1 to 9.

Images