Large model-based interview target dynamic planning method, computer device and medium

Abstract

This invention relates to a dynamic programming method for interview objectives based on a large model, as well as computer equipment and media. The method includes acquiring basic information about the interviewee; executing a dialogue interview step, guiding the interviewee to output feedback information based on the basic information and a preset structured feature library; performing verification processing on the feedback information to obtain labeled feature dimensions; retrieving the structured feature library and comparing it with the labeled feature dimensions to obtain the feature dimensions to be followed up; and returning to execute the dialogue interview step based on the feature dimensions to be followed up until all feature dimensions in the structured feature library have been accessed and labeled. The beneficial effects are: dialogue with the interviewee based on basic information and a structured feature library guides the interviewee to output feedback information, avoiding the limitations of a fixed question sequence and making the dialogue more relevant to the interviewee's background; verification processing improves the quality of interview data; and comparison processing identifies the dimensions to be followed up, avoiding omissions and repetitions.

Description

Technical Field

[0001] This invention relates to the field of large model technology, and in particular to a dynamic programming method for interview objectives based on large models, computer equipment, and media. Background Technology

[0002] Currently, most event interviews use a fixed sequence of questions, making it impossible to customize interview questions based on the interviewee's basic information, resulting in insufficient question targeting; there is a lack of an effective verification mechanism for feedback information, making it difficult to ensure that the feedback information fits the investigation needs, and easily generating invalid interview data; without establishing a structured feature library and comparison logic with the information already obtained, it is impossible to accurately identify the dimensions to be questioned, which easily leads to omissions or repetitions in the investigation. Summary of the Invention

[0003] The technical problem to be solved by the embodiments of the present invention is to provide a dynamic programming method, computer equipment, and medium for interview targets based on a large model. This method uses acquired basic information and a pre-set structured feature library to engage in dialogue with the interviewee, thereby guiding the interviewee to output feedback information. This avoids the limitations of a fixed question sequence and makes the dialogue more relevant to the interviewee's background. Verification processing is used to validate the interviewee's feedback information, filtering out irrelevant content and improving the quality of interview data. A comparison processing step is performed based on the structured feature library and the labeled feature dimensions to identify dimensions to be followed up. Simultaneously, based on these dimensions, the dialogue interview steps are repeated until all feature dimensions of the structured feature library are covered, ensuring the efficiency of the event interview and avoiding omissions and repetitions.

[0004] To address the aforementioned technical problems, this invention provides a dynamic programming method for interview objectives based on a large model, comprising: Obtain basic information about the interviewees; The dialogue interview step is executed, and the interviewee is guided to output feedback information based on the basic information and the preset structured feature library; Receive the feedback information and perform verification processing on the feedback information to obtain the labeled feature dimensions; In response to the feature dimension query command, the structured feature library is retrieved and compared with the labeled feature dimensions to obtain the feature dimension to be explored. Based on the feature dimensions to be followed up, the dialogue interview steps are executed again until all feature dimensions in the structured feature library have been accessed and marked.

[0005] In one exemplary embodiment, obtaining the basic information of the interviewee includes, In response to the initialization command, the model outputs its own role information and guides the interviewee to output a consultation request; If the inquiry request has been received, the system will respond to the inquiry request and then guide the interviewee to introduce themselves. If not, the system will directly guide the interviewee to introduce themselves to obtain the basic information.

[0006] In one exemplary embodiment, the dialogue interview step includes asking the interviewee questions based on the interview questions, guiding the interviewee to output the feedback information; the interview questions are formulated in combination with the interview objectives and the basic information.

[0007] In one exemplary embodiment, the verification process includes, Determine whether the feedback information matches the interview question. If not, modify the interview question and ask the interviewee the modified interview question until the feedback information matches the interview question. If yes, identify and mark the feature dimension corresponding to the feedback information.

[0008] In one exemplary embodiment, the workaround for the access problem includes, The feedback information is analyzed to extract the theme and scene elements, and the access problem is analyzed to extract the missing elements. Based on the theme scene elements and the missing elements, a response text corresponding to the feedback information is generated; the response text includes a first guidance text that matches the theme scene elements and a second guidance text that matches the missing elements.

[0009] In an exemplary embodiment, the comparison process includes removing the labeled feature dimension from the structured feature library and selecting one of the remaining feature dimensions from the structured feature library to obtain the feature dimension to be investigated.

[0010] In an exemplary embodiment, after returning and executing the dialogue interview steps based on the feature dimensions to be questioned until all feature dimensions in the structured feature library have been accessed and marked, the large model-based dynamic programming method for interview objectives further includes, Abnormal text in the feedback information is extracted based on the exaggerated statement recognition algorithm; The interviewee is asked questions based on the abnormal text, and it is determined whether the interviewee answers directly. If yes, the interview ends and an evaluation report is generated; if no, questions are asked again based on the abnormal text until the interviewee answers directly, at which point the interview ends and the evaluation report is generated.

[0011] In an exemplary embodiment, the evaluation report includes scores for all feature dimensions in the structured feature library; the structured feature library includes background feature dimensions, target feature dimensions, action feature dimensions, and outcome feature dimensions; the basic information includes the interviewee's work background information and personality information.

[0012] Accordingly, this application also relates to a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the dynamic programming method for interview objectives based on a large model.

[0013] Accordingly, this application also relates to a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the dynamic programming method for interview objectives based on a large model.

[0014] Implementing this invention has the following beneficial effects: Based on the acquired basic information and a pre-set structured feature library, the system engages in dialogue with interviewees, guiding them to provide feedback. This avoids the limitations of a fixed question sequence and makes the dialogue more relevant to the interviewees' backgrounds. Verification processing filters out irrelevant content, improving the quality of the interview data. A comparison process is performed between the structured feature library and the labeled feature dimensions to identify dimensions to be followed up. Simultaneously, the system returns to the dialogue interview steps based on these dimensions until all feature dimensions in the structured feature library are covered, ensuring efficient event interviews and preventing omissions and duplications.

[0015] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the steps of a dynamic programming method for interview objectives based on a large model, according to one embodiment of the present invention. Figure 2 This is a schematic diagram of step S100 of one embodiment of the present invention; Figure 3 This is another embodiment of the present invention: a dynamic programming method for interview objectives based on a large model. Figure 4 This is a flowchart of a dynamic programming method for interview objectives based on a large model, according to one embodiment of the present invention. Figure 5 This is a schematic diagram of the hardware structure of the computer device of the present invention. Detailed Implementation

[0017] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Many specific details are set forth in the following description to provide a thorough understanding of the present invention. However, the present invention can be practiced in many other ways different from those described herein, and those skilled in the art can make similar modifications without departing from the spirit of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0018] It should be noted that when a component is said to be "fixed to" another component, it can be directly attached to the other component or there may be an intervening component. When a component is said to be "connected to" another component, it can be directly connected to the other component or there may be an intervening component. The terms "vertical," "horizontal," "left," "right," and similar expressions used in this document are for illustrative purposes only.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0020] Current event interview techniques often employ fixed question sequences, which can easily lead to repetitive questioning, resulting in low interview efficiency and a rigid experience. Assessments rely on subjective judgment and lack standardized, structured scoring criteria, leading to significant variations in evaluation results across different scenarios and low credibility. They are also prone to collecting vague or off-target information, affecting assessment accuracy. Furthermore, the lack of dynamic control logic covering all dimensions often results in the omission of core competency dimensions, leading to incomplete and one-sided assessments.

[0021] This application provides a dynamic programming method for interview objectives based on a large model, referring to... Figure 1 ,include, Step S100: Obtain basic information about the interviewees; Step S200: Perform the dialogue interview step, and guide the interviewee to output feedback information based on basic information and a preset structured feature library; Step S300: Receive feedback information and perform verification processing on the feedback information to obtain the labeled feature dimensions; Step S400: In response to the feature dimension query command, the structured feature library is retrieved and compared with the labeled feature dimensions to obtain the feature dimensions to be explored. Step S500: Based on the feature dimensions to be followed up, return to the execution of the dialogue interview step, that is, return to the execution of step S200, until all feature dimensions in the structured feature library have been accessed and marked.

[0022] In an exemplary embodiment, the structured feature library includes background feature dimension S, target feature dimension T, action feature dimension A, and outcome feature dimension R; the basic information includes the interviewee's work background information and personality information.

[0023] Specifically, S (Situation) focuses on the background and environment in which the event occurs, and it is necessary to clarify "under what circumstances it occurs", such as "the background of the project facing the departure of core members and a tight delivery cycle"; T (Task) focuses on the specific goals and responsibilities in the event, and needs to clearly define the "tasks to be completed", such as "being responsible for reorganizing the project team, developing a new delivery plan, and ensuring on-time launch"; A (Action) focuses on the specific measures taken and individual contributions to achieve the goal, and needs to specify "what proactive actions were taken", such as "coordinating 3 cross-departmental personnel to fill in, optimizing key process links, and synchronizing progress and risks daily"; R (Result) focuses on the final results and impact of the action, and needs to specify "what results were ultimately achieved", such as "the project was delivered 2 days ahead of schedule, costs were reduced by 15%, and customer satisfaction reached 92%".

[0024] The "S / T / A / R" dimensions of the structured feature library correspond to the core logic of competency assessment (a complete behavioral chain from scenario background to final result), ensuring that questions always focus on the key aspects of competency assessment; basic information includes work background and personality information, providing anchors for question customization, making questions strongly correlated with the interviewee's real experiences, and improving the effectiveness of feedback information.

[0025] The “S / T / A / R” dimensions constitute a complete closed loop of capability performance (background constraints → goal decomposition → action implementation → result presentation), with each dimension corresponding to different core elements of capability.

[0026] For example, refer to Figure 2 In step S100, obtaining basic information about the interviewee includes: In step S101, the dialogue model responds to the initialization command, outputs its own role information, and guides the interviewee to output a consultation request. Step S102: Check if an inquiry request has been received. If yes, respond to the inquiry request and guide the interviewee to introduce themselves. If no, directly guide the interviewee to introduce themselves to obtain basic information.

[0027] In step S101, the dialogue model responds to the initialization command, outputs its own role information, and guides the interviewee to output a consultation request. The model's own role information can be the interview identity played by the dialogue model, the interview responsibilities, and the interview interaction rules.

[0028] For example, in step S101, the dialogue model outputs, "Hello! I am the intelligent interview assistant for this competency assessment, responsible for conducting a dialogue with you regarding 'project management skills.' You can consult me ​​about the interview process and assessment dimensions, and I will answer your questions." In this embodiment, the role information needs to match the professional speaking style of a real assessor. The interviewer's role in the dialogue model is that of an assessor, and the interview responsibility is to conduct an interview focusing on 'project management skills.' The interview interaction rules are open-ended interactive questions. This avoids mechanical instructions while implicitly guiding the interview objective and directing the interviewee to focus on the interview topic.

[0029] For example, in step S102, if the interviewee sends a consultation request to the dialogue model, and the dialogue model detects the consultation request, the dialogue model will prioritize responding to the consultation question. At this time, the dialogue model needs to call the preset interview consultation knowledge base (the base stores common consultation questions and standard answers, such as "This assessment is expected to last 20-30 minutes" and "The assessment dimensions include background feature dimension, target feature dimension, action feature dimension, and result feature dimension"). After responding, a "natural transition guidance" is used to obtain the interviewee's own information, such as: "The above is the answer to your question. Next, in order to better understand your experience, please briefly introduce your previous job positions, years of work experience, and your communication style in your work (such as being more proactive in coordination / meticulous in implementation)."

[0030] For example, in step S100, the dialogue model will first introduce itself, and then ask the interviewee if they have any questions. If so, it will answer the interviewee's questions first and then let the interviewee introduce themselves. If not, it will directly guide the interviewee to introduce themselves.

[0031] In an exemplary embodiment, step S200, the dialogue interview step, includes asking questions to the interviewee based on the interview questions, and guiding the interviewee to output feedback information; the interview questions are formulated in combination with the interview objectives and basic information.

[0032] For example, in step S200, questions are asked to the interviewee based on the interview questions. The interview questions are formulated by combining the preset interview objectives and basic information. The macro interview objectives (such as "project management ability") are broken down into sub-dimensions in the corresponding structured feature library (such as background S, task T, action A, and result R), and the assessment requirements for each sub-dimension are clarified (such as the background S dimension requiring an assessment of "the background constraints at the time of project initiation"). For example, for an interviewee with "3 years of experience in To B products", the interview question for the background S dimension can be customized as "Please explain the core business constraints you faced during the initiation phase of the To B product project you led (such as changes in customer requirements, insufficient resources, etc.)", rather than the general "Please explain the initiation background of the project you participated in", ensuring that the questions accurately reach the interviewee's real experience and improve the effectiveness of the feedback information.

[0033] For example, the dialogue model can take the form of "question + supplementary explanation", such as: "Please explain the specific way you break down the task objectives in your project (supplementary explanation: you can explain how you break down the overall objectives into actionable tasks for team members in the context of a specific project)". The supplementary explanation is used to clarify the granularity requirements of the feedback information and avoid the interviewee outputting ambiguous content due to misunderstanding.

[0034] In one exemplary embodiment, the verification process in step S300 includes, Determine whether the feedback information matches the interview question. If not, modify the interview question and ask the interviewee the modified interview question until the feedback information matches the interview question. If it does, identify and mark the feature dimensions corresponding to the feedback information to obtain the marked feature dimensions.

[0035] For example, the dialogue model performs semantic analysis on the interviewee's answers, then compares the analysis results with the semantics of the interview questions. When the analysis results of the interviewee's answers and the semantic analysis results of the interview questions are on the same topic, the feedback information matches the interview questions. The aforementioned alternative interview questions include... First, analyze the feedback information to extract the thematic scene elements from the feedback information, and analyze the access questions to extract the missing elements in the access questions; Secondly, based on the theme scene elements and missing elements, a response text corresponding to the feedback information is generated; the response text includes a first guidance text that matches the theme scene elements and a second guidance text that matches the missing elements.

[0036] Specifically, the interview question is: Please describe a project you felt most accomplished about in the last six months. If the interviewee answers: "I had a cup of milk tea today." This answer clearly doesn't fit the above interview question. At this point, the dialogue model will change the question. The changed question still revolves around the original question, but it will respond to the interviewee's previous answer in a friendly way. For example: If the interviewee answers: "I had a cup of milk tea today," you can ask: "You had milk tea today? Drinking milk tea must have been a pleasant experience. Could you please recall the project you felt most accomplished about in the last six months?" Extract thematic scene elements that reflect the emotions or daily scenarios from the feedback of the interviewees (for example, "drinking milk tea" in the feedback corresponds to "relaxed scenario" and "pleasant emotion"); compare it with the original interview questions' objectives (for example, to understand "projects that have brought a sense of accomplishment in the past six months"), and identify the missing elements in the feedback that were not covered by the interview questions (for example, "time" and "project accomplishment" were not mentioned). These are the "information gaps" that need to be filled.

[0037] The first guiding text is generated based on the theme and scenario elements, using phrases that fit the interviewee's emotions or the context (such as "Drinking milk tea today, drinking milk tea should make you feel happy") to reduce resistance to responses; the second guiding text is generated based on missing elements, explicitly reminding the interviewee to fill in the missing information (such as "Could you please recall what your most fulfilling project has been in the last six months?"), ensuring that the interviewee does not deviate from the original question's objective; the two parts of the text are then naturally combined to form a modified interview question that is both friendly and guides the interviewee to provide feedback that fits their needs.

[0038] In one exemplary embodiment, the comparison process includes removing the labeled feature dimensions from the structured feature library and selecting one of the remaining feature dimensions from the structured feature library to obtain the feature dimension to be investigated.

[0039] For example, if there are multiple abilities to be assessed in an interview, it is not necessary to ask questions about each ability individually. For instance, a single context can serve as the context for multiple abilities. For example, a clothing store encounters a customer who likes several items but is hesitant. Salesperson A's colleague B is a new employee and has limited ability to persuade the customer to place more orders. In this situation, what should store manager C do? This assesses store manager C's communication skills and ability to guide others. Both abilities can be assessed using the same context S.

[0040] In one exemplary embodiment, refer to Figure 3 Step S500, after returning to the dialogue interview steps based on the feature dimensions to be questioned until all feature dimensions in the structured feature library have been accessed and marked, the dynamic programming method for interview goals based on the large model also includes, Step S600: Extract abnormal text from the feedback information based on the exaggerated statement recognition algorithm; Step S700: Ask questions to the interviewee based on the abnormal text and determine whether the interviewee answers positively. If yes, end the interview and output an evaluation report; if no, ask questions again based on the abnormal text until the interviewee answers the questions positively, then end the interview and output an evaluation report.

[0041] For example, an algorithm for identifying exaggerated statements can be used to extract anomalous text from feedback information, extracting points from the interviewee's answers that are illogical or seem exaggerated. For instance, if the interviewee is a primary school basketball teacher, and the interview is assessing his technical and tactical abilities, but the teacher answers that he frequently plays and communicates with national team players, this is clearly an exaggeration. In this case, the question would be asked again, focusing on the anomalous text "frequently plays and communicates with national team players," such as, "Can you specifically describe some of the most popular modern basketball techniques and tactics they have taught you?"

[0042] For example, by comparing the feedback from interviewees on the same topic, if there is a clear logical conflict between different statements (e.g., first stating "only 3 people participated in the project," then saying "the workload of 10 people was completed in 1 week"), the conflicting segments are marked as abnormal text. The core is to identify the inconsistencies between the statements.

[0043] Retrieve pre-defined industry standard data (such as project cycle for the same position, reasonable range of performance growth, etc.). If the quantitative statements in the feedback (such as "cost reduction of 50%" or "completion of annual target in 1 month") significantly exceed the industry standard range without special explanation, then mark the quantitative fragment as abnormal text. The core is to identify content in the data that deviates from common sense.

[0044] If the feedback contains vague statements without specific factual support (such as "the project is very effective" or "the leadership highly approves"), without mentioning specific performance indicators or specific scenarios of approval, then the vague fragment will be marked as abnormal text. The core is to identify content that is vague and lacks detail.

[0045] In one exemplary embodiment, the evaluation report includes scores for all feature dimensions in the structured feature library. Here, all feature dimensions in the structured feature library refer to the feature dimensions corresponding to the capabilities being assessed in the current dialogue.

[0046] Before conducting the dialogue assessment, each ability is pre-set with corresponding behavioral points and scoring rules. The dialogue model will score the degree of matching between the behavioral points and scoring rules, and divide the matching degree into five levels: excellent, good, medium, passable, and poor. Each dimension has a score range. The dialogue model determines which score range each feature dimension belongs to and outputs a score range.

[0047] Before the dialogue assessment begins, for each assessed capability (such as project management capability), its corresponding structured feature dimensions, behavioral points, and scoring rules are first clarified. Specifically, the behavioral points for project management capability are the concrete manifestations that meet the requirements under this dimension (e.g., the behavioral points for the Background S dimension are: clearly explaining the event background, timeline, and constraints). The scoring rules divide the degree of matching of behavioral points into three levels: Excellent, Good, and Poor. Each level corresponds to a fixed score range (e.g., Excellent: 85-100 points, Good: 70-84 points, Poor: 0-69 points), and the criteria for judging each level are clearly defined (e.g., "Excellent" must completely cover all behavioral points, "Poor" must not cover any, and "Good" must partially cover the behavioral points).

[0048] After the dialogue assessment, the dialogue model retrieves the entire dialogue record and extracts content related to the "behavioral points" of each feature dimension from the interviewee's feedback. A text matching algorithm is then used to determine the degree of overlap between the feedback content and the behavioral points. If the feedback completely covers all the behavioral points of a certain feature dimension (e.g., the feedback for dimension S includes "Q2 2024, changes in customer needs, and insufficient team manpower", which perfectly matches the preset behavioral points), it is judged as "excellent"; If the feedback only partially covers the key points of the behavior (e.g., only mentions "changes in customer needs" without mentioning time and manpower constraints), it should be judged as "good". If the feedback does not mention any behavioral points, it is judged as "poor".

[0049] Based on the matching results, the dialogue model assigns a score within the corresponding range to each feature dimension according to the preset "level-score range" correspondence. Finally, it summarizes the range scores for all feature dimensions to form the scoring section of the evaluation report. After scoring is completed, the entire dialogue interview process ends.

[0050] Reference Figure 4 , Figure 4 This is a dynamic programming process for interview objectives based on a large model. The process is divided into four stages, progressing step-by-step according to the logic of "warm-up → fact reconstruction → follow-up questioning and internalization → scoring report," as detailed below: Phase 1: Warm-up Phase; The process begins with "Start" and enters the first stage (warm-up stage); the operation of "obtaining and refining the basic user information of the interviewees" is performed, and the second stage is entered after completion.

[0051] Phase Two: The Fact-Reconstruction Phase; Perform the "Judgment of Interviewee Responses" operation to proceed to the "Whether the Response Meets Requirements" judgment stage: If the judgment is "No", return to "Redesign Questions Based on Initial Questions", adjust the questions and ask the interviewee again until the interviewee's response meets the requirements; If the judgment is "Yes": Proceed to the "Dynamic Programming Module Judges Results Based on STAR" operation, and then perform the "Whether All Principles Pass" judgment: If the judgment is "Yes", proceed to the third stage; If the judgment is "No": Perform the "Design Questions for Uninterviewed Principles Based on Dynamic Module Judgment Results" operation, and after completion, return to the "Dynamic Programming Module Judges Results Based on STAR" stage until all principles pass.

[0052] The third stage: the stage of questioning the facts and internalization; Perform the operation of "finding points of exaggeration in the interviewee's answers and asking questions based on the dialogue process" to enter the judgment stage of "judging whether the answer meets the requirements": if the judgment is "no": return to the "finding points of exaggeration in the interviewee's answers and asking questions" stage until the interviewee answers the model's questions positively; if the judgment is "yes": perform the "end the interview" operation to enter the fourth stage.

[0053] Phase Four: Scoring and Report Generation; Perform the "scoring and generating assessment report" operation. Once completed, the process will enter the "end" stage, and the entire interview process will terminate.

[0054] This application obtains basic information about the interviewees through a dialogue model and customizes interview questions for them based on the interview objectives, avoiding the limitations of a fixed question sequence and making the questions more relevant to the interviewees' backgrounds. Validation processing verifies the interviewees' feedback to ensure it aligns with the interview questions, filtering out irrelevant content and improving the quality of the interview data. A comparison process is performed between the structured feature library and the labeled feature dimensions to identify dimensions to be followed up on. Simultaneously, based on these dimensions, the dialogue interview process is repeated until all feature dimensions in the structured feature library are covered, ensuring the efficiency of the event interview and avoiding omissions and duplications.

[0055] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 5As shown, the computer device includes a processor, memory, input / output interfaces, a communication interface, a display unit, and an input device. The processor, memory, and input / output interfaces are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interfaces are used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When executed by the processor, the computer program implements a dynamic programming method for interview objectives based on a large model. The display unit is used to form a visually visible image and can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0056] Those skilled in the art will understand that Figure 5 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0057] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0058] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the above method embodiments.

[0059] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0060] 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, data stored, data displayed, 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 the relevant data must comply with relevant regulations.

[0061] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0062] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0063] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent should be determined by the appended claims.

Claims

1. A dynamic programming method for interview objectives based on a large model, characterized in that, include, Obtain basic information about the interviewees; The dialogue interview step is executed, and the interviewee is guided to output feedback information based on the basic information and the preset structured feature library; Receive the feedback information and perform verification processing on the feedback information to obtain the labeled feature dimensions; In response to the feature dimension query command, the structured feature library is retrieved and compared with the labeled feature dimensions to obtain the feature dimension to be explored. Based on the feature dimensions to be followed up, the dialogue interview steps are executed again until all feature dimensions in the structured feature library have been accessed and marked.

2. The dynamic programming method for interview objectives based on a large model according to claim 1, characterized in that, The basic information obtained from the interviewees includes... In response to the initialization command, the model outputs its own role information and guides the interviewee to output a consultation request; Check whether the consultation request has been received. If so, respond to the consultation request and then guide the interviewee to introduce themselves. If not, then directly guide the interviewee to introduce themselves to obtain the basic information.

3. The dynamic programming method for interview objectives based on a large model according to claim 1, characterized in that, The dialogue interview steps include asking questions to the interviewee based on the interview questions, and guiding the interviewee to output the feedback information; the interview questions are formulated in combination with the interview objectives and the basic information.

4. The dynamic programming method for interview objectives based on a large model according to claim 3, characterized in that, The verification process includes, Determine whether the feedback information matches the interview question. If not, modify the interview question and ask the interviewee the modified interview question until the feedback information matches the interview question. If yes, identify and mark the feature dimension corresponding to the feedback information.

5. The dynamic programming method for interview objectives based on a large model according to claim 4, characterized in that, The workarounds for the access problem include... The feedback information is analyzed to extract the theme and scene elements, and the access problem is analyzed to extract the missing elements. Based on the theme scene elements and the missing elements, a response text corresponding to the feedback information is generated; the response text includes a first guidance text that matches the theme scene elements and a second guidance text that matches the missing elements.

6. The dynamic programming method for interview objectives based on a large model according to claim 1, characterized in that, The comparison process includes removing the labeled feature dimensions from the structured feature library and selecting one of the remaining feature dimensions from the structured feature library to obtain the feature dimension to be investigated.

7. The dynamic programming method for interview objectives based on a large model according to claim 1, characterized in that, After returning to the dialogue interview steps based on the feature dimensions to be questioned until all feature dimensions in the structured feature library have been accessed and marked, the large model-based dynamic programming method for interview objectives further includes: Abnormal text in the feedback information is extracted based on the exaggerated statement recognition algorithm; The interviewee is asked questions based on the abnormal text, and it is determined whether the interviewee answers directly. If yes, the interview ends and an evaluation report is generated; if no, questions are asked again based on the abnormal text until the interviewee answers directly, at which point the interview ends and the evaluation report is generated.

8. The dynamic programming method for interview objectives based on a large model according to claim 7, characterized in that, The evaluation report includes scores for all feature dimensions in the structured feature library; the structured feature library includes background feature dimensions, target feature dimensions, action feature dimensions, and outcome feature dimensions; the basic information includes the interviewee's work background information and personality information.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the dynamic programming method for interview objectives based on a large model as described in any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the dynamic programming method for interview objectives based on a large model as described in any one of claims 1 to 8.

Images