Human resource agent cue word generation method, equipment and product
By establishing a mapping relationship between variable names and variable placeholders in the human resources intelligent agent, prompt words are dynamically generated, solving the problem that the intelligent agent prompt word template cannot flexibly respond to changes in business logic and data analysis, and realizing efficient prompt word iteration and enterprise-level unified management.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- QIAN JIN NETWORK INFORMATION TECH SHANGHAI LTD
- Filing Date
- 2025-12-23
- Publication Date
- 2026-05-12
AI Technical Summary
In existing technologies, the intelligent agent prompt word templates in the human resources industry cannot flexibly respond to changes in business logic and data analysis, resulting in long iteration cycles, poor reusability, and an inability to achieve unified enterprise-level management.
By acquiring variable name configuration information and prompt word templates, a mapping relationship between variable names and variable placeholders is established, and intelligent agent prompt words are dynamically generated to achieve flexible data configuration and automated processing.
It improves the iteration efficiency of prompt words and the agility of business response, reduces development and maintenance costs, and enables flexible configuration and rapid optimization of prompt words.
Smart Images
Figure CN122019587A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of Internet technology, as well as artificial intelligence and prompt word generation technology, and in particular to a method, device, program product and storage medium for generating prompt words for a human resources intelligent agent. Background Technology
[0002] Currently, large-scale models and the intelligent agents built upon them are widely used in the human resources industry for core scenarios such as resume screening, job matching, interview assessment, and career planning. In the implementation of these applications, prompt words, as the core instruction set guiding the large-scale model to understand tasks and execute specialized analyses, directly determine the accuracy and practicality of the intelligent agent's output based on their quality and adaptability. Currently, prompt word templates in the industry are generally static templates written by algorithm engineers or business experts. If specific candidate resume data or job description information needs to be introduced as the basis for analysis, the data fields are usually embedded into the template in a hard-coded form.
[0003] When business logic iterates or data analysis dimensions change, such as when the evaluation model expands its focus from "years of service" to "gender" and "age," or when the structure of job information fields is adjusted, developers must intervene and manually modify the application code that calls the prompt words to adapt to the new variable input requirements. This results in the prompt word optimization iteration cycle being constrained by the development schedule, making it impossible to respond quickly to business feedback. Furthermore, the same set of prompt word templates has poor reusability in different applications or scenarios, resulting in fragmented management and making it difficult to achieve unified enterprise-level control. Summary of the Invention
[0004] In view of this, embodiments of this application provide a method for generating intelligent agent prompt words based on a prompt word management platform, an electronic device, a computer-readable storage medium, and a computer program product, thereby solving at least one technical problem.
[0005] This application provides a method for generating prompt words for a human resources intelligent agent, comprising: obtaining variable name configuration information, the variable name configuration information including at least one variable name, each variable name including a main variable name and a sub-variable name, wherein the main variable name is used to identify a data category, the data category including a resume information category and a recruitment information category, and the sub-variable name is used to identify a specific data field under the data category; obtaining a prompt word template, the prompt word template including natural language text and at least one variable defined by a predetermined identifier, each variable placeholder being defined by a predetermined identifier, and the name of the variable placeholder being consistent with the variable name in the variable name configuration information; parsing the prompt word template, identifying multiple variable placeholders defined by the predetermined identifier, and extracting the variable name corresponding to each variable placeholder; establishing a mapping relationship, associating each variable name in the variable name configuration information with the variable placeholders identified in the prompt word template, forming a one-to-one correspondence between variable names and variable placeholders; and obtaining corresponding field values from user resumes and / or recruitment information according to the variable name configuration information; Based on the mapping relationship, each variable placeholder in the prompt word template is replaced with the corresponding field value to generate an agent prompt word.
[0006] Optionally, the method according to the embodiments of this application further includes: inputting the generated agent prompt word into a large model selected by the prompt word writer, so that the selected large model outputs a corresponding result according to the prompt word.
[0007] Optionally, according to the method of this application embodiment, the main variable name includes resume and job title, which are used to indicate that the field value comes from user resume information or job posting information, respectively.
[0008] Optionally, according to the method of this application embodiment, when the main variable name is resume, its corresponding sub-variable name includes at least one of gender, age, personal strengths and years of work experience; when the main variable name is job title, its corresponding sub-variable name includes at least one of job title and job requirements.
[0009] Optionally, the method according to the embodiments of this application further includes: obtaining a prompt word template and corresponding variable name configuration information according to a business code, wherein the business code has a one-to-one correspondence with the prompt word template and the corresponding variable name configuration information.
[0010] Optionally, according to the method of this application embodiment, before establishing the mapping relationship, the method further includes: parsing the variables in the prompt word template, verifying the semantic consistency between the variables and the variable names based on natural language processing technology, and issuing a warning or suggesting adjustment to the prompt word writer if there is a discrepancy.
[0011] Optionally, the method according to the embodiments of this application further includes: dynamically reconstructing the paragraph order or format of the prompt word template based on the type or number of variable names, so as to adapt to the input requirements of different large models or optimize readability.
[0012] This application provides a human resources intelligent agent prompt word generation device, comprising: a first acquisition module, configured to acquire variable name configuration information, the variable name configuration information including at least one variable name, each variable name including a main variable name and a sub-variable name, wherein the main variable name is used to identify a data category, the data category including a resume information category and a recruitment information category, and the sub-variable name is used to identify a specific data field under the data category; a second acquisition module, configured to acquire a prompt word template, the prompt word template including natural language text and at least one variable defined by a predetermined identifier, each variable placeholder being defined by a predetermined identifier, and the name of the variable placeholder being consistent with the variable name in the variable name configuration information; a parsing module, configured to parse the prompt word template, identify multiple variable placeholders defined by the predetermined identifier, and extract the variable name corresponding to each variable placeholder; an association module, configured to establish a mapping relationship, associating each variable name in the variable name configuration information with the variable placeholders identified in the prompt word template, forming a one-to-one correspondence between variable names and variable placeholders; and a third acquisition module, configured to acquire corresponding field values from user resumes and / or recruitment information according to the variable name configuration information. The replacement module is used to replace each variable placeholder in the prompt word template with the corresponding field value based on the mapping relationship, so as to generate intelligent agent prompt words.
[0013] This application provides an electronic device, which includes a processor and a memory storing computer program instructions; the electronic device executes the computer program instructions to implement the method described above.
[0014] This application provides a computer program product, which includes computer program instructions that, when executed, implement the method described above.
[0015] This application provides a computer-readable storage medium storing computer program instructions, which, when executed by a processor, implement the method described above.
[0016] This application decouples the data requirements for prompt words from the application code, allowing business personnel to flexibly add, delete, or modify the resume or job fields on which the prompt words depend through a configuration interface without touching the underlying code. This directly eliminates the frequent code modification and release cycles caused by fine-tuning of business logic, greatly improving the iteration efficiency of intelligent agent prompt words and the agility of business response. Attached Figure Description
[0017] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings of the embodiments of this application will be briefly described below.
[0018] Figure 1 This is a schematic diagram of the system architecture of an embodiment of this application.
[0019] Figure 2 This is a flowchart illustrating the human resources intelligent agent prompt word generation method according to an embodiment of this application.
[0020] Figure 3 This is a schematic diagram of the front-end page of a configurable large model prompt word management platform for the human resources industry, according to an embodiment of this application.
[0021] Figure 4 This is a schematic diagram of data interaction between the front-end and back-end of the prompt word management platform and the intelligent agent in an embodiment of this application.
[0022] Figure 5 This is a schematic diagram of an electronic device used to implement the intelligent agent prompt word generation method based on the prompt word management platform in the embodiments of this application. Detailed Implementation
[0023] The principles and spirit of this application will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are provided to make the principles and spirit of this application clearer and more thorough. The exemplary embodiments provided herein are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art without inventive effort are within the scope of protection of this application.
[0024] Embodiments of this application relate to terminal devices and / or servers. Implementations of this application can be a system, terminal, device, method, computer-readable storage medium, or computer program product, and can be specifically implemented as entirely hardware, entirely software, or a combination of hardware and software. Figure 1This diagram illustrates a system architecture according to an embodiment of the present application, including a terminal device 102 and a server 104. The terminal device 102 may include at least one of the following: a smartphone, tablet computer, laptop computer, desktop computer, smart TV, various wearable devices, augmented reality (AR) devices, virtual reality (VR) devices, etc. A client application, such as an app, mini-program, or browser-based client, can be installed on the terminal device 102. Users can input commands through the client, and the terminal device 102 can send request information containing the commands to the server 104. Upon receiving the request information, the server 104 performs corresponding processing and returns the processing result information to the terminal device 102. The server 104 may be a local server or a cloud server, and may be a single server or a server cluster, etc.
[0025] In this document, the terms "first," "second," "third," etc., are used only to distinguish one entity (or operation) from another in textual description, and do not require or imply any sequential order between these entities (or operations).
[0026] Figure 2 This is a flowchart illustrating the human resource intelligent agent prompt word generation method according to an embodiment of this application. For example... Figure 2 As shown, the method includes the following steps: S101: Obtain variable name configuration information. Variable name configuration information includes at least one variable name. Each variable name includes a main variable name and a sub-variable name. The main variable name is used to identify the data category. The data categories include resume information category and recruitment information category. The sub-variable name is used to identify the specific data field under the data category. S102: Obtain the prompt word template, which includes natural language text and at least one variable delimited by a predefined identifier. Each variable placeholder is delimited by a predefined identifier, and the name of the variable placeholder is consistent with the variable name in the variable name configuration information. S103: Parse the prompt word template, identify multiple variable placeholders defined by predefined identifiers, and extract the variable name corresponding to each variable placeholder; S104: Establish a mapping relationship, associate each variable name in the variable name configuration information with the variable placeholders identified in the prompt word template, and form a one-to-one correspondence between variable names and variable placeholders; S105: Based on the variable name configuration information, retrieve the corresponding field values from the user's resume and / or recruitment information; S106: Based on the mapping relationship, replace each variable placeholder in the prompt word template with the corresponding field value to generate the agent prompt word.
[0027] The human resources intelligent agent prompt word generation method proposed in this application constructs a dynamic and configurable prompt word generation mechanism to address the problem of frequent prompt word iteration caused by changes in business logic or data requirements in the development of intelligent agents in the human resources field.
[0028] In this embodiment, it is necessary to obtain variable name configuration information predefined by the administrator or prompt word engineer. This variable name configuration information includes at least one variable name, each of which mainly includes a main variable name and a sub-variable name. The main variable name acts as an identifier for the data category, such as resume or job title, to indicate the data source category to which the variable belongs. The sub-variable name is used to specify the specific attribute fields under the data category, such as age and years of work experience under the resume category, or job title and job information under the job title category. Thus, when searching the resume and job information uploaded or filled in by users (referring to users using the app with the intelligent agent) stored in the backend, the system will determine the job title based on the main variable name "job title" and then determine the job title in the job title information based on the sub-variable name "job title". If the job title in the job title information is "algorithm engineer", then the system will finally fill in "algorithm engineer" as the variable name (job title, job title) in the variable position of the prompt word template. In this embodiment, the prompt word writer refers to the business personnel of the app who can write prompt words for the intelligent agent.
[0029] In this embodiment, the prompt word template is a text that integrates a natural language task description and variable placeholders. The variable placeholders are defined by predetermined identifiers, such as the following prompt word template: ## Role I excel at resume evaluation. I will comprehensively assess a candidate's abilities based on their qualifications (CV information). ## Task Description XXX ## Enter the following information: CV information: "gender": Resume. Gender , "age": Resume. Age , "Personal strengths": Resume. Personal Strengths , Years of work experience: Resume. Years of work experience , in Resume. Gender , Resume. Personal Strengths These elements, delimited by double-flower symbols, are variable placeholders. The names of these variable placeholders are the same as the variable name configuration information obtained by the system.
[0030] In some embodiments of this application, optionally, variable name configuration information and prompt word templates can be input into the system through a prompt word management platform. The prompt word management platform at least includes a variable name providing module and a prompt word editing module. The variable name providing module provides variable name configuration information for user selection, and the template editing module receives prompt word templates input by the prompt word writer. In some embodiments of this application, optionally, the variable name configuration information obtained by the prompt word management platform is stored in the backend in the form of a list. Subsequently, the system retrieves the prompt word template uploaded by the prompt word writer and the variable names selected by the prompt word writer from the backend, fills in the actual information, and obtains the prompt word for the intelligent agent.
[0031] Furthermore, in this embodiment, establishing a mapping relationship involves associating each variable placeholder extracted from the parsed prompt template with the corresponding variable name defined in the variable name configuration information, forming a one-to-one binding relationship. When generating the final prompt, the data to be filled in at a specific location in the template (identified by a placeholder) should come from which specific field (sub-variable name) under which data category (main variable name). Then, based on the variable names (especially the sub-variable names) pointed to in the established mapping relationship, the system actively retrieves and extracts the corresponding actual field values from the user's resume and / or recruitment information. For example, for the variable name "resume.age", the system will retrieve the age value "30" from the resume record of the currently evaluated candidate; for "position.position name", the system will retrieve the position name "Java Development Engineer" from the target position record.
[0032] By establishing a mapping relationship between variable names and variable placeholders, the prompt word template is able to dynamically acquire data. Ultimately, the agent extracts actual information from the resume and job information uploaded by the user based on the variable names selected by the writer, thereby generating personalized and complete prompt words in batches that can be directly used by large models.
[0033] This application configurates data, templates, and mapping relationships, enabling prompt writers to modify variable names, update the natural language of prompt templates, and name variable placeholders to handle the same changes in requirements. This completely avoids the code modification, testing, and deployment processes, greatly improving the efficiency and flexibility of prompt writing and iteration. It also allows business experts without technical backgrounds to directly participate in prompt optimization, significantly reducing development and maintenance costs caused by changes in business requirements. Ultimately, this enables human resources services to adapt to the diverse and rapid optimization requirements of prompts in different intelligent agent applications.
[0034] Optionally, in some embodiments of this application, it may also include: The generated agent prompts are input into the large model selected by the prompt writer, so that the selected large model outputs the corresponding results based on the prompts.
[0035] In this embodiment, the system also provides a variety of large models for prompt writers to choose from. This allows prompt writers to select a suitable large model from the provided list based on business needs (such as requirements for cost, speed, and professional field). Then, when the agent is working, it can pass the complete prompt, model name, model parameters, etc., to the interface of the large model service. After obtaining the return results from the large model, it sends job information matching the user's resume to the user based on the returned results.
[0036] In this embodiment, the prompt word writer can select different large models without needing to develop and interface with each model separately, significantly improving the platform's versatility and the deployment efficiency of business intelligence applications. Furthermore, after obtaining the agent prompt words, the prompt words can be directly input into the selected large model service interface to obtain the required analysis, evaluation, or generation results. The entire process achieves automated flow from abstract configuration to concrete instance generation.
[0037] In some embodiments of this application, optionally, the main variable name includes resume and job title, which are used to indicate that the field value comes from user resume information or job posting information, respectively.
[0038] This application provides a typical example of a main variable name, which divides the scope of main variable names into two categories: resume and job title. This allows the intelligent agent to quickly retrieve the actual information corresponding to the variable name configuration information from the database.
[0039] In some embodiments of this application, optionally, when the main variable name is resume, its corresponding sub-variable name includes at least one of gender, age, personal strengths and years of work experience; When the main variable is named "job title", its corresponding sub-variable names include at least one of "job title" and "job requirements".
[0040] This application provides typical examples of main variable names and corresponding sub-variable names. It categorizes main variable names into two main types: "resume" and "job title," specifically specifying that sub-variable names for the "resume" variable name can include gender, age, personal strengths, and years of work experience, while sub-variable names for the "job title" variable name can include job title and job information. By standardizing and classifying core human resource data variable names, this provides prompt writers with intuitive and business-scenario-appropriate data selection, reducing the understanding cost of template construction and ensuring the accuracy and effectiveness of the data sources required for prompts.
[0041] In some embodiments of this application, optionally, the prompt word template and the corresponding variable name configuration information are obtained according to the business code, wherein the business code and the prompt word template and the corresponding variable name configuration information have a one-to-one correspondence.
[0042] In this embodiment, a unique business code is assigned to each prompt word template and its corresponding selected variable name. The system can retrieve the corresponding prompt word template and variable name based on the business code. Specifically, the system assigns a unique identifier to each arranged prompt word template. This business code acts as a globally unique index for the prompt word template and its complete configuration (including the mapping relationship between variables and variable names). Through this code, the specified prompt word template and all its associated information can be accurately and quickly located and invoked, realizing the systematic and identifiable management of a large number of prompt words, greatly facilitating template reuse, version control, and integration with external business systems.
[0043] In some embodiments of this application, optionally, before establishing the mapping relationship, the method further includes: parsing the variables in the prompt word template, verifying the semantic consistency between the variables and the variable names based on natural language processing technology, and issuing a warning or suggesting adjustment to the prompt word writer if there is a discrepancy.
[0044] Specifically, before establishing the final mapping between variable names and variable placeholders, the prompt word template is first parsed, and natural language processing techniques are used to understand the variable placeholders (such as...). Job title The system assesses the semantic relevance between the selected variable name and the actual job title (e.g., "resume, ideal job"). If inconsistencies exist, the AI proactively warns the user or provides adjustment suggestions, thereby preventing potential data mismatches upstream in the process.
[0045] By performing semantic consistency verification before mapping, human configuration errors can be proactively detected and warned, significantly reducing the risk of incorrect prompt word generation or unreliable output results of large models due to mismatch between dependent variables and variable names.
[0046] In some embodiments of this application, the method may optionally further include: dynamically reconstructing the paragraph order or format of the prompt word template based on the type or number of variable names, in order to adapt to the input requirements of different large models or optimize readability.
[0047] This application's embodiments no longer treat the prompt word template as fixed, static text, but rather allow for dynamic adjustment based on the characteristics of the selected variable names. For example, when multiple "project experience" sub-variable names are detected, the prompt word template can be automatically reconstructed into a bullet-point listing format to enhance readability; or, to adapt to the sensitive requirements of a specific large model for input length, the paragraph order can be dynamically adjusted based on the estimated length of the variable name content, placing key information at the beginning.
[0048] The method described in this application can dynamically optimize the structure and format of the final prompt words, thereby better adapting to the characteristics and input requirements of different large models and improving the understanding quality of the prompt words and the final output results of the large models.
[0049] The implementation methods and advantages of the embodiments of this application have been described above through multiple examples. The specific processing procedures of the embodiments of this application are described in detail below with reference to specific examples.
[0050] Figure 3 This is a schematic diagram of the front-end page of a prompt word management platform according to an embodiment of this application. Combined with... Figure 3 As shown, the prompt word generation process in this application embodiment is as follows: Step 1: In the model selection box 401 of the prompt word management platform, select the large model for which you want to input prompt words; Step 2: In the variable name selection box 402 below the input field, select the desired variable name. In one of the variable names shown in the figure, the main variable name is "Position" and the sub-variable name is "Position Name". Step 3: Enter the prompt template in the system prompt box 403 or the user prompt box 404. Use the variable placeholders in the prompt template. Mark it. The content inside must match the selected variable name, for example... Job Title. Position Here is an example of a prompt word template: ## Role I am skilled in resume evaluation. Based on the candidate-job fit, the original job requirements (JD), and the candidate's qualifications (CV), I will comprehensively evaluate whether the candidate meets the needs of the current position. ## Task Description XXXXX ## Enter the following information: Job title : Job Title
[0051] Job Information : Job Position. Job Information
[0052] CV information : "gender": Resume. Gender , "age": Resume. Age , "Personal strengths": Resume. Personal Strengths , Years of work experience: Resume. Years of work experience , Step 4: Save the prompt word settings and assign a business code, such as RGMATCH. The backend stores the information, which will officially take effect after the user publishes it. The prompt word management platform can also manage this set of prompt words, including which model is used, model parameters, and other information. Step 5: If further adjustments are needed, you can click on the existing prompts to edit them. After clicking "Publish," the latest version number will be generated, and the old version will be discontinued, but the page can still be traced. Step 6: Implement the person-job matching program (intelligent agent). In the code, use the business logic `RGMATCH` to retrieve the latest prompt word template and corresponding variable list from the management platform. The variable list reveals the required variable names. After obtaining the corresponding variable names from the resume and job information, the prompt word template will be updated accordingly. The wrapped variables are replaced with their true values to obtain complete cue words for the large model. The output cue words are as follows: ## Role I am skilled in resume evaluation. Based on the candidate-job fit, the original job requirements (JD), and the candidate's qualifications (CV), I will comprehensively evaluate whether the candidate meets the needs of the current position. ## Task Description XXXXX ## Enter the following information: Job title : Java development engineer Job Information : Proficient in Spring Boot...... CV information : Gender: Male Age: 30 "Personal Strengths": Years of experience as a Java development manager... Years of work experience: 8 Step 7: The agent sends the complete prompt words, model name, model parameters, etc. to the interface of the large model service to obtain the large model parsing results.
[0053] Figure 4 This is a schematic diagram illustrating the data interaction between the front-end and back-end of the prompt word management platform and the intelligent agent in an embodiment of this application. For example... Figure 4 As shown, after the prompt writer inputs the prompt template on the prompt management platform page, the platform's backend saves the input prompt template and the variable names selected by the prompt writer, with the variable names stored as a list. Thus, when the developed agent receives a resume uploaded or written online by a user, it automatically retrieves the prompt template and the stored variable name list from the prompt management platform's backend. Based on the retrieved variable name list, it extracts the corresponding information from the user's resume and then fills the variable content positions in the prompt template with the information extracted from the resume. Finally, the agent calls the large model interface, inputting the filled-in prompt template into the large model interface to obtain the large model's parsing results. In this application, the large model's parsing results can be used to recommend positions matching the user's resume to the user.
[0054] Accordingly, embodiments of this application also provide a human resource intelligent agent prompt word generation device, including: The first acquisition module is used to acquire variable name configuration information, which includes at least one variable name. Each variable name includes a main variable name and a sub-variable name. The main variable name is used to identify the data category, which includes resume information category and recruitment information category. The sub-variable name is used to identify the specific data field under the data category. The second acquisition module is used to acquire a prompt word template, which includes natural language text and at least one variable defined by a predetermined identifier. Each variable placeholder is defined by a predetermined identifier, and the name of the variable placeholder is consistent with the variable name in the variable name configuration information. The parsing module is used to parse the prompt word template, identify multiple variable placeholders defined by the predetermined identifier, and extract the variable name corresponding to each variable placeholder; The association module is used to establish a mapping relationship, associating each variable name in the variable name configuration information with the variable placeholders identified in the prompt word template, forming a one-to-one correspondence between variable names and variable placeholders; The third acquisition module is used to obtain the corresponding field values from the user's resume and / or recruitment information based on the variable name configuration information; The replacement module is used to replace each variable placeholder in the prompt word template with the corresponding field value based on the mapping relationship, so as to generate intelligent agent prompt words.
[0055] Based on at least one of the above embodiments, the electronic device in the embodiments of this application may be a user terminal device, a server, other computing devices, or a cloud server. Figure 5 The diagram illustrates the hardware structure of an electronic device according to an embodiment of this application. The electronic device may include a processor 601 and a memory 602 storing computer program instructions. The processor 601 reads and executes the computer program instructions stored in the memory 602 to implement the process or function of any of the methods described above.
[0056] Specifically, processor 601 may include a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application. Memory 602 may include a mass storage device for data or instructions. For example, memory 602 may be at least one of the following: a hard disk drive (HDD), read-only memory (ROM), random access memory (RAM), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, universal serial bus (USB) drive, or other physical / tangible memory storage device. Alternatively, memory 602 may include removable or non-removable (or fixed) media. Furthermore, memory 602 may be internal or external to the integrated gateway disaster recovery device. Memory 602 may be non-volatile solid-state memory. In other words, typically memory 602 includes a tangible (non-transitory) computer-readable storage medium (such as a memory device) encoded with computer-executable instructions, and when the software is executed (e.g., by one or more processors), it can perform the operations described in the methods of the embodiments of this application.
[0057] As an example, Figure 5The illustrated electronic device may also include a communication interface 603 and a bus 610. The processor 601, memory 602, and communication interface 603 are connected via bus 610 and communicate with each other. Bus 610 may include hardware, software, or both, and may couple components of an online data traffic metering device together. The bus may include at least one of the following: Accelerated Graphics Port (AGP) or other graphics bus, Enhanced Industry Standard Architecture (EISA) bus, Front Side Bus (FSB), HyperTransport (HT) Interconnect, Industry Standard Architecture (ISA) bus, Infinite Bandwidth Interconnect, Low Pin Count (LPC) bus, memory bus, Microchannel Architecture (MCA) bus, Peripheral Component Interconnect (PCI) bus, PCI-Express (PCI-X) bus, Serial Advanced Technology Attachment (SATA) bus, Video Electronics Standards Association Local (VLB) bus, or other suitable bus. Bus 610 may include one or more buses. Although specific buses are described or shown in the embodiments of this application, any suitable bus or interconnection method is contemplated in the embodiments of this application.
[0058] In conjunction with the methods in the above embodiments, this application also provides a computer-readable storage medium storing computer program instructions, which, when executed by a processor, implement the process or function of any of the methods in the above embodiments.
[0059] This application also provides a computer program product that stores computer program instructions, which, when executed by a processor, implement the process or function of any of the methods described above.
[0060] The flowcharts and / or block diagrams of methods, terminals, systems, and computer program products according to embodiments of this application have been exemplarily described above, and related aspects have been described. It should be understood that each block or combination thereof in the flowcharts and / or block diagrams may be implemented by computer program instructions, by dedicated hardware performing a specified function or action, or by a combination of dedicated hardware and computer instructions. For example, these computer program instructions may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to form a machine such that these instructions, executed via such processor, enable the implementation of the function / action specified in each block or combination thereof in the flowcharts and / or block diagrams. Such a processor may be a general-purpose processor, a dedicated processor, a special-purpose application processor, or a field-programmable logic circuit.
[0061] The functional blocks shown in the structural block diagrams of this application can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc.; when implemented in software, they are programs or code segments used to perform the required tasks. Programs or code segments can be stored in memory or transmitted over a transmission medium or communication link via data signals carried on a carrier wave. Code segments can be downloaded via computer networks such as the Internet or intranets.
[0062] It should be noted that this application is not limited to the specific configurations and processes described above or shown in the figures. The above are merely specific embodiments of this application. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the described systems, devices, terminals, modules, or units can be referred to the corresponding processes in the method embodiments, and need not be repeated here. It should be understood that the scope of protection of this application is not limited thereto. Any equivalent modifications or substitutions that can be conceived by those skilled in the art within the scope of the technology disclosed in this application should be covered within the scope of protection of this patent application.
Claims
1. A method for generating prompt words for a human resource intelligent agent, characterized in that, include: Obtain variable name configuration information, which includes at least one variable name. Each variable name includes a main variable name and a sub-variable name. The main variable name is used to identify the data category, which includes resume information category and recruitment information category. The sub-variable name is used to identify the specific data field under the data category. Obtain a prompt word template, which includes natural language text and at least one variable defined by a predetermined identifier. Each variable placeholder is defined by a predetermined identifier, and the name of the variable placeholder is consistent with the variable name in the variable name configuration information. The prompt word template is parsed to identify multiple variable placeholders defined by the predetermined identifier, and the variable name corresponding to each variable placeholder is extracted. Establish a mapping relationship, associating each variable name in the variable name configuration information with the variable placeholders identified in the prompt word template, forming a one-to-one correspondence between variable names and variable placeholders; Based on the variable name configuration information, obtain the corresponding field values from the user's resume and / or recruitment information; Based on the mapping relationship, each variable placeholder in the prompt word template is replaced with the corresponding field value to generate an agent prompt word.
2. The method according to claim 1, characterized in that, Also includes: The generated agent prompts are input into the large model selected by the prompt writer, so that the selected large model outputs the corresponding results based on the prompts.
3. The method according to claim 1, characterized in that, The main variable name includes "resume" and "job title", which are used to indicate that the field value comes from user resume information or job posting information, respectively.
4. The method according to claim 3, characterized in that, When the main variable is named "Resume", its corresponding sub-variable names include at least one of gender, age, personal strengths, and years of work experience; When the main variable is named "job title", its corresponding sub-variable name includes at least one of "job title" and "job requirements".
5. The method according to claim 1, characterized in that, Also includes: The prompt word template and corresponding variable name configuration information are obtained based on the business code, wherein the business code, the prompt word template, and the corresponding variable name configuration information have a one-to-one correspondence.
6. The method according to claim 1, characterized in that, Before establishing the mapping relationship, the process also includes: parsing the variables in the prompt word template, verifying the semantic consistency between the variables and variable names based on natural language processing technology, and issuing a warning or suggesting adjustments to the prompt word writer if there is a discrepancy.
7. The method according to claim 5, characterized in that, Also includes: Based on the type or number of variable names, the paragraph order or format of the prompt word template is dynamically reconstructed to adapt to the input requirements of different large models or to optimize readability.
8. An electronic device, characterized in that, The electronic device is a terminal device or a server. The electronic device includes a processor and a memory storing computer program instructions. When the electronic device executes the computer program instructions, it implements the method as described in any one of claims 1-7.
9. A computer program product, characterized in that, It includes computer program instructions that, when executed, implement the method as described in any one of claims 1-7.
10. A computer-readable storage medium, characterized in that, It stores computer program instructions that, when executed, implement the method as described in any one of claims 1-7.