Recruitment page building system and method for realizing personalized recommendation

Abstract

The invention discloses a recruitment page building system and method for realizing personalized recommendation. The system comprises a strategy configuration module used for providing a configuration page and configuring a recommendation strategy of a target recruitment scene in response to an operation of an operator; the page management module is used for providing an editing page, establishing a recruitment page framework in response to a dragging operation of an operator on a page component in the editing page and a binding operation of a recommendation component, and binding a recommendation strategy for the recommendation component; the recommendation engine or the search engine is used for recalling and generating a personalized recommendation content list from the database according to the user portrait data and a recommendation strategy bound with the recommendation component; and the page rendering module is used for filling the personalized recommendation content list into the recommendation component, calling an adaptation rule corresponding to the access client for rendering, and outputting a recruitment page adapted to the access client. According to the method and the device, an operator can conveniently and quickly complete establishment of personalized recommendation recruitment pages for different recruitment scenes.

Description

Technical Field

[0001] This application relates to the field of recruitment technology, and in particular to a recruitment page building system and method for implementing personalized recommendations. Background Technology

[0002] Online recruitment platforms have become a core channel connecting job seekers and employers. To improve recruitment efficiency and user experience, operations staff often need to build recruitment pages for different recruitment themes or scenarios.

[0003] However, current recruitment platform activity pages generally use fixed templates or require custom development by the technical team. This approach cannot support operations staff in independently and flexibly building pages according to rapidly changing business needs. At the same time, existing solutions for adapting to multiple terminals (such as PCs, mobile devices, and mini-programs) usually only stop at the level of basic layout adjustments, making it difficult to ensure the consistency of interactive logic and visual experience across different terminals.

[0004] Secondly, current recruitment platforms typically rely on a globally unified algorithm for recommendation logic, making it difficult for operations staff to easily configure it for specific recruitment scenarios (such as internships or urgent high-paying jobs). This results in a disconnect between page content and the campaign theme, leading to low recommendation accuracy. Furthermore, from selecting positions and configuring recommendation rules to building pages and final rendering and publishing, operations staff currently need to switch between multiple independent backend systems. This fragmented process is inefficient and makes it difficult to achieve personalized recruitment pages that allow for rapid deployment on the operations side and real-time user experience. Summary of the Invention

[0005] In view of this, embodiments of this application provide a recruitment page building system and method for implementing personalized recommendations, in order to solve at least one of the above-mentioned technical problems.

[0006] In a first aspect, embodiments of this application provide a recruitment page building system for implementing personalized recommendations, comprising: The strategy configuration module provides a configuration page and responds to operations personnel's actions on the recommendation strategy configuration component in the configuration page, configuring at least one recommendation strategy applicable to the target recruitment scenario; wherein, the recommendation strategy includes filtering rules, sorting rules and fallback rules for recommended content, and the recommended content includes job information and / or company information; The page management module provides an editing page and responds to drag-and-drop operations of page components and binding operations of recommendation components by operators. It builds the recruitment page framework and binds recommendation strategies to recommendation components, which include job recommendation components and / or enterprise recommendation components. Recommendation engines or search engines are used to respond to job seekers' login actions, obtain job seekers' user profile data, and retrieve and generate personalized recommendation content lists from the database based on the user profile data and the recommendation strategy bound to the recommendation component. The page rendering module is used to populate the personalized recommendation content list into the recommendation component, and to render the recruitment page framework by calling the corresponding adaptation rules for the accessing client from the preset multi-terminal adaptation rules, and output the recruitment page adapted to the accessing client.

[0007] According to some embodiments of this application, optionally, the screening rules include at least one of the following: job function, city of work, company identity, company size, industry, years of work experience requirement, education requirement, language requirement, salary range, and job search radius centered on the job seeker's address or desired work location; the sorting rules include at least one of the following: comprehensive sorting, posting time sorting, commuting distance sorting, job application rate sorting, company reputation sorting, and matching degree with user's expected salary sorting; the fallback rule includes supplementing the recalled recommended content with a preset popular strategy when the number of recommended content recalled according to the set screening rules is less than a preset threshold.

[0008] According to some embodiments of this application, optionally, the recruitment page building system further includes a tag configuration module; the tag management module is used to respond to the tag configuration rules applicable to the target recruitment scenario input by the operators, to perform batch scanning of positions and / or companies in the database, and to assign corresponding tags to positions and / or companies that meet the tag configuration rules; the tags are one of the filtering rules.

[0009] According to some embodiments of this application, optionally, the tag configuration rule includes one or more tag configuration conditions, which include industry, work experience requirements, recruitment urgency, company size, or job salary range; and assigning corresponding tags to positions and / or companies that meet the tag configuration rule, including: assigning sub-tags corresponding to each configuration condition in the tag configuration rule and scene tags representing the target recruitment scenario to positions and / or companies that meet the tag configuration rule.

[0010] According to some embodiments of this application, optionally, the tag management module is further configured to, when the tag configuration rules change, determine whether the positions and / or enterprises that have been tagged with historical scenario tags comply with the changed tag configuration rules, and delete the historical scenario tags of positions and / or enterprises that do not comply with the changed tag configuration rules; obtain the current status data of positions and / or enterprises that have been tagged with historical sub-tags; determine whether the current status data meets the tag configuration conditions on which the historical sub-tags were generated; if not, delete the historical sub-tags.

[0011] According to some embodiments of this application, optionally, the page management module is also used to add page materials suitable for the target recruitment scenario to the recruitment page framework in response to the upload operation of the operator. The page materials include images, videos, audio and / or text templates.

[0012] According to some embodiments of this application, the system may optionally include an advertising management module; the advertising management module is used to configure advertising placement information in the recruitment page framework in response to the operation of the operator, and the advertising placement information includes advertising materials, placement channels, placement time periods and target audience targeting conditions.

[0013] According to some embodiments of this application, optionally, the page component further includes a user incentive component; the page management module is also used to respond to the operation of the operator, set the user incentive component in the recruitment page framework, and configure the user behavior trigger conditions and incentive resources of the user incentive component. The user behavior trigger conditions include at least one of completing job application, filling in resume, or participating in page interaction; the system also includes an incentive trigger module, used to obtain the job seeker's behavior data, and when the job seeker's behavior data meets the preset trigger conditions, to issue corresponding incentive resources to the job seeker through the user incentive component.

[0014] According to some embodiments of this application, optionally, the page component further includes an AI chatbot component; the page management module is also used to set the AI ​​chatbot component in the recruitment page framework in response to the operation of the operator; wherein, the AI ​​chatbot component includes: a knowledge base construction unit, used to slice job information, company information and preset question and answer text from the database into multiple knowledge fragments, and to vectorize each knowledge fragment to generate a corresponding vectorized representation, and to store the knowledge fragments and their vectorized representations in a vector knowledge base; an intent recognition and retrieval unit, used to vectorize the job seeker's question to obtain a question vector, and to calculate the similarity between the question vector and the vectorized representation of each knowledge fragment in the vector knowledge base, and to recall one or more related knowledge fragments based on the similarity; a retrieval enhancement generation model, used to integrate the job seeker's question, the recalled knowledge fragments and the dialogue context to generate target response content; and to perform relevance reflection verification on the target response content; a response output unit, used to output the target response content if the relevance reflection verification passes, and to output preset default response content if the relevance reflection verification fails.

[0015] According to some embodiments of this application, optionally, the recommendation strategy includes a first type of recommendation strategy based on a recommendation engine and a second type of recommendation strategy based on a search engine. Specifically, when the recommendation component is bound to the first type of recommendation strategy, the recommendation engine queries the database for job information and / or company information that match the filtering rules and fallback rules based on user profile data and the filtering and fallback rules in the first type of recommendation strategy, and sorts the queried job information and / or company information according to the sorting rules in the first type of recommendation strategy to obtain a personalized recommendation content list. Specifically, when the recommendation component is bound to the second type of recommendation strategy, the search engine queries the database for job information and / or company information that match the search keywords, filtering rules, and fallback rules based on the search keywords entered by the job seeker, user profile data, and the filtering and fallback rules in the second type of recommendation strategy, and sorts the queried job information and / or company information according to the sorting rules in the second type of recommendation strategy to obtain a personalized recommendation content list.

[0016] According to some embodiments of this application, optionally, the page management module is further configured to, in response to the configuration operation of the page components by the operators, configure at least one display style of the page components and the user attribute triggering conditions corresponding to each display style; the page rendering module is further configured to, based on the user attributes in the user profile data, determine the target display style of the page components, the user attributes including at least one of job seeker age, industry, and job seeker identity; when rendering the recruitment page framework, adjust the visual style of the page components in the recruitment page framework according to the target display style, the visual style including at least one of color, font, icon, layout density, and image-text ratio.

[0017] Secondly, embodiments of this application provide a method for building a recruitment page to implement personalized recommendations, including: In response to the operations personnel's operation on the recommendation strategy configuration component in the configuration page, at least one recommendation strategy applicable to the target recruitment scenario is configured; wherein, the recommendation strategy includes filtering rules, sorting rules and fallback rules for recommended content, and the recommended content includes job information and / or company information; In response to the drag-and-drop operations of page components on the editing page and the binding operations of recommendation components on the page components by the operators, a recruitment page framework is built, and recommendation strategies are bound to the recommendation components; wherein, the recommendation components include job recommendation components and / or enterprise recommendation components; In response to a job seeker's login action, the system obtains the job seeker's user profile data and, based on the user profile data and the recommendation strategy bound to the recommendation component, retrieves and generates a personalized list of recommended content from the database. The personalized recommendation list is populated into the recommendation component, and the corresponding adaptation rule for the accessing client is called from the preset multi-terminal adaptation rules to render the recruitment page framework and output the recruitment page adapted to the accessing client.

[0018] The recruitment page building system and method for implementing personalized recommendations, as described in the embodiments of this application, integrate core functions such as strategy configuration, drag-and-drop page building, recommendation component binding, and multi-terminal rendering into a unified platform. This allows non-technical operations personnel to break free from dependence on the development team and independently and quickly complete the entire process from event planning to page launch within a visual interface. This shortens the response cycle for creating personalized pages for various recruitment scenarios (such as seasonal recruitment and industry-specific events), solving the pain points of slow customization iteration and low efficiency in multi-system operation in traditional models, and improving the operational efficiency and flexibility of recruitment business. On the other hand, by providing a flexibly configurable recommendation strategy rule library (including filtering rules, sorting rules, and fallback rules), operations personnel can finely define the recommendation logic for positions or companies based on different recruitment event themes. Combined with the real-time processing of user profile data by recommendation engines or search engines, highly personalized recommendation content lists can be dynamically generated and presented for different job seekers within the same page framework. This effectively overcomes the problems of traditional recruitment platforms having a single recommendation dimension and being disconnected from the scenario theme, thereby improving the accuracy of job matching, the exploration efficiency of job seekers, and the relevance of page content. On the other hand, through the collaboration of the preset multi-terminal adaptation rule library and the page rendering module, the corresponding layout and interaction specifications can be automatically invoked for rendering according to the user's access terminal, ensuring that a personalized recruitment page with consistent adaptation and experience can be obtained on various clients such as PC, mobile browser, mini program and App, providing users with a smooth, stable and personalized cross-terminal browsing experience. Attached Figure Description

[0019] 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.

[0020] Figure 1 This is a structural block diagram of a recruitment page building system for implementing personalized recommendations, provided in an embodiment of this application.

[0021] Figure 2 This is a schematic diagram of a configuration page.

[0022] Figure 3 This is another structural block diagram of a recruitment page building system for implementing personalized recommendations, provided in an embodiment of this application.

[0023] Figure 4 A diagram illustrating a configuration page for setting rules for tags.

[0024] Figure 5 This is an illustration of uploading page materials for a target recruitment scenario on an editing page.

[0025] Figure 6 A structural diagram of an AI chatbot component.

[0026] Figure 7 This is an operational diagram of an AI chatbot component.

[0027] Figure 8 This is a diagram illustrating how to configure a recommendation strategy on a configuration page.

[0028] Figure 9 A schematic diagram of the configuration page provided for the strategy configuration module.

[0029] Figure 10 This is a flowchart illustrating a method for building a recruitment page to implement personalized recommendations, as provided in an embodiment of this application.

[0030] Figure 11 This is a schematic diagram of the hardware structure of an electronic device according to an embodiment of this application. Detailed Implementation

[0031] The features and exemplary embodiments of various aspects of this application will be described in detail below. To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain this application and not to limit it. For those skilled in the art, this application can be implemented without some of these specific details. The following description of the embodiments is merely to provide a better understanding of this application by illustrating examples.

[0032] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes said element.

[0033] It should be understood that the term "and / or" used in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0034] Various modifications and variations can be made to this application without departing from its spirit or scope, which will be apparent to those skilled in the art. Therefore, this application is intended to cover modifications and variations falling within the scope of the corresponding claims (the claimed technical solutions) and their equivalents. It should be noted that the implementation methods provided in the embodiments of this application can be combined with each other without contradiction.

[0035] Before describing the technical solutions provided in the embodiments of this application, in order to facilitate understanding of the embodiments of this application, this application first specifically explains the problems existing in the related technologies: Online recruitment platforms have become a core channel connecting job seekers and employers. To improve recruitment efficiency and user experience, operations staff often need to build recruitment pages for different recruitment themes or scenarios.

[0036] However, current recruitment platform activity pages generally use fixed templates or require custom development by the technical team. This approach cannot support operations staff in independently and flexibly building pages according to rapidly changing business needs. At the same time, existing solutions for adapting to multiple terminals (such as PCs, mobile devices, and mini-programs) usually only stop at the level of basic layout adjustments, making it difficult to ensure the consistency of interactive logic and visual experience across different terminals.

[0037] Secondly, current recruitment platforms typically rely on globally unified algorithms for recommendation logic, making it difficult for operations staff to easily configure them for specific recruitment scenarios (such as internships or urgent high-paying jobs). For example, operations staff currently struggle to quickly configure a customized recommendation strategy for an internship-focused page, such as "filtering positions requiring no work experience and sorting by company reputation," resulting in a disconnect between page content and the event theme, and low recommendation accuracy. Furthermore, from selecting positions, configuring recommendation rules, building the page, to final rendering and publishing, operations staff currently need to switch between multiple independent backend systems. This fragmented process is inefficient and hinders the creation of personalized recruitment pages that allow for rapid deployment on the operations side and real-time user experience.

[0038] To address at least one of the aforementioned technical problems, embodiments of this application provide a recruitment page building system and method for implementing personalized recommendations.

[0039] The following section first introduces the recruitment page building system for implementing personalized recommendations provided in the embodiments of this application.

[0040] Figure 1 This is a structural block diagram of a recruitment page building system for implementing personalized recommendations, provided in an embodiment of this application. Figure 1 As shown, the recruitment page building system 10 for implementing personalized recommendations provided in this application embodiment may include a strategy configuration module 101, a page management module 102, a recommendation engine or search engine 103, and a page rendering module 104.

[0041] The strategy configuration module 101 can be used to provide a configuration page and, in response to operations personnel's actions on the recommendation strategy configuration component in the configuration page, configure at least one recommendation strategy suitable for the target recruitment scenario. The recommendation strategy may include filtering rules, sorting rules, and fallback rules for recommended content, and the recommended content may include job information and / or company information.

[0042] Figure 2 This is a schematic diagram of a configuration page. For example... Figure 2 As shown, the configuration page includes a recommendation strategy configuration component 20. This component can include a filtering rule component 201, a sorting rule component 202, and a fallback rule component 203. For example, operators can use the filtering rule component 201 to set the conditions that the retrieved recommended content must meet, such as job function, city, salary range, and company size. They can also use the sorting rule component 202 to set the order of the recommended content, such as sorting by overall ranking, posting time, commuting distance, or company reputation. Finally, the fallback rule component 203 can be used to set backup strategies when the retrieved content according to the filtering rules is insufficient, such as displaying popular jobs or companies across the platform. Through the configuration of these visual components, operators can quickly configure recommendation logic tailored to different scenarios, such as campus recruitment and urgent hiring events, without needing to code.

[0043] In some embodiments, the recommended content can be job information as the main body, with company information as an attribute of the job information. That is, the core unit for recalling and displaying is the specific job, while the company information (such as company name, logo, and size) that posted the job is presented as supplementary information. In other embodiments, the recommended content can also be a display of company information as the main body, with job information as supplementary information. For example, in a scenario of a featured company zone, a list of eligible companies is recalled and displayed first, and the company's open positions are displayed below each company's information or on the details page. In still other embodiments, the recommended content can be multiple lists of job information and company information, where some lists are mainly job information and some lists are mainly company information, which can be flexibly configured by operators according to page area and scenario.

[0044] The page management module 102 can be used to provide an editing page and respond to the drag-and-drop operations of page components in the editing page and the binding operations of recommendation components in the page components by the operators, build the recruitment page framework, and bind recommendation strategies to the recommendation components.

[0045] For example, the editing page can include, but is not limited to, title components, image and text components, separator components, list components, and core recommendation components. Recommendation components may include job recommendation components and / or company recommendation components. Operators can build the recruitment page framework by dragging and dropping these page components onto the canvas. Furthermore, after adding recommendation components to the page, operators can bind a pre-configured recommendation strategy from the strategy configuration module 101 to the recommendation component. This binding operation establishes a connection between the recommendation component and the backend recommendation logic, enabling the recommendation component to dynamically populate content according to the recommendation strategy during final rendering.

[0046] The recommendation engine or search engine 103 can be used to respond to job seekers' login actions, obtain job seekers' user profile data, and retrieve and generate a personalized list of recommended content from the database based on the user profile data and the recommendation strategy bound to the recommendation component.

[0047] For example, user profile data may include job seekers' job intentions, such as desired city, job function, salary, and industry, and may also include implicit preferences derived from analysis of job seekers' historical behavior. When a job seeker logs into a recruitment platform or visits a recruitment page frame, the recommendation engine or search engine 103 can retrieve and generate a personalized recommendation list specifically for that job seeker from the database based on the user profile data and the recommendation strategy bound to the recommendation component. For example, on the same campus recruitment page, the job list seen by computer science graduates and finance graduates will differ due to their different professional backgrounds.

[0048] The page rendering module 104 can be used to populate the personalized recommendation content list into the recommendation component, and call the corresponding adaptation rules for the accessing client from the preset multi-terminal adaptation rules to render the recruitment page framework and output the recruitment page adapted to the accessing client.

[0049] Specifically, the page rendering module 104 populates the corresponding recommendation component in the recruitment page framework with a list of personalized recommendations, completing the content population. Then, based on the detected type of access client (or access terminal) (e.g., PC, mobile, App, mini-program), the page rendering module 104 retrieves preset adaptation rules from the rule base to render the recruitment page framework for that access client (or access terminal). For example, for PCs, a multi-column layout is used, supporting mouse hover effects. For mobile devices, a single-column flow layout is used, supporting pull-to-refresh. Finally, the page rendering module 104 outputs a recruitment page that includes personalized recommendations while providing a native experience on the current device.

[0050] In some embodiments, each page component can have a built-in multi-platform attribute configuration module, such as the job list component supporting separate settings for the default number of jobs on PC and mobile. A multi-platform adaptation rule base stores adaptation rules for each client (e.g., PC / mobile / App / mini-program), which can include interaction specifications and layout rules. For example, the interaction specifications could allow PC users to click to load more jobs, while mobile users could support pull-down loading. The layout rules could allow PC components to be arranged horizontally, while mobile components are arranged vertically. The page rendering module 104 can render the recruitment page framework by calling the preset adaptation rules for that client from the rule base based on the detected access terminal type (e.g., PC, mobile, App, or mini-program), achieving a consistent multi-platform experience.

[0051] The recruitment page building system and method provided in this application for implementing personalized recommendations, on the one hand, integrate core functions such as strategy configuration, drag-and-drop page building, recommendation component binding, and multi-terminal rendering into a unified platform. This allows non-technical operations personnel to break free from dependence on the development team and independently and quickly complete the entire process from event planning to page launch within a visual interface. This shortens the response cycle for creating personalized pages for various recruitment scenarios (such as seasonal recruitment and industry-specific events), solving the pain points of slow customization iteration and low efficiency of multi-system operation in the traditional model, and improving the operational efficiency and flexibility of recruitment business. On the other hand, by providing a flexibly configurable recommendation strategy rule library (including filtering rules, sorting rules, and fallback rules), operations personnel can finely define the recommendation logic for positions or companies based on different recruitment event themes. Combined with the real-time processing of user profile data by recommendation engines or search engines, highly personalized recommendation content lists can be dynamically generated and presented for different job seekers within the same page framework. This effectively overcomes the problems of traditional recruitment platforms having a single recommendation dimension and being disconnected from the scenario theme, thereby improving the accuracy of job matching, the exploration efficiency of job seekers, and the relevance of page content. On the other hand, through the collaboration of the preset multi-terminal adaptation rule library and the page rendering module, the corresponding layout and interaction specifications can be automatically invoked for rendering according to the user's access terminal, ensuring that a personalized recruitment page with consistent adaptation and experience can be obtained on various clients such as PC, mobile browser, mini program and App, providing users with a smooth, stable and personalized cross-terminal browsing experience.

[0052] According to some embodiments of this application, the screening rules may optionally include at least one of the following: job function, city of work, company identity, company size, industry, years of work experience requirement, education requirement, language requirement, salary range, and job search radius centered on the job seeker's address or the job seeker's desired work location.

[0053] like Figure 2 As shown, in some examples, the filtering rules may include at least one of the following: job function, city, company name, company size, industry, years of work experience required, education requirement, language requirement, salary range, and a job search radius centered on the job seeker's address or desired work location. The job search radius centered on the job seeker's address or desired work location can also be referred to as latitude and longitude and the radius range. For example, the job search radius could be 5 kilometers, 10 kilometers, or other distances. Each item in the filtering rules can be clicked and further edited. For example, clicking on the city allows you to further edit the city to city A; clicking on the industry allows you to further edit the industry to internet, finance, or manufacturing, etc., and so on. Further details are omitted.

[0054] The sorting rules include at least one of the following: overall sorting, posting time sorting, commuting distance sorting, job application rate sorting, company reputation sorting, and matching degree with user's expected salary.

[0055] like Figure 2 As shown, in some examples, sorting rules may include comprehensive sorting, posting time sorting, and commuting distance sorting. Furthermore, sorting rules may also include job application rate sorting, company reputation sorting, and matching degree with user's expected salary, etc. Operations personnel can select appropriate sorting rules to rank the recommended content according to different recruitment scenarios and goals. Each sorting can be ascending or descending order, and can be flexibly set according to actual circumstances; this application does not impose any limitations on this.

[0056] A fallback rule can include supplementing the retrieved recommended content with a preset popular strategy when the number of recommended content retrieved according to the set filtering rules is less than a preset threshold. Specifically, the fallback rule ensures that job seekers can still see a certain number of recommended content even when the number of recommended content retrieved according to the set filtering rules is less than the preset threshold. Operations personnel can flexibly set the fallback rule according to the needs of the scenario.

[0057] For example, when recalling content according to the set filtering rules, if the number of recalled recommendations is less than a preset threshold (e.g., 10 items), a fallback rule is activated. Preset popular strategies may include displaying popular jobs or companies across the platform. For instance, a recruitment platform can pre-calculate jobs with high application rates and attention over a period of time as popular jobs, or companies with high visibility and significant hiring needs as popular companies. When there is insufficient recalled content, these popular jobs or companies are added to the recommended content list.

[0058] Thus, the recruitment page building system of this application provides operators with powerful configuration capabilities. By flexibly setting filtering rules, operators can provide job seekers with more accurate filtering options based on different recruitment scenarios and target audiences, improving the matching degree between recommended content and job seekers' needs, and reducing the time and effort job seekers spend searching for suitable positions in massive amounts of information. Diverse sorting rule settings allow operators to choose the most suitable sorting method based on recruitment priorities and job seeker characteristics, improving the relevance and attractiveness of recommended content, enhancing job seekers' exploration efficiency and the accuracy of job matching. The fallback rule ensures that even under strict filtering conditions, job seekers can still see a certain amount of recommended content, avoiding a decline in user experience due to insufficient recommended content, and enhancing the attractiveness and competitiveness of the recruitment platform. At the same time, the recruitment page building system also improves the work efficiency of operators, enabling them to quickly and flexibly build recruitment pages that meet the needs of different recruitment scenarios.

[0059] Figure 3 This is another structural block diagram of a recruitment page building system for implementing personalized recommendations, provided as an embodiment of this application. For example... Figure 3 As shown, according to some embodiments of this application, optionally, the recruitment page building system 10 for implementing personalized recommendations may also include a tag configuration module 105.

[0060] The tag management module 105 can respond to tag configuration rules input by operations personnel that are applicable to the target recruitment scenario, perform batch scanning of positions and / or companies in the database, and assign corresponding tags to positions and / or companies that meet the tag configuration rules. Combined with... Figure 2 As shown, tags can be used as one of the filtering rules, making it easier to achieve more accurate personalized recommendations based on tags.

[0061] In some specific examples, tag configuration rules may include one or more tag configuration conditions. For example, tag configuration conditions may include, but are not limited to, industry, work experience requirements, hiring urgency, company size, or job salary range. Figure 4 This is a diagram illustrating a configuration page for setting rules for tags. For example... Figure 4 As shown, the tag configuration rule configuration page provides an intuitive visual interface, making it easy for operations personnel to input and manage tag configuration rules. In the basic information area, operations personnel can set the basic attributes of the tags. The basic information area can include the tag it belongs to, enumeration value, enumeration ID, whether it is pushed, and remarks, etc.

[0062] For example, you can select the relevant tag from a dropdown menu. For enumeration values, you can enter them in a text input box, such as "Manufacturing Job Title." For enumeration IDs, you can enter them in a text input box. Whether to push notifications can be controlled via a toggle button. For notes, you can add descriptions in a multi-line text box, which also displays a character count.

[0063] In the rule label area, operations staff can click the "Add" button to create new rule groups. Each rule group is enclosed in curly braces {} and contains multiple condition entries. Each condition entry consists of a field, an operator, and a value. For example, fields can be selected via dropdown menus, such as Industry 1 / job_indutp1; minimum years of work experience (0 means no experience required, -1); job urgency (rating) / b_joburgen, etc. Operators can also be selected via dropdown menus, such as include any of the following, equal to, etc. Values ​​can be entered via text input boxes or label selectors, such as selecting the industry label, entering a numerical value for years of work experience, or selecting labels like urgency. Each condition entry also provides "Add Group," "Add," and "Delete" buttons on the right, allowing operations staff to flexibly adjust rules.

[0064] For example, in some examples, the tag configuration rules for the target recruitment scenario can be: Industry = Manufacturing, Work Experience Requirement = No Experience Required, and Recruitment Urgency = High.

[0065] After the operations staff completes the input of the tag configuration rules on the configuration page and clicks the save button, the tag management module 105 can immediately respond and perform a batch scan of the job information and / or company information in the database. During the scan, it can check one by one whether each job and / or company meets all the tag configuration conditions in the tag configuration rules set by the operations staff. For jobs and / or companies that meet the tag configuration rules, the corresponding tags can be applied. Specifically, sub-tags corresponding to each configuration condition in the tag configuration rules can be applied. For example, if the tag configuration rules include the industry condition of manufacturing, then the sub-tag of manufacturing can be applied to jobs and / or companies that meet the tag configuration rules. In some embodiments, scene tags that represent the target recruitment scenario, such as spring recruitment, manufacturing special sessions, etc., can also be applied to jobs and / or companies that meet the tag configuration rules, so as to display and classify them more clearly on the recruitment page.

[0066] Thus, by setting up the tag management module 105, operators can flexibly configure tag rules according to different recruitment scenarios, accurately classifying and labeling positions and companies in the database. This not only improves the efficiency of organizing and managing recruitment information, but also provides job seekers with more accurate and personalized recruitment information recommendations.

[0067] According to some embodiments of this application, optionally, the tag management module 105 can also be used to determine whether the positions and / or enterprises that have been tagged with historical scene tags conform to the changed tag configuration rules when the tag configuration rules are changed, and delete the historical scene tags of the positions and / or enterprises that do not conform to the changed tag configuration rules; obtain the current status data of the positions and / or enterprises that have been tagged with historical sub-tags; determine whether the current status data meets the tag configuration conditions on which the historical sub-tags were generated; if not, delete the historical sub-tags.

[0068] Specifically, the first step is to conduct a comprehensive scan of all positions and companies that have been tagged with historical scenario labels. The relevant information for each position and company with a historical scenario label is then compared one by one with the revised label configuration rules to determine if it complies with the new rules. For example, for a position previously tagged with the spring recruitment scenario label, if the definition of spring recruitment changes after the label configuration rules are revised, such as adjusting the recruitment timeframe, the label management module 105 can reassess whether the position still meets the requirements for spring recruitment based on the revised label configuration rules.

[0069] For positions and companies that are determined not to comply with the revised tag configuration rules, the tag management module 105 can automatically delete their historical scenario tags. Taking industry-specific scenario tags as an example, if the rules have changed and the definition of an industry-specific event has added restrictions on company size, and some companies that have already been tagged with this tag no longer meet the new company size requirements, then the tag management module 105 can delete the historical scenario tags for these companies' industry-specific events. This ensures that the scenario tags displayed on the recruitment page match the actual recruitment scenario rules, making the recommended positions and / or companies highly compatible with the recruitment scenario.

[0070] In addition, the tag management module 105 can obtain the current status data of positions and companies that have been tagged with historical sub-tags. Current status data may include the latest job requirements, the latest information about the company, and the number of views and applications for the job and / or company. For example, for a job tagged with the "high-salary" sub-tag, the tag management module 105 can obtain data such as the current salary range and benefits for that job. For a company tagged with the "large enterprise" sub-tag, the tag management module 105 can obtain data such as the current number of employees and asset size of that company.

[0071] The tag management module 105 can compare the current status data obtained with the tag configuration conditions used when generating historical sub-tags. For example, if the historical sub-tag "experienced candidates preferred" for a job was generated based on a work experience requirement of 5 years or more, but the current work experience requirement for the job has changed to 3 years or more, then it is determined that the current status data of the job does not meet the conditions for generating the "experienced candidates preferred" sub-tag.

[0072] If the determination result indicates that the current status data of the position and / or the company does not meet the tag configuration conditions used when generating the historical sub-tag, the tag management module 105 can delete the historical sub-tag for the position and / or the company. Continuing with the example of the high-salary position mentioned above, if the salary range of the position decreases due to company adjustments and no longer meets the salary standards set for the high-salary position sub-tag, the tag management module 105 can delete the high-salary position sub-tag from the position.

[0073] Thus, by updating tags in real time or on a schedule through the tag management module 105, the tags for positions and companies in the database can always accurately reflect the latest recruitment scenarios and requirements. This makes the positions / companies displayed on the recruitment page more authentic and reliable, providing job seekers with more precise screening and reference criteria. At the same time, it also improves the management efficiency of recruitment information, avoids information chaos caused by inconsistencies between tags and rules, and further enhances the user experience.

[0074] According to some embodiments of this application, optionally, the page management module 102 can also be used to add page materials suitable for the target recruitment scenario in response to the upload operation of the operator. The page materials include images, videos, audio and / or text templates.

[0075] Specifically, the page management module 102 can respond to the upload operation of the operators and add page materials suitable for the target recruitment scenario to the recruitment page framework. The page materials include images, videos, audio and / or text templates.

[0076] Figure 5 This is an illustration of uploading page materials representing a target recruitment scenario to an editing page. For example... Figure 5 As shown in some examples, operations staff can customize the colors of page components according to the needs of the target recruitment scenario. For instance, by entering the corresponding color codes in the input boxes for theme color, swipe-up overlay color, and button color, the corresponding color blocks will be displayed on the right, allowing operations staff to intuitively view the color effects. For example, if the target recruitment scenario has a lively and enthusiastic style, operations staff can set the button color to a bright orange (such as #FF721B) to attract job seekers' attention.

[0077] In the component background image area, operations staff can click the upload area to upload images. If a gradient background image has already been uploaded, they can also click the "+" button on the right to add a new image. This allows operations staff to change the mobile page background image according to different seasons or recruitment themes, creating an atmosphere that matches the target recruitment scenario. For example, when recruiting in the fall, a background image with autumn elements can be uploaded.

[0078] The "More Job Background Images" section allows operations staff to upload more job background images, such as by adding new images via the "+" button on the right, to update the visual effects of the job display area. The "Special Zone Name" section allows operations staff to enter a special zone name, such as "Autumn Recruitment," in the input box, clearly indicating the theme of the recruitment zone or recruitment scenario.

[0079] In the icon section of the recruitment zone, operations staff can upload icons representing their target recruitment scenarios, such as icons with a blue circular background and white text indicating autumn recruitment. The example links below help staff view icon examples to ensure uploaded icons match the overall design style. In the subtitle section of the recruitment area, operations staff can enter promotional slogans such as "Autumn Recruitment Starts, Apply Now!" to enhance the attractiveness of recruitment information.

[0080] In the PC widescreen browsing area, a toggle button allows operators to control whether the feature is enabled. Operators can decide whether to provide a widescreen browsing experience for PC users based on actual needs.

[0081] After the operations staff completes the uploading and configuration of page materials, they can perform further processing using the operation buttons in the lower right corner. Clicking the "Cancel" button will discard the current changes. Clicking the "Save and Preview" button will save the current progress and preview the actual display effect of the recruitment page, allowing for timely identification and adjustment of any unsatisfactory aspects. Clicking the "Submit for Review" button will submit the final configuration for review. Once approved, these page materials will be officially applied to the recruitment page for the target recruitment scenario.

[0082] Thus, by responding to the upload operations of operations personnel through the page management module 102, page materials suitable for the target recruitment scenario can be flexibly added to the recruitment page framework. This allows the recruitment page to be personalized according to different recruitment themes and target audiences, enhancing the attractiveness and relevance of recruitment information. A rich variety of page materials (such as images, videos, audio, and text templates) can showcase information about recruiting companies and positions from multiple dimensions, providing job seekers with a better browsing experience, increasing their attention and participation in recruitment information, and ultimately improving recruitment effectiveness. At the same time, the clear layout and operation process of the editing page can improve the work efficiency of operations personnel, making it convenient for them to quickly complete the configuration and updating of page materials.

[0083] According to some embodiments of this application, optionally, the recruitment page building system 10 for implementing personalized recommendations may also include an advertising management module. The advertising management module can be used to configure advertising placement information in the recruitment page framework in response to operations by operators. The advertising placement information includes advertising materials, placement channels, placement time periods, and target audience targeting conditions.

[0084] Specifically, the advertising management module allows operations staff to select and upload suitable advertising creatives for recruitment pages. Advertising creatives can include various formats such as images, videos, and text ads. For example, for a recruitment ad from a technology company, operations staff can upload the company logo, product images, or a video showcasing the company's work environment as advertising creatives.

[0085] Additionally, the system allows operations staff to select appropriate advertising channels based on the characteristics of the target audience and recruitment needs. To improve advertising effectiveness, the advertising management module allows operations staff to set advertising time slots based on recruitment deadlines and the target audience's active times. For example, if the target audience is primarily office workers, operations staff can choose to run ads on weekday evenings or weekends to increase ad exposure. Operations staff can set specific dates and time periods, and the system will automatically run and stop ads according to these settings.

[0086] The advertising management module also allows operations staff to set target audience targeting criteria to ensure that ads are accurately delivered to potential job seekers. Target audience targeting criteria can include multiple dimensions such as age, gender, work experience, education, and industry. For example, if recruiting marketing personnel with over 5 years of experience, operations staff can set targeting criteria such as age 28-40, relevant marketing experience, and a bachelor's degree or higher. Based on the target audience targeting criteria, the system filters out qualified candidates and delivers ads to them.

[0087] Thus, through the advertising management module, the recruitment page building system can achieve more precise and effective advertising. It allows operations personnel to flexibly configure advertising information according to different recruitment scenarios and target audiences, improving ad exposure and conversion rates. Precise target audience targeting settings ensure that ads reach potential job seekers, improving recruitment efficiency and effectiveness.

[0088] According to some embodiments of this application, the page component may optionally include a user incentive component. The user incentive component can be used to increase job seeker participation and motivation.

[0089] Accordingly, the page management module 102 can also be used to respond to the operations of the operators, set up a user incentive component in the recruitment page framework, and configure the user behavior trigger conditions and incentive resources of the user incentive component. The user behavior trigger conditions include at least one of completing job application, filling in resume or participating in page interaction.

[0090] Specifically, operations personnel can set up user incentive components within the recruitment page framework through drag-and-drop operations. Secondly, they can configure user behavior trigger conditions for these components based on recruitment needs and goals. These trigger conditions can include at least one of completing a job application, filling out a resume, or participating in page interactions. For example, for an urgent recruitment campaign, operations personnel can set the trigger condition to completing a job application; that is, the incentive mechanism is triggered when a job seeker successfully submits their application. For recruitment scenarios that require collecting more job seeker information, filling out a resume can be used as the trigger condition. Furthermore, to increase page interactivity, operations personnel can also set participation in page interactions (such as liking, commenting, and sharing) as trigger conditions.

[0091] In addition to setting user behavior triggers, operations staff can also configure incentive resources. These incentive resources include, but are not limited to, coupons, points, virtual gifts, and physical gifts.

[0092] Accordingly, the recruitment page building system 10 for implementing personalized recommendations can also have an incentive trigger module, which is used to obtain job seekers' behavioral data. When the job seeker's behavioral data meets the preset trigger conditions, the corresponding incentive resources are issued to the job seeker through the user incentive component.

[0093] Specifically, the incentive trigger module can be used to acquire job seekers' behavioral data. For example, when a job seeker completes a job application on a recruitment page, the incentive trigger module can record information such as the job title and application time; when a job seeker fills out a resume, the incentive trigger module can acquire basic information about the resume, work experience, educational background, and other data.

[0094] Then, the incentive triggering module compares the acquired job seeker behavior data with preset user behavior trigger conditions. If the job seeker's behavior data meets the user behavior trigger conditions, the incentive distribution process is immediately initiated.

[0095] When a job seeker's behavioral data meets the user behavior trigger conditions, the incentive trigger module can distribute corresponding incentive resources to the job seeker through the user incentive component. The specific distribution method can vary depending on the type of incentive resource. For example, for coupons, coupon codes can be automatically sent to the job seeker's pre-registered email address or mobile phone number. For points or virtual gifts, the corresponding points or virtual gifts can be directly added to the job seeker's account.

[0096] Therefore, by setting up user incentive components in the recruitment page system and combining them with the working mechanism of the incentive trigger module, job seeker participation and enthusiasm can be improved. Appropriately configured user behavior trigger conditions and incentive resources can guide job seekers to complete key steps in the recruitment process, such as submitting job applications and filling out resumes, thereby improving recruitment efficiency and effectiveness. At the same time, increasing page interaction trigger conditions can also enhance the activity of the recruitment page and expand the reach of recruitment information.

[0097] According to some embodiments of this application, the page component may optionally include an AI chatbot component. The page management module 102 can also be used to set the AI ​​chatbot component in the recruitment page framework in response to operations by operators. Operators can add the AI ​​chatbot component to the recruitment page framework via drag-and-drop operations on the editing page provided by the page management module 102.

[0098] Figure 6 A structural diagram of an AI chatbot component. Figure 7 This is a schematic diagram illustrating the operation of an AI chatbot component. (Combined with...) Figure 6 and Figure 7As shown, the AI ​​chatbot component 60 may include a knowledge base construction unit 601, an intent recognition and retrieval unit 602, a retrieval enhancement generation model 603, and a response output unit 604.

[0099] The knowledge base construction unit 601 can be used to slice job information, enterprise information and preset question and answer text from the database into multiple knowledge fragments; and to vectorize each knowledge fragment, converting it into a vector form that the computer can understand and process, generating the corresponding vectorized representation; and to store the knowledge fragments and their vectorized representations in the vector knowledge base.

[0100] The intent recognition and retrieval unit 602 can be used to vectorize the job seeker's question to obtain a question vector, calculate the similarity between the question vector and the vectorized representation of each knowledge fragment in the vector knowledge base, and recall one or more related knowledge fragments based on the similarity.

[0101] Specifically, when a job seeker asks a question to the AI ​​chatbot on a job posting page, the intent recognition and retrieval unit 602 first vectorizes the job seeker's question, converting the natural language question into a question vector. Then, it calculates the similarity between the question vector and the vectorized representations of various knowledge segments in the vector knowledge base. For example, if a job seeker asks, "Does this position offer five social insurances and one housing fund?", the intent recognition and retrieval unit 602 can convert this question into a question vector and search for knowledge segments with high similarity in the vector knowledge base. Based on the similarity, it recalls one or more relevant knowledge segments, such as information about the position offering five social insurances and one housing fund.

[0102] The retrieval-enhanced generative model 603 can integrate the job seeker's question, recalled knowledge fragments, and dialogue context to generate target response content. Combining the job seeker's question, "Does this position offer five social insurances and one housing fund?", the recalled knowledge fragment, "This position provides five social insurances and one housing fund," and previous dialogue content, the target response content is generated, such as "Hello, this position does offer five social insurances and one housing fund." Then, the retrieval-enhanced generative model 603 can perform relevance reflection and verification on the target response content, checking whether the response content is relevant to the job seeker's question and whether it is accurate and reasonable.

[0103] The response output unit 604 can be used to output the target response content and display it to the job seeker if the relevance reflection verification passes; otherwise, it can output a preset default response content. The preset default response content can be flexibly adjusted according to the actual situation, and this application does not limit it.

[0104] In this way, AI chatbot components can provide job seekers with accurate and timely information, improving their experience and recruitment efficiency. In practical applications, they can be used in various scenarios, such as when job seekers inquire about job details, learn about company information, or ask about the interview process. For example, when a job seeker asks, "What is the interview process like?", the AI ​​chatbot component can quickly retrieve relevant information from its knowledge base and generate a detailed response, helping the job seeker understand each stage of the interview.

[0105] Figure 8 This is a diagram illustrating how to configure a recommendation strategy on a configuration page. For example... Figure 8 As shown, the basic information area on the configuration page allows operations personnel to define the basic attributes of the recommendation strategy. For example, they can enter a name for the recommendation strategy, such as "25 Autumn Recruitment - YPS - Industry - Consumer Retail," which is the unique identifier for this strategy and typically includes information such as time, business line, industry, or scenario for easy identification and management. The filtering rules are the core logic configuration area for the recommendation strategy, determining which positions will be recommended to target users. For example, filtering rules can include job intention, company ID, job function, work city, upper and lower limits of monthly salary, operations job section, company code, and custom career tags. In addition, operations personnel can also configure sorting rules and fallback rules (…). Figure 8 (Not shown).

[0106] According to some embodiments of this application, the recommendation strategy may optionally include a first type of recommendation strategy based on a recommendation engine and a second type of recommendation strategy based on a search engine. The first type of recommendation strategy and the second type of recommendation strategy may be different or the same, and can be flexibly configured according to actual conditions. This application does not limit this.

[0107] Specifically, when the recommendation component is bound to the first type of recommendation strategy, the recommendation engine can query job information and / or company information that match the filtering rules and fallback rules in the database based on user profile data and the filtering rules and fallback rules in the first type of recommendation strategy, and sort the queried job information and / or company information according to the sorting rules in the first type of recommendation strategy to obtain a personalized recommendation content list.

[0108] The difference between a search engine and a recommendation engine lies in the fact that a search engine can receive search keywords entered by job seekers, thereby making more accurate recommendations for that job seeker. Specifically, when the recommendation component is bound to a second type of recommendation strategy, a search engine can, based on the search keywords entered by the job seeker, user profile data, and the filtering and fallback rules in the second type of recommendation strategy, query the database for job information and / or company information that matches the search keywords, filtering rules, and fallback rules, and sort the retrieved job information and / or company information according to the sorting rules in the second type of recommendation strategy to obtain a personalized recommendation list.

[0109] According to some embodiments of this application, optionally, the page management module 102 can also be used to configure at least one display style of the page component and the user attribute triggering conditions corresponding to various display styles in response to the configuration operation of the operation personnel on the page component.

[0110] Specifically, in some embodiments, the editing page provided by the page management module 102 allows operators to configure the display style of page components, such as selecting one or more preset display styles, or customizing the display style. For each display style, operators can set specific visual style parameters, such as color scheme, font type, icon style, layout density, and / or text-image ratio. For example, for a fashionable display, operators can choose bright color combinations, such as pink and blue; use distinctive fonts, such as handwritten fonts; and pair them with cute icons to increase the flexibility of the layout and the diversity of text-image ratios.

[0111] Additionally, operations staff can set various user attribute trigger conditions for different display styles based on job seekers' age, industry, and job application status. For example, for job seekers aged 20-25, a fashion-themed display can be triggered; for job seekers in the internet industry, a tech-themed display can be triggered.

[0112] Accordingly, the page rendering module 104 can also be used to determine the target display style of the page components based on the user attributes in the user profile data. The user attributes include at least one of the job seeker's age, industry, and job seeker status. When rendering the recruitment page frame, the visual style of the page components in the recruitment page frame is adjusted according to the target display style. The visual style includes at least one of color, font, icon, layout density, and text-image ratio.

[0113] Specifically, the page rendering module 104 can obtain user attribute information such as the job seeker's age, industry, and job seeker status from user profile data. Based on the obtained user attribute information, it matches it with the trigger conditions configured in the page management module 102 to determine the target display style of the page components. For example, if the job seeker is 22 years old, works in the internet industry, and is a recent graduate, the target display style is determined to be a fashionable, tech-savvy, and vibrant style based on the configured trigger conditions. Then, the page rendering module 104 adjusts the visual style of the page components in the recruitment page framework according to the target display style. The visual style includes at least one of the following: color, font, icons, layout density, and text-to-image ratio.

[0114] Thus, through the collaborative work of the aforementioned page management module and page rendering module, the recruitment page building system of this application can dynamically adjust the display style of page components according to user attributes, providing a personalized recruitment page experience for different job seekers, thereby improving job seeker participation and recruitment effectiveness.

[0115] In addition, some embodiments allow operators to personalize the appearance, location, and triggering method of the AI ​​chatbot component to better match the overall style and needs of the recruitment page. For example, operators can set the AI ​​chatbot component to float in the lower right corner of the page, making it convenient for job seekers to initiate conversations at any time; they can also set it to automatically pop up the AI ​​chatbot window after a job seeker stays on the page for a certain period of time, proactively interacting with the job seeker.

[0116] Figure 9 This is a diagram illustrating the configuration page provided for the strategy configuration module. For example... Figure 9 As shown in some embodiments of this application, optionally, the configuration page allows operators to configure recommendation strategies and select the client application for the target recruitment scenario or recommendation strategy, such as PC, mobile H5, App, and / or mini-program. Secondly, operators can select channels to reach job seekers, such as email, App Push, in-app messages, SMS, official account template messages, and mini-program subscription messages. Through the selected channels, messages about establishing target recruitment scenarios / topics or personalized recommendations recalled according to the recommendation strategy can be pushed to job seekers, thereby increasing job seekers' attention to and response rate to recruitment information, and attracting more suitable talent for recruiting companies.

[0117] Additionally, the configuration page allows operations personnel to select the engine used for the target recruitment scenario or recommendation strategy, such as a recommendation engine or a search engine. Furthermore, the configuration page may include version management components, A / B testing components, and analytics reporting components.

[0118] The version management component can automatically generate new versions when operations staff edit or modify recommendation strategies. Data from the old and new versions will be presented separately in various data reports, making it easy for operations staff to observe the data effects before and after the edit.

[0119] A / B testing components can be used in various ways. For example, operations staff can create two different versions of a job posting page and then use A / B testing to randomly distribute users to these two versions. By observing various data metrics of the different job posting pages, such as click-through rate and user dwell time, they can determine which job posting page performs better. For instance, if operations staff modify the copy of a job posting page, A / B testing can determine whether the new copy attracts more users to click on the job details.

[0120] The analytics reporting component can be used to provide operational staff with analytical reports. For example, these reports may include detailed data on various key metrics such as click-through rate (CTR), conversion rate (CVR), impressions, and user dwell time. Through these reports, operational staff can understand the actual effectiveness of their recommendation strategies, identify existing problems, and adjust and optimize their strategies based on the analysis results. For instance, if a particular version of the strategy shows a low CTR, operational staff can use the analytics report to find the reasons, such as whether the recall rules are set improperly or the ranking rules fail to highlight key positions, and then make targeted improvements.

[0121] Based on the recruitment page building system 10 for implementing personalized recommendations provided in the above embodiments, this application also provides a method for building a recruitment page for implementing personalized recommendations. For example, this method can be implemented based on the recruitment page building system 10 for implementing personalized recommendations provided in any of the above embodiments.

[0122] Figure 10 This is a flowchart illustrating a method for building a recruitment page to implement personalized recommendations, provided in an embodiment of this application. Figure 10 As shown in the embodiments of this application, the method for building a recruitment page to implement personalized recommendations may include the following steps: S101: In response to the operation of the operation personnel on the recommendation strategy configuration component in the configuration page, configure at least one recommendation strategy applicable to the target recruitment scenario; wherein, the recommendation strategy includes filtering rules, sorting rules and fallback rules for recommended content, and the recommended content includes job information and / or company information; S102: In response to the drag-and-drop operations of page components on the editing page and the binding operations of recommendation components in the page components by the operators, build the recruitment page framework and bind recommendation strategies to the recommendation components; wherein, the recommendation components include job recommendation components and / or enterprise recommendation components; S103: In response to the job seeker's login operation, obtain the job seeker's user profile data, and based on the user profile data and the recommendation strategy bound to the recommendation component, retrieve and generate a personalized recommendation content list from the database; S104: Populate the personalized recommendation list into the recommendation component, and render the recruitment page framework by calling the corresponding adaptation rule from the preset multi-terminal adaptation rules for the accessing client, and output the recruitment page adapted to the accessing client.

[0123] The specific processes of steps S101 to S104 above have been described in detail above and will not be repeated here.

[0124] The recruitment page building method for personalized recommendations provided in this application integrates core functions such as strategy configuration, drag-and-drop page building, recommendation component binding, and multi-terminal rendering into a unified platform. This allows non-technical operations personnel to break free from dependence on development teams and independently and quickly complete the entire process from event planning to page launch within a visual interface. This shortens the response cycle for creating personalized pages for various recruitment scenarios (such as seasonal recruitment and industry-specific events), solving the pain points of slow customization iteration and low efficiency in multi-system operation in traditional models, and improving the operational efficiency and flexibility of recruitment business. On the other hand, by providing a flexibly configurable recommendation strategy rule library (including filtering rules, sorting rules, and fallback rules), operations personnel can finely define the recommendation logic for positions or companies based on different recruitment event themes. Combined with the real-time processing of user profile data by recommendation engines or search engines, highly personalized recommendation content lists can be dynamically generated and presented for different job seekers within the same page framework. This effectively overcomes the problems of traditional recruitment platforms having a single recommendation dimension and being disconnected from the scenario theme, thereby improving the accuracy of job matching, the exploration efficiency of job seekers, and the relevance of page content. On the other hand, through the collaboration of the preset multi-terminal adaptation rule library and the page rendering module, the corresponding layout and interaction specifications can be automatically invoked for rendering according to the user's access terminal, ensuring that a personalized recruitment page with consistent adaptation and experience can be obtained on various clients such as PC, mobile browser, mini program and App, providing users with a smooth, stable and personalized cross-terminal browsing experience.

[0125] In some embodiments, the screening rules may optionally include at least one of the following: job function, city of work, company identity, company size, industry, years of work experience requirement, education requirement, language requirement, salary range, and job search radius centered on the job seeker's address or the job seeker's desired work location; The sorting rules include at least one of the following: overall sorting, posting time sorting, commuting distance sorting, job application rate sorting, company reputation sorting, and matching degree with users' expected salary. The fallback rule includes supplementing the recalled recommended content with a preset popular strategy when the number of recommended content recalled according to the set filtering rules is less than a preset threshold.

[0126] In some embodiments, the method may optionally include: in response to the tag configuration rules for the target recruitment scenario input by the operations personnel, performing a batch scan of positions and / or companies in the database, and assigning corresponding tags to positions and / or companies that meet the tag configuration rules; the tags are one of the filtering rules.

[0127] In some embodiments, optionally, the tag configuration rule includes one or more tag configuration conditions, which include industry, work experience requirements, recruitment urgency, company size, or job salary range; and assigning corresponding tags to positions and / or companies that meet the tag configuration rule, including: assigning sub-tags corresponding to each configuration condition in the tag configuration rule and scene tags representing the target recruitment scenario to positions and / or companies that meet the tag configuration rule.

[0128] In some embodiments, the method may optionally include: when the tag configuration rules change, determining whether the positions and / or enterprises that have been tagged with historical scenario tags comply with the changed tag configuration rules, and deleting the historical scenario tags of the positions and / or enterprises that do not comply with the changed tag configuration rules; obtaining the current status data of the positions and / or enterprises that have been tagged with historical sub-tags; determining whether the current status data meets the tag configuration conditions on which the historical sub-tags were generated; if not, deleting the historical sub-tags.

[0129] In some embodiments, the method may optionally include: in response to an upload operation by an operator, adding page materials suitable for the target recruitment scenario to the recruitment page framework, the page materials including images, videos, audio and / or text templates.

[0130] In some embodiments, the method may optionally include: in response to an operator's action, configuring advertising information in the recruitment page framework, the advertising information including advertising materials, advertising channels, advertising time periods, and target audience targeting conditions.

[0131] In some embodiments, the method may optionally include: in response to the operation of the operator, setting a user incentive component in the recruitment page framework, and configuring the user behavior trigger conditions and incentive resources of the user incentive component, wherein the user behavior trigger conditions include at least one of completing job application, filling in resume or participating in page interaction; obtaining job seeker behavior data, and when the job seeker behavior data meets the preset trigger conditions, issuing corresponding incentive resources to the job seeker through the user incentive component.

[0132] In some embodiments, the method may optionally include: in response to an operator's operation, setting an AI chatbot component in the recruitment page framework; the AI ​​chatbot component includes: a knowledge base construction unit, used to slice job information, company information and preset question and answer text from a database into multiple knowledge fragments, and to vectorize each knowledge fragment to generate a corresponding vectorized representation, and to store the knowledge fragments and their vectorized representations in a vector knowledge base; an intent recognition and retrieval unit, used to vectorize the job seeker's question to obtain a question vector, and to calculate the similarity between the question vector and the vectorized representations of each knowledge fragment in the vector knowledge base, and to recall one or more relevant knowledge fragments based on the similarity; a retrieval enhancement generation model, used to integrate the job seeker's question, the recalled knowledge fragments and the dialogue context to generate target response content; and to perform relevance reflection verification on the target response content; a response output unit, used to output the target response content if the relevance reflection verification passes, and to output a preset default response content if the relevance reflection verification fails.

[0133] In some embodiments, the method may optionally include: in response to the operator's configuration operation on the page component, configuring at least one display style of the page component and the user attribute triggering conditions corresponding to each display style; Based on user attributes in the user profile data, determine the target display style of the page components. User attributes include at least one of the following: job seeker age, industry, and job seeker status. When rendering the recruitment page framework, adjust the visual style of the page components in the recruitment page framework according to the target display style. The visual style includes at least one of the following: color, font, icon, layout density, and text-image ratio.

[0134] It should be noted that the method for building a recruitment page to achieve personalized recommendations can have the same or corresponding technical features as the recruitment page building system for achieving personalized recommendations provided in the above product embodiments, and produce the same technical effects. For the sake of brevity, it will not be elaborated further here.

[0135] The electronic device in this application embodiment may be a user terminal device, a server, other computing devices, or a cloud server. Figure 11 This is a schematic diagram of the hardware structure of an electronic device according to an embodiment of this application. The electronic device may include a processor 1101 and a memory 1102 storing computer program instructions. When the processor 1101 executes the computer program instructions, it implements the process or function of the recruitment page building method for personalized recommendation in any of the above embodiments.

[0136] Specifically, processor 1101 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 1102 may include mass storage for data or instructions. For example, memory 1102 may be at least one of the following: 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 1102 may include removable or non-removable (or fixed) media. Furthermore, memory 1102 may be internal or external to an electronic device. Memory 1102 may be non-volatile solid-state memory. In other words, typically memory 1102 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. The processor 1101 reads and executes computer program instructions stored in the memory 1102 to implement any of the processes or functions of the recruitment page building method for personalized recommendations in the above embodiments.

[0137] In one example Figure 11 The illustrated electronic device may also include a communication interface 1103 and a bus 1110. The processor 1101, memory 1102, and communication interface 1103 are connected via bus 1110 and communicate with each other. Communication interface 1103 is primarily used to enable communication between modules, devices, units, and / or equipment in the embodiments of this application. Bus 1110 includes hardware, software, or both, and can couple components of the online data traffic billing device together. For example, 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 buses. Bus 1110 may include one or more buses. Although specific buses are described or illustrated in the embodiments of this application, any suitable bus or interconnection method may be considered in the embodiments of this application.

[0138] In conjunction with the methods in the above embodiments, this application also provides a computer-readable storage medium storing computer program instructions. When executed by a processor, the computer program instructions implement the process or function of any of the methods for building a recruitment page for personalized recommendations in the above embodiments.

[0139] In addition, this application also provides a computer program product that stores computer program instructions. When the computer program instructions are executed by a processor, they implement the process or function of any of the recruitment page building methods for personalized recommendations described above.

[0140] The flowcharts and / or block diagrams of methods, apparatuses, 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 that enables the implementation of the function / action specified in each block or combination thereof in the flowcharts and / or block diagrams, executable via such processor. Such a processor may be a general-purpose processor, a dedicated processor, a special-purpose application processor, or a field-programmable logic circuit.

[0141] 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.

[0142] 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 descriptions 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, 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 person skilled in the art can conceive of various equivalent modifications or substitutions within the scope of the technology disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application.

Claims

1. A recruitment page building system for implementing personalized recommendations, characterized in that, include: The strategy configuration module provides a configuration page and responds to operations personnel's actions on the recommendation strategy configuration component in the configuration page, configuring at least one recommendation strategy applicable to the target recruitment scenario; wherein, the recommendation strategy includes filtering rules, sorting rules and fallback rules for recommended content, and the recommended content includes job information and / or company information; The page management module provides an editing page and responds to drag-and-drop operations of page components on the editing page and binding operations of recommendation components among the page components by operators. It builds the recruitment page framework and binds recommendation strategies to the recommendation components; wherein, the recommendation components include job recommendation components and / or enterprise recommendation components. Recommendation engines or search engines are used to respond to job seekers' login actions, obtain job seekers' user profile data, and retrieve and generate personalized recommendation content lists from the database based on the user profile data and the recommendation strategy bound to the recommendation component. The page rendering module is used to populate the personalized recommendation content list into the recommendation component, and to render the recruitment page framework by calling the corresponding adaptation rules for the accessing client from the preset multi-terminal adaptation rules, and output the recruitment page adapted to the accessing client.

2. The system according to claim 1, wherein the screening rules include at least one of the following: job function, city of work, company identity, company size, industry, years of work experience requirement, education requirement, language requirement, salary range, and job search radius centered on the job seeker's address or the job seeker's desired work location; The sorting rules include at least one of the following: comprehensive sorting, posting time sorting, commuting distance sorting, job application rate sorting, company reputation sorting, and matching degree with user's expected salary. The fallback rule includes supplementing the recalled recommended content with a preset popular strategy when the number of recommended content recalled according to the set filtering rules is less than a preset threshold.

3. The system according to claim 1, characterized in that, The system also includes a tag configuration module; The tag management module is used to respond to the tag configuration rules input by the operations personnel that are applicable to the target recruitment scenario, to perform batch scanning of positions and / or companies in the database, and to assign corresponding tags to positions and / or companies that meet the tag configuration rules; the tags are one of the filtering rules.

4. The system according to claim 3, characterized in that, Tag configuration rules include one or more tag configuration conditions, which may include industry, work experience requirements, recruitment urgency, company size, or job salary range. To assign corresponding tags to positions and / or companies that meet the tag configuration rules, including: To assign sub-tags corresponding to the various configuration conditions in the tag configuration rules and scene tags representing the target recruitment scenario to positions and / or companies that meet the tag configuration rules.

5. The system according to claim 4, characterized in that, The tag management module is also used to determine whether the positions and / or enterprises that have been tagged with historical scenario tags comply with the changed tag configuration rules when the tag configuration rules are changed, and to delete the historical scenario tags of positions and / or enterprises that do not comply with the changed tag configuration rules. Retrieve the current status data of positions and / or enterprises that have been tagged with historical sub-tags; determine whether the current status data meets the tag configuration conditions used to generate the historical sub-tags; if not, delete the historical sub-tags.

6. The system according to claim 1, characterized in that, The page management module is also used to respond to the upload operation of the operators and add page materials suitable for the target recruitment scenario to the recruitment page framework. The page materials include images, videos, audio and / or text templates.

7. The system according to claim 1, characterized in that, The system also includes an advertising management module; The advertising management module is used to respond to the operations of the operators and configure advertising placement information in the recruitment page framework. The advertising placement information includes advertising materials, placement channels, placement time periods, and target audience targeting conditions.

8. The system according to claim 1, characterized in that, The page components also include user incentive components; The page management module is also used to respond to the operations of the operators, set up a user incentive component in the recruitment page framework, and configure the user behavior trigger conditions and incentive resources of the user incentive component. The user behavior trigger conditions include at least one of completing job application, filling in resume or participating in page interaction. The system also includes an incentive triggering module, which is used to acquire job seekers' behavioral data. When the job seeker's behavioral data meets preset triggering conditions, corresponding incentive resources are issued to the job seeker through the user incentive component.

9. The system according to claim 1, characterized in that, The page components also include an AI chatbot component; The page management module is also used to respond to the operations of the operators by setting up an AI chatbot component in the recruitment page framework; The AI ​​chatbot component includes: The knowledge base construction unit is used to slice job information, enterprise information and preset question and answer text from the database into multiple knowledge fragments, and to vectorize each knowledge fragment to generate a corresponding vectorized representation, and to store the knowledge fragments and their vectorized representations into a vector knowledge base. The intent recognition and retrieval unit is used to vectorize the job seeker's question to obtain a question vector, calculate the similarity between the question vector and the vectorized representation of each knowledge fragment in the vector knowledge base, and recall one or more relevant knowledge fragments based on the similarity. A retrieval-enhanced generative model is used to integrate job seekers' questions, recalled knowledge fragments, and dialogue context to generate target response content; and to perform relevance-based verification on the target response content. The response output unit is used to output the target response content if the relevance reflection verification passes, and to output the preset default response content if the relevance reflection verification fails.

10. The system according to claim 1, characterized in that, The recommendation strategies include a first type of recommendation strategy based on recommendation engines and a second type of recommendation strategy based on search engines; The recommendation engine is specifically used to query job information and / or company information that meet the filtering rules and fallback rules from the database when the recommendation component is bound to the first type of recommendation strategy, based on user profile data and the filtering rules and fallback rules in the first type of recommendation strategy, and sort the queried job information and / or company information according to the sorting rules in the first type of recommendation strategy to obtain a personalized recommendation content list. Specifically, when the recommendation component is bound to the second type of recommendation strategy, the search engine queries the database for job information and / or company information that match the search keywords, filtering rules, and fallback rules, based on the search keywords entered by the job seeker, user profile data, and the filtering and fallback rules in the second type of recommendation strategy. The search engine then sorts the retrieved job information and / or company information according to the sorting rules in the second type of recommendation strategy to obtain a personalized recommendation content list.

11. The system according to claim 1, characterized in that, The page management module is also used to respond to the configuration operations of the operators on the page components, and to configure at least one display style of the page components and the user attribute triggering conditions corresponding to each display style. The page rendering module is also used to determine the target display style of the page components based on the user attributes in the user profile data. The user attributes include at least one of the job seeker's age, industry, and job seeker status. When rendering the recruitment page frame, the visual style of the page components in the recruitment page frame is adjusted according to the target display style. The visual style includes at least one of color, font, icon, layout density, and image-text ratio.

12. A method for building a recruitment page to implement personalized recommendations, characterized in that, include: In response to the operations personnel's operation on the recommendation strategy configuration component in the configuration page, at least one recommendation strategy applicable to the target recruitment scenario is configured; wherein, the recommendation strategy includes filtering rules, sorting rules and fallback rules for recommended content, and the recommended content includes job information and / or company information; In response to the drag-and-drop operations of operators on page components in the editing page and the binding operations of recommendation components in the page components, a recruitment page framework is built, and recommendation strategies are bound to the recommendation components; wherein, the recommendation components include job recommendation components and / or enterprise recommendation components; In response to a job seeker's login action, the system obtains the job seeker's user profile data and, based on the user profile data and the recommendation strategy bound to the recommendation component, retrieves and generates a personalized list of recommended content from the database. The personalized recommendation list is populated into the recommendation component, and the corresponding adaptation rule for the accessing client is called from the preset multi-terminal adaptation rules to render the recruitment page framework and output the recruitment page adapted to the accessing client.

Images