Building worker recruitment method and system

Abstract

The invention provides a building worker recruitment method and system, and relates to the technical field of data processing. The method comprises the following steps: acquiring application information of an applicant; determining one or more reference workers based on the target post and / or resume information; and based on the historical rectification information of the one or more reference workers, determining the assessment content of the applicant. The system comprises an acquisition module used for acquiring application information of applicants; the first determination module is used for determining one or more reference workers based on the target post and / or resume information; and the second determination module is used for determining the assessment content of the applicant based on the historical rectification information of the one or more reference workers. According to the invention, enterprise own data and external data can be fully utilized to formulate recruitment assessment content more suitable for enterprise management styles or special scenes, and the assessment result of the recruitment process is utilized to customize targeted training content for recruitment personnel.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method and system for recruiting construction workers. Background Technology

[0002] In the construction industry, traditional recruitment processes rely heavily on resume matching (e.g., determining which candidates advance to the interview stage) and using a fixed question bank for candidate assessment (e.g., written tests). It's important to note that a candidate's past performance is significantly influenced by company management style (e.g., frequency of changes, salary system) and project type (e.g., high-rise building projects, subway projects). However, existing recruitment systems, when matching resumes, only compare keywords (e.g., job type, length of service, projects participated in) and cannot determine a candidate's actual work capabilities in a specific environment. Furthermore, using a fixed question bank fails to dynamically reflect whether a candidate will pose a specific engineering risk after joining the company (e.g., welding defects in nuclear power projects), and post-employment training programs are not linked to assessment results. For example, high-scoring candidates lack advanced training content, while low-scoring candidates' weaknesses are not addressed.

[0003] Based on this, we aim to propose a method and system for recruiting construction workers, which integrates multi-source data (enterprise / project / rectification records) to build a predictive model and achieve scenario-based and precise recruitment assessment. Summary of the Invention

[0004] To address the issues of how to avoid a mismatch between recruited construction workers and recruitment needs, and how to develop targeted training programs for recruited workers, this invention provides a method and system for recruiting construction workers.

[0005] The invention includes a method for recruiting construction workers. The method is applied to a recruitment platform and includes: obtaining user operation instructions through the recruitment platform; responding to the operation instructions, retrieving applicant information through a first terminal device; invoking a first determination module to determine one or more reference workers based on target positions and / or resume information; invoking a second determination module to determine the applicant's assessment criteria based on the historical performance information of the one or more reference workers; and sending the assessment criteria to the user.

[0006] The invention includes a construction worker recruitment system. The system comprises: an acquisition module, used to retrieve applicant information via a first terminal device in response to a user's operation command; a first determination module, used to determine one or more reference workers based on target positions and / or resume information; a second determination module, used to determine the assessment criteria for applicants based on the historical performance information of the one or more reference workers; and a sending module, used to send the assessment criteria to the user.

[0007] The invention includes a construction worker recruitment device, comprising a processor for executing a construction worker recruitment method.

[0008] The invention includes a computer-readable storage medium that stores computer instructions. When a computer reads the computer instructions from the storage medium, the computer executes a method for recruiting construction workers.

[0009] The beneficial effects of the above invention include, but are not limited to: (1) determining the assessment content by selecting reference workers similar to the applicants and extracting historical rectification information, instead of using a general question bank for assessment, making the assessment content more targeted, avoiding disconnection from job requirements, and reducing the mismatch rate between applicants and job positions; (2) transforming post-remedial rectification information into information for pre-screening, which can identify potential risks during the recruitment stage, reducing the possibility of applicants having similar problems after joining the company, thereby reducing rework, delays, quality problems, etc. caused by similar problems, and optimizing construction quality and efficiency; (3) using historical rectification information of multiple reference workers for information complementarity when determining the assessment content. (3) Ensure the completeness and diversity of assessment content, and avoid poor quality of assessment content due to lack of data or information silos; (4) Use the assessment results as an important reference source for training content, so that training can address the shortcomings exposed by the applicants in the assessment in a targeted manner; (5) By strengthening the training on the problems corresponding to the historical rectification information, the applicants can fully understand the possible risk points before taking up the post, master the correct operating methods and preventive measures, so as to avoid repeating the mistakes of the reference workers in future work to the greatest extent; (6) By developing targeted training content for applicants with high assessment scores, the high-scoring applicants can advance on the basis of being competent in the target position and be able to handle more and more difficult tasks. Attached Figure Description

[0010] This specification will be further described by way of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting; in these embodiments, the same reference numerals denote the same structures, wherein: Figure 1 This is a schematic diagram illustrating an application scenario of a recruitment platform according to some embodiments of the present invention; Figure 2 This is a block diagram of a construction worker recruitment system according to some embodiments of the present invention; Figure 3 This is an exemplary flowchart of a construction worker recruitment method according to some embodiments of the present invention; Figure 4 This is an exemplary flowchart illustrating the determination of a reference worker according to some embodiments of the present invention; and Figure 5 This is an exemplary flowchart illustrating the generation of assessment content according to some embodiments of the present invention. Detailed Implementation

[0011] To more clearly illustrate the technical solution of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are merely some examples or embodiments of the present invention. For those skilled in the art, these drawings can be applied to other similar scenarios without creative effort. Unless obvious from the context or otherwise specified, the same reference numerals in the drawings represent the same structures or operations.

[0012] It should be understood that the terms “system,” “device,” “unit,” and / or “module” used herein are one way to distinguish different components, elements, parts, sections, or assemblies at different levels. However, if other terms can achieve the same purpose, they may be replaced by other expressions.

[0013] Unless the context clearly indicates an exception, words such as "a," "an," "a kind," and / or "the" do not specifically refer to the singular and may also include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of explicitly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.

[0014] Flowcharts are used in this specification to illustrate the operations performed by the system according to the present invention. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, the steps can be processed in reverse order or simultaneously. Furthermore, other operations can be added to these processes, or one or more steps can be removed from them.

[0015] In the construction industry, traditional recruitment processes rely heavily on matching resumes with candidates and using a fixed question bank for assessment. The limitations of this process include at least the following: 1. Ignoring environmental variables: Traditional recruitment processes ignore the significant impact of company management style (e.g., frequency of internal restructuring and training, timeliness of salary payments) and project type (e.g., high-rise building projects, tunnel engineering projects) on candidate performance, leading to a disconnect between assessment results and actual job requirements. 2. Insufficient utilization of historical behavioral data: Large companies accumulate data (e.g., historical restructuring data) but this is not used in the recruitment process; while SMEs, due to their small size, lack data and cannot build effective assessment standards, relying heavily on subjective experience. 3. Rigid assessment and training content: Fixed assessment question banks cannot dynamically adapt to specific engineering risks (e.g., welding defects in nuclear power projects, hoisting errors in prefabricated buildings). 4. The training program is disconnected from the assessment results: those who scored low in the assessment did not receive targeted reinforcement training on frequently remedial issues; those who scored high in the assessment lacked advanced ability development based on the benchmarks of outstanding personnel (e.g., advanced ability training for high-difficulty tasks).

[0016] Based on this, the present invention provides a method and system for recruiting construction workers. This method and system fully utilize both the company's own data and data from external databases to develop recruitment assessment content more suited to the company's management style or specific scenarios, and uses the assessment results from the recruitment process to customize targeted training content for recruited personnel to improve their abilities.

[0017] Figure 1 This is a schematic diagram illustrating an application scenario of a recruitment platform according to some embodiments of the present invention. In some embodiments, such as... Figure 1 As shown, the recruitment platform 100 includes a processing device 110, a network 120, a storage device 130, and one or more terminal devices 140. In some embodiments, a construction worker recruitment method (e.g., processes 300-500) can be applied to the recruitment platform 100. For example, the processing device 110 can obtain user operation instructions, and in response to the operation instructions, retrieve the applicant's application information through a first terminal device (e.g., terminal device 140), determine the applicant's assessment content based on the applicant's application information (e.g., by calling a first determination module and a second determination module), and send it to the user (e.g., the applicant and / or recruiter).

[0018] Processing device 110 can access data or information from storage device 130 and / or terminal device 140 via network 120, or directly connect to storage device 130 and / or terminal device 140 to access information and / or data. Processing device 110 can process the accessed information and / or data. For example, processing device 110 can obtain job application information from terminal device 140 and, based on the application information, identify one or more candidate workers. As another example, processing device 110 can determine the assessment criteria for job applicants based on the historical performance information of one or more candidate workers.

[0019] Network 120 can connect various components of recruitment platform 100 and / or connect recruitment platform 100 with external resources. In some embodiments, one or more components of recruitment platform 100 (e.g., processing device 110, storage device 130, and / or terminal device 140) can exchange information and / or data via network 120. For example, terminal device 140 can send job application information and / or target company information (e.g., target company information and project information) to processing device 110 via network 120. As another example, processing device 110 can send assessment content to terminal device 140 via network 120.

[0020] In some embodiments, network 120 can be any one or more of a wired network or a wireless network. In some embodiments, network 120 may include one or more network access points. For example, network 120 may include wired or wireless network access points (e.g., base stations and / or network switching points) through which one or more components of recruitment platform 100 can connect to network 120 to exchange data and / or information.

[0021] Storage device 130 refers to a device used to store data, instructions, and / or any other information. In some embodiments, storage device 130 may store data and / or information obtained from processing device 110 and / or terminal device 140, etc. For example, storage device 130 may store job application information obtained from terminal device 140. As another example, storage device 130 may store information about reference workers (e.g., reference worker resume information, skill level, etc.). As another example, storage device 130 may store company information, project information undertaken by the company, etc. As another example, storage device 130 may store workers' historical task data. In some embodiments, storage device 130 may include one or any combination of mass storage, removable memory, etc.

[0022] Terminal device 140 refers to one or more electronic terminals used by a user (e.g., personnel involved in the recruitment process). Personnel involved in the recruitment process include job applicants (e.g., applicants for jobs), recruitment interviewers, etc. In some embodiments, such as Figure 1 As shown, terminal device 140 may include, but is not limited to, smartphone 141, tablet computer 142, laptop computer 143, desktop computer 144, etc. In some embodiments, terminal device 140 can interact with other components in recruitment platform 100 via network 120. For example, terminal device 140 can send job application information to processing device 110 or storage device 130 via network 120. As another example, terminal device 140 can receive assessment content sent by processing device 110 via network 120.

[0023] In some embodiments, the recruitment platform 100 may include one or more terminal devices 140. For example, the recruitment platform 100 may include a first terminal device used by job applicants and a second terminal device used by recruiters / companies. Different job applicants may use the same or different first terminal devices; for example, each job applicant may have their own electronic terminal, or different job applicants may share the same electronic terminal. For example, job applicant A may use job applicant B's first terminal device to send application information and / or application requests. Similarly, different recruiters / companies may use the same or different second terminal devices.

[0024] Figure 2 This is a block diagram of a construction worker recruitment system according to some embodiments of the present invention. In some embodiments, such as Figure 2 As shown, the construction worker recruitment system 200 may include an acquisition module 210, a first determination module 220, and a second determination module 230. In some embodiments, the construction worker recruitment system 200 may be part of a recruitment platform 100, for example, integrated into a processing device 110.

[0025] The acquisition module 210 is used to retrieve the application information of job applicants through the first terminal device in response to user operation instructions. Users may include job applicants (e.g., job seekers) and / or recruiters (e.g., interviewers conducting recruitment assessments). Operation instructions may include at least one of a recruitment request, an application request, or an assessment content generation request. For example, the acquisition module 210 may retrieve the application information of a job applicant through the first terminal device in response to an application request initiated by the job applicant via the first terminal device. As another example, the acquisition module 210 may retrieve the application information of a job applicant through the first terminal device in response to a recruitment request and / or an assessment content generation request initiated by a recruiter through a second terminal device.

[0026] The first determination module 220 is used to determine one or more reference workers based on the target job position and / or resume information.

[0027] In some embodiments, the first determining module 220 is further configured to determine multiple candidate workers from the target company based on resume information; and to determine the candidate workers with work experience in the target position among the multiple candidate workers as reference workers based on the target position.

[0028] In some embodiments, the first determining module 220 is further configured to obtain company information of the target company and project information undertaken through the second terminal device; construct a virtual work scenario based on the target position, the company information and project information; and select one or more reference workers from the worker database based on the virtual work scenario.

[0029] In some embodiments, the first determining module 220 is further configured to determine a job vector based on the job description information of the target job; determine a company vector based on the proportion of workers, number of job types, rectification frequency, and payroll timeliness in the company information; determine a project vector based on the project type, project cycle, and current project stage in the project information; and determine a virtual scenario vector based on the job vector, company vector, and project vector.

[0030] In some embodiments, the first determining module 220 is further configured to retrieve historical task data of at least one worker from the worker database; determine a historical scenario vector corresponding to the historical task data for each of the at least one worker; and determine one or more reference workers based on the similarity between the historical scenario vector and the virtual scenario vector.

[0031] The second determination module 230 is used to determine the assessment content of applicants based on the historical rectification information of one or more reference workers.

[0032] In some embodiments, the second determining module 230 is further configured to determine the target rectification task from the historical rectification information of one or more reference workers based on the virtual scenario vector; and generate assessment content based on the information record of the target rectification task.

[0033] In some embodiments, the construction worker recruitment system 200 also includes a generation module 240.

[0034] The generation module 240 is used to obtain the assessment results of the applicants through the first terminal device; in response to the assessment results indicating that the assessment has been passed, it generates training content for the applicants.

[0035] In some embodiments, the generation module 240 is further configured to generate training content based on the historical rectification information of one or more reference workers in response to the assessment result belonging to the first segment.

[0036] In some embodiments, the generation module 240 is further configured to, in response to the assessment result belonging to the second segment, determine a first group and a second group from one or more reference workers; and generate training content based on the differences between the historical rectification information corresponding to the first group and the second group.

[0037] For more information on the acquisition module 210, the first determination module 220, the second determination module 230, and the generation module 240, please refer to [link to documentation]. Figure 3 And its related descriptions.

[0038] It should be noted that the above description of the construction worker recruitment system 200 and its modules is for ease of description only and should not be construed as limiting the invention to the scope of the embodiments described. It is understood that those skilled in the art, after understanding the principles of the system, may arbitrarily combine the various modules or construct subsystems connected to other modules without departing from these principles. In some embodiments, Figure 2 The acquisition module 210, the first determination module 220, the second determination module 230, and the generation module 240 disclosed herein can be different modules within a single system, or a single module can implement the functions of two or more of the aforementioned modules. For example, the modules can share a single storage module, or each module can have its own separate storage module. As another example, the construction worker recruitment system 200 may also include a sending module for sending assessment content and / or training content to recruiters and / or applicants. Such variations are all within the scope of protection of this invention.

[0039] Figure 3 This is an exemplary flowchart illustrating a construction worker recruitment method according to some embodiments of the present invention. Figure 3 As shown, process 300 includes the following steps. In some embodiments, process 300 may be executed by recruitment platform 100 (e.g., processing device 110) or construction worker recruitment system 200.

[0040] Step S310: Retrieve the job application information of the applicants through the first terminal device. In some embodiments, step S310 is performed by the processing device 110.

[0041] In some embodiments, the recruitment platform 100 (e.g., processing device 110) can retrieve the job application information of applicants through a first terminal device in response to user operation instructions. In some embodiments, user operation instructions can be obtained through the recruitment platform 100, for example, by obtaining operation instructions input by the user through terminal device 140.

[0042] Users can include recruiters and / or job applicants. For example, a user can be a job applicant or a company's recruitment manager (e.g., an interviewer conducting a recruitment assessment).

[0043] Operation instructions can be requests initiated by users through terminal devices. For example, a user can access the webpage or application / mini-program of recruitment platform 100 through terminal device 140 and initiate a relevant request by clicking the relevant button in the webpage or application / mini-program. Operation instructions can include at least one of recruitment requests, application requests, assessment content generation requests, etc. For example, an applicant initiates an application request to the recruitment platform through a first terminal device, and a recruiter initiates a recruitment request and / or assessment content generation request to the recruitment platform through a second terminal device.

[0044] The first terminal device refers to the terminal device used by the job applicant. The second terminal device refers to the terminal device used by the recruiter.

[0045] In some embodiments, the recruitment platform 100 (e.g., processing device 110) may, in response to a user's operation instructions, retrieve the applicant's application information through a second terminal device or an internal database (e.g., storage device 130).

[0046] An applicant is someone who participates in the selection and assessment process of a recruiting party (such as a company or organization) in order to obtain a job. For example, an applicant could be a worker participating in the recruitment process of a construction company.

[0047] Application information refers to a collection of data related to an applicant's application process. Application information may include information about the recruiting party and / or the applicant. In some embodiments, application information includes the target position at the company the applicant is applying for and the applicant's resume information.

[0048] The target company refers to the enterprise or organization that the applicant is applying to.

[0049] The target position refers to the job or type of work that the applicant is applying for. For example, if the target company is a construction company, the target position could be a concrete pourer or a scaffolder.

[0050] In some embodiments, the target position includes the corresponding job level (e.g., manager level, team leader level, and general worker) and the specific position / job type. For example, the target position could be manager level - project manager, team leader level - construction team leader, general worker - scaffolder, etc.

[0051] A resume contains information reflecting an applicant's work experience and / or personal details. For example, a resume may include the applicant's basic information, skills and qualifications, skill level, and work experience.

[0052] Basic information refers to the applicant's basic personal information. For example, basic information may include the applicant's photo, name, gender, age, length of service, and physical condition (e.g., height, weight, whether physical indicators are normal, whether there are any diseases, etc.).

[0053] Skills qualification information refers to the qualifications that applicants possess related to their job skills. For example, skills qualification information may include the applicant's skilled worker qualifications (e.g., welder, electrician, lathe operator, fitter, etc.), special equipment (e.g., elevator, lifting machinery, special-purpose vehicle) operation qualifications, and special operation (e.g., electrical work, welding and thermal cutting, working at heights, scaffolding erection, special equipment installation / dismantling, etc.) qualifications. For instance, skills qualification information could be "possessing welding qualifications and working at heights qualifications."

[0054] Skill level reflects the level of skills corresponding to skill qualifications. For example, skill levels can include basic, intermediate, and advanced, representing an increasing level of skill among job applicants.

[0055] Work experience reflects information about the jobs an applicant has participated in. For example, work experience may include the companies the applicant has worked for, the types of projects they have participated in (e.g., building construction projects, bridge construction projects, hydropower / nuclear power plant construction projects, etc.), the position / job type they held in the projects, and the duration of hands-on work.

[0056] In some embodiments, the job application information may be filled out by the applicant through the first terminal device (e.g., terminal device 140) they use, and the recruitment platform 100 (e.g., processing device 110) retrieves the applicant's job application information through the first terminal device.

[0057] In some embodiments, the recruitment platform 100 may communicate with a highly trusted platform (e.g., a government-operated real-name human resources management platform) and retrieve and collect the application information of current applicants from the data filed on the aforementioned platform.

[0058] Step S320: The first determining module is invoked to determine one or more reference workers based on the target job position and / or resume information. In some embodiments, step S320 is performed by the processing device 110.

[0059] Reference workers refer to workers whose information is similar to that of the job applicants and whose historical data (e.g., historical rectification information) is complete.

[0060] In some embodiments, the recruitment platform 100 may determine one or more reference workers from the worker database by invoking a first determining module (e.g., first determining module 220).

[0061] A worker database refers to a database that stores worker information. The worker database may be configured in a storage device (e.g., storage device 130). The worker database may store worker information for multiple workers (hereinafter referred to as current workers), who are currently employed by or have previously been employed by multiple companies, including the target company.

[0062] Worker information refers to the relevant information about existing workers stored in the worker database. Worker information may include the company (the company the current worker is currently employed by or has previously employed by), job position (the current position the current worker holds or the position they held in a company they have left), and resume information and / or historical reassignment information. Similar to resume information, resume information may include basic information about existing workers, skills and qualifications, skill level, work experience, etc. An explanation of historical reassignment information is provided below.

[0063] Understandably, a worker's information may include more than one company and one job title. For example, a worker may currently work for company A as a scaffolder, and may have previously worked for companies B and C as a concrete pourer. In this case, the worker's information would include companies A, B, and C, and job titles including scaffolder and concrete pourer.

[0064] It should be noted that the worker database can store worker information for multiple existing workers. These existing workers may belong to the target company, or to other companies or related platforms. The employment status of these existing workers can be either currently employed or retired (i.e., information on retired workers can still be retained in the worker database).

[0065] In some embodiments, the recruitment platform 100 may determine reference workers by invoking the first determining module 220, based on the applicant's target position and / or resume information, as well as the worker information of existing workers.

[0066] In some embodiments, the recruitment platform 100 can, by invoking the first determining module 220, identify existing workers whose job positions are the same as the applicant's target positions and whose resume information is the same or partially the same as the applicant's resume information as reference workers. For example, the first determining module 220 can identify existing workers whose job positions are the same as the applicant's target positions as reference workers. For instance, taking the applicant's target position as a concrete pourer as an example, the first determining module 220 can identify existing workers whose job positions include concrete pourer (which may be the current position or a previously held position) as reference workers. As another example, the first determining module 220 can identify existing workers whose job positions are the same as the applicant's target positions and whose skill qualification information and skill level are respectively the same as the applicant's skill qualification information and skill level as reference workers. For example, if the applicant's target position is scaffolder, the skill qualification information is "possessing high-altitude operation qualification", and the skill level is intermediate, then the first determination module 220 can determine the existing workers among multiple existing workers whose job positions include scaffolder (which may be the current position or a position they have held), who possess high-altitude operation qualification, and whose skill level is intermediate as reference workers.

[0067] In some embodiments, the first determining module 220 determines multiple candidate workers from the target company based on resume information.

[0068] Candidate workers refer to existing workers of the target company (including those currently employed by the target company and those who have previously worked for the target company but have now left) whose key information in their resumes (hereinafter referred to as secondary key information) is similar to the key information in the resumes of job applicants (hereinafter referred to as primary key information).

[0069] Key information refers to relatively important information in a resume or professional background report. For example, the first key information may include the applicant's length of service, skill qualifications, skill level, and / or the types of projects they have participated in, as found in their resume. Conversely, the second key information may include the same information in an existing worker's professional background report. Which information in the resume or professional background report is considered key information can be pre-selected by the recruiter (e.g., HR personnel of the target company) or automatically determined by the first determination module 220 based on the target position.

[0070] The aforementioned similarity refers to the first key information and the second key information meeting a first preset condition. The first preset condition includes that the similarity between the second key information and the first key information (hereinafter referred to as the first similarity) is not lower than a first similarity threshold. The first similarity refers to the proportion of information with identical content in the first key information and the second key information to all key information.

[0071] For example, taking a scenario where both the first and second key information include years of service, skill qualification information, skill level, and types of projects participated in (the number of key information items is 4), and the first similarity threshold is 50%, if applicant A's first key information includes: 20 years of service, skill qualification information "possesses welding qualification", skill level is primary, and the type of projects participated in is building construction projects; and existing worker B's second key information includes: 20 years of service, skill qualification information "possesses welding qualification", skill level is intermediate, and the type of projects participated in is building construction projects, then existing worker B's second key information contains 3 items that are the same as applicant A. Therefore, the first similarity between existing worker B's second key information and the first key information is 3 / 4 = 75% > 50%, and existing worker B can be selected as a candidate worker.

[0072] As another example, if worker C has 15 years of service, his skill qualification information is "possesses electrician qualification", his skill level is junior, and the type of project he has participated in is bridge construction project, then in worker C's second key information, there is one piece of information that is the same as applicant A. Therefore, worker C's first similarity with the first key information is 1 / 4 = 25% < 50%, and worker C cannot be selected as a candidate worker.

[0073] In some embodiments, the recruitment platform 100 may, by invoking the first determining module 220, determine one or more candidate workers with work experience in the target position among a plurality of candidate workers as one or more reference workers based on the target position.

[0074] For example, taking the target position as scaffolder as an example, the first determining module 220 can determine the candidate worker whose job position in the worker information includes scaffolder as the reference worker from among multiple candidate workers.

[0075] In some embodiments, the historical rectification information in the worker information of the reference workers is not empty. For example, after determining multiple reference workers based on the method described above, the first determining module 220 can eliminate reference workers whose historical rectification information is empty and determine the remaining reference workers as the final reference workers. Further explanation regarding historical rectification information can be found in the relevant description below.

[0076] In some embodiments of the present invention, by limiting the reference workers to the target company, the assessment criteria are ensured to match the actual management requirements of the target company, thereby making the determined assessment content more targeted.

[0077] In some embodiments, the recruitment platform 100 obtains company information and project information of the target company; constructs a virtual work scenario based on the target position, company information, and project information; and selects one or more reference workers from the worker database based on the virtual work scenario. Further details regarding this embodiment can be found in [link to relevant documentation]. Figure 4 And its related descriptions.

[0078] Step S330 involves invoking the second determining module to determine the assessment criteria for the applicant based on historical rectification information from one or more reference workers. In some embodiments, step S330 is executed by the processing device 110.

[0079] Historical rectification information refers to the recorded data on the rectification actions of existing workers. Rectification refers to rework required of existing workers (e.g., rework due to substandard construction quality or building appearance not meeting aesthetic requirements). Historical rectification information may include rectification time, rectification issue, rectification task, and rectification result. Rectification time includes the time when workers began rework and the estimated time for rework; rectification issue refers to the problem requiring rectification, such as deviation in the position of erected scaffolding or incorrect rebar spacing; rectification task refers to the task of resolving the rectification issue, such as adjusting the scaffolding position or adjusting the rebar spacing; rectification result includes whether the rectification issue has been resolved or not.

[0080] It is understandable that some existing workers may never have been asked to rework, therefore the above information regarding the historical rectification of these existing workers can be empty.

[0081] In some embodiments, when the historical rectification information of existing workers is included as part of the worker information, the historical rectification information can be stored together with the worker information in the worker database.

[0082] In some embodiments, historical rectification information of existing workers can be stored separately in the rectification database.

[0083] The rectification database refers to a database that stores historical rectification information. The rectification database can be configured in a storage device (e.g., storage device 130). The rectification database and the worker database can be two databases in the same storage device, or databases in two different storage devices, or the rectification database can be a sub-database of the worker database. The rectification database can store historical rectification information for multiple existing workers. The recruitment platform 100 can invoke the second determination module 230, which communicates with the storage device (e.g., storage device 130) where the rectification database is located (e.g., via network 120) to obtain the historical rectification information for each reference worker in the rectification database.

[0084] The assessment content refers to the test content used to evaluate whether the applicant is competent for the target position. The assessment content may include written test, interview and / or practical skills assessment.

[0085] Written examination refers to assessing applicants through paper-based interviews to obtain assessment results (e.g., test scores).

[0086] Interview assessment refers to the process of evaluating applicants through questions asked by interviewers (e.g., recruiters or project managers from the target company) and answers given by the applicants. The assessment result is the interviewer's evaluation of the applicant's answers.

[0087] Practical skills assessment refers to evaluating the applicant's actual operational abilities (e.g., pouring concrete, erecting scaffolding), and the assessment result is the examiner's evaluation of the operational results (e.g., concrete pouring situation, scaffolding erection quality / speed).

[0088] Written tests and / or interviews can be conducted online or offline.

[0089] In some embodiments, the second determining module 230 determines the assessment content of the applicant based on the rectification tasks in the historical rectification information of one or more reference workers.

[0090] In some embodiments, the second determining module 230 may determine the target worker from one or more reference workers, and determine the assessment content of the applicant based on the rectification tasks in the target worker's historical rectification information.

[0091] The target worker refers to the reference worker (one or more) used to determine the assessment content. For example, the second determination module 230 can determine the reference worker with the highest similarity between the second key information and the first key information among one or more reference workers as the target worker, or determine the reference workers in the top N positions after sorting the first similarity from largest to smallest, or determine the reference workers whose first similarity is greater than a preset threshold (greater than the first similarity threshold) as the target worker. For an explanation of the first key information, second key information, and first similarity, please refer to the previous text.

[0092] For example, assuming the rectification task in the target worker's historical rectification information is "adjusting the scaffolding position", the assessment content can be a practical assessment, specifically including scaffolding erection or scaffolding position adjustment, or the assessment content can be an interview assessment, specifically including the examiner asking the applicant questions about the scaffolding erection / adjustment process and precautions, and the applicant answering in person, or the assessment content includes the aforementioned practical assessment and interview assessment.

[0093] In some embodiments, the second determining module 230 determines the target rectification task from the historical rectification information of one or more reference workers based on a virtual scenario vector; and generates assessment content based on the information record of the target rectification task. Further details regarding this embodiment can be found in [link to relevant documentation]. Figure 5 And its related descriptions.

[0094] In some embodiments, such as Figure 3 As shown, process 300 further includes steps S340 and S350. In some embodiments, steps S340 and S350 are performed by processing device 110.

[0095] Step S340: Obtain the assessment results of the applicants through the first terminal device.

[0096] The assessment result refers to the evaluation of an applicant's performance after completing the assessment content. The assessment result can be represented by an assessment score; a higher score indicates better performance. For example, the assessment score can be a value within the range of [0, 100]. The assessment result can also be represented by an assessment rating, such as Fail, Pass, Good, and Excellent, representing progressively better performance.

[0097] In some embodiments, after the assessment is completed, the applicant can upload the assessment results to the recruitment platform 100 via a first terminal device. In some embodiments, after the assessment is completed, the examiner can upload the assessment results to the recruitment platform 100 via a second terminal device (e.g., terminal device 140). The processing device 110 or the generation module 240 can directly receive the assessment results sent by the first terminal device or the second terminal device, or retrieve the assessment results from the storage device.

[0098] Step S350: In response to the assessment result indicating that the assessment has been passed, training content is generated for the applicant.

[0099] The assessment indicates whether the applicant's performance is satisfactory. In some embodiments, the recruitment platform 100 generates training content for the applicant by calling the generation module 240.

[0100] In some embodiments, the generation module 240 can determine whether the assessment result meets the third preset condition and determine whether the applicant has passed the assessment.

[0101] The third preset condition can be that the assessment score is greater than a score threshold, such as 60, 80 or 90.

[0102] The third prerequisite can also be that the performance evaluation rating is not lower than the preset rating. For example, if the preset rating is "pass", then if the applicant's performance evaluation rating is pass, good, or excellent, then the applicant passes the evaluation; if the applicant's performance evaluation rating is fail, then the applicant fails the evaluation.

[0103] Training content refers to the pre-employment training provided to applicants who have passed the assessment. For example, training content may include rebar tying training, scaffolding erection training, and concrete pouring training.

[0104] In some embodiments, the generation module 240 can develop different training content for applicants with different assessment results. For example, for applicants with a passing or good assessment rating, the training content corresponds to the assessment content (e.g., if the assessment content is "rebar tying," then the training content is also "rebar tying," to improve the applicant's operational skills in rebar tying and reduce the probability of errors). Alternatively, for applicants with an excellent assessment rating, the training content is empty (i.e., no training is required for applicants with excellent assessment ratings).

[0105] In some embodiments, in response to the assessment result belonging to the first segment, the generation module 240 generates training content based on the historical rectification information of one or more reference workers.

[0106] The first segment refers to the range of results to which the assessment results belong.

[0107] For example, if the assessment result is represented by an assessment score, then the first segment could be [60, 80]. Based on this, if an applicant's assessment score is 64 (within the range of [60, 80]), then the applicant's assessment result belongs to the first segment.

[0108] For example, if the assessment results are represented by assessment scores, the first segment can include "pass" and "good". Based on this, if an applicant's assessment rating is "pass", then the applicant's assessment result belongs to the first segment.

[0109] In some embodiments, the generation module 240 obtains high-frequency rectification issues based on the historical rectification information of one or more reference workers, and determines training content based on the rectification tasks corresponding to the high-frequency rectification issues.

[0110] High-frequency rectification issues refer to rectification issues that occur at least once or twice. For example, with a frequency threshold of 5, if rectification issue D occurs twice, rectification issue E occurs 7 times, rectification issue F occurs 6 times, and rectification issue G occurs 3 times in all historical rectification information of one or more reference workers, then rectification issues E and F will be considered high-frequency rectification issues.

[0111] For example, if the frequently rectified issues are "incorrect spacing of rebar tying" and "incorrect selection of rebar type", then the generated training content will be rebar type identification and rebar tying training.

[0112] In some embodiments, in response to an assessment result belonging to a second segment, the processing device 110 or the generation module 240 determines a first group and a second group from one or more reference workers.

[0113] The second segment refers to the range of results that are higher than the first segment.

[0114] For example, if the assessment result is represented by an assessment score and the first segment is [60, 80], then the second segment could be (80, 100). Based on this, if an applicant's assessment score is 87 (within the range of (80, 100)), then the applicant's assessment result belongs to the second segment.

[0115] For example, if the assessment results are represented by assessment scores and the first segment includes "pass" and "good," then the second segment could be "excellent." Therefore, if an applicant's assessment rating is "excellent," then that applicant's assessment result belongs to the second segment.

[0116] The first and second groups refer to the different grouping results for the reference workers. The first group includes workers whose overall rectification rate is below the threshold, and the second group includes workers whose work experience in the target position has no history of rectification and whose overall rectification rate is above the threshold (hereinafter referred to as the rectification rate threshold).

[0117] The overall rectification rate refers to the percentage of all construction tasks completed by a worker (e.g., erecting scaffolding in project A, pouring concrete in project B, etc.) that require rectification. For example, assuming a worker completed 80 construction tasks, of which 15 were required to be rectified, the worker's overall rectification rate is 15 / 80 × 100% = 18.75%.

[0118] In some embodiments, the generation module 240 generates training content based on the differences between the historical rectification information corresponding to the first group and the second group.

[0119] In some embodiments, historical rectification information may also include the overall rectification rate, the recurrence rate of rectified issues, and the pass rate for high-difficulty tasks. Therefore, the aforementioned differences include differences in the overall rectification rate, the recurrence rate of rectified issues, and the pass rate for high-difficulty tasks.

[0120] In some embodiments, historical rectification information may further include any one or two of the following: overall rectification rate, recurrence rate of rectified issues, and pass rate for high-difficulty tasks. Based on this, the aforementioned differences include any one or two of the following: differences in overall rectification rate, differences in recurrence rate of rectified issues, and differences in pass rate for high-difficulty tasks.

[0121] In some embodiments, historical rectification information may include other information (e.g., the average time it takes for workers to complete rectification tasks), without limitation.

[0122] The recurrence rate of rectified problems refers to the percentage of construction tasks that reproduce rectified problems out of all construction tasks completed by a worker. A reproduced rectified problem means that after a worker reworks a rectified problem, the same problem reappears (i.e., the rectification is incomplete). For example, suppose a worker completes 80 construction tasks, and 6 of them reproduce rectified problems. Then, the worker's recurrence rate of rectified problems is 6 / 80 × 100% = 7.5%.

[0123] The pass rate for high-difficulty tasks refers to the probability that a worker completes a high-difficulty task and that the task is not required to be rectified. For example, suppose a worker completes 16 high-difficulty tasks, of which 3 require rectification due to rectification issues. Then the worker's pass rate for high-difficulty tasks is (16-3) / 16×100%=81.25%.

[0124] The difference between the overall rectification rates of the first group and the second group can be represented by the difference between the mean of the overall rectification rates of all reference workers in the first group and the mean of the overall rectification rates of all reference workers in the second group (hereinafter referred to as the first difference). For example, assuming that the mean of the overall rectification rate of all reference workers in the first group is 15% and the mean of the overall rectification rate of all reference workers in the second group is 38%, then the difference between the overall rectification rates of the first group and the second group is 38% - 15% = 23%.

[0125] Similar to the difference in the overall rectification rate, the difference in the recurrence rate of rectified problems between the first group and the second group can be characterized by the difference between the mean recurrence rate of rectified problems of each reference worker in the first group and the mean recurrence rate of rectified problems of each reference worker in the second group (hereinafter referred to as the second difference); the difference in the pass rate of high-difficulty tasks between the first group and the second group can be characterized by the difference between the mean pass rate of high-difficulty tasks of each reference worker in the first group and the mean pass rate of high-difficulty tasks of each reference worker in the second group (hereinafter referred to as the third difference).

[0126] In some embodiments, the generation module 240 determines training content based on the magnitude relationship between the first difference, the second difference, and the third difference, including: If the largest of the first, second, and third differences is the first difference, then the training content focuses on the stability and comprehensiveness of construction quality. For example, the training content includes comprehensive learning of construction specifications (covering the target trades), refined operation training, and cross-trade collaboration processes.

[0127] If the largest of the first, second, and third differences is the second difference, then the training content focuses on cultivating the ability to analyze the root causes of construction problems and the ability to solve problems. For example, the training content includes explanations of problem examples, demonstrations of correct and incorrect operating procedures, etc.

[0128] If the largest of the first, second, and third differences is the third difference, then the training content focuses on the ability to handle high-difficulty tasks or complex working conditions. For example, the training content includes training on high-difficulty tasks and problem-solving procedures for problems that arise under complex working conditions.

[0129] In some embodiments of the present invention, the assessment results are used as an important reference source for training content, enabling training to specifically address the shortcomings revealed by applicants during the assessment. By providing intensive training on issues corresponding to historical rectification information, applicants can fully understand potential risks and master correct operating methods and preventive measures before starting work, thereby minimizing the likelihood of them repeating the mistakes of the reference workers in future work. By developing targeted training content for applicants with high assessment scores, high-scoring applicants can advance beyond their target positions and handle more and more challenging tasks.

[0130] In some embodiments of the present invention, the assessment content is determined by selecting reference workers similar to the applicants and extracting historical rectification information, rather than using a general question bank. This makes the assessment content more targeted, avoids being out of touch with job requirements, and reduces the mismatch rate between applicants and positions. Rectification information, which is typically remedial after the fact, is transformed into information for pre-screening, allowing potential risks to be identified during the recruitment stage. This reduces the likelihood of applicants exhibiting similar problems after joining the company, thereby reducing rework, delays, and quality issues caused by similar problems, and optimizing construction quality and efficiency. When determining the assessment content, historical rectification information from multiple reference workers is used for information complementarity, ensuring the completeness and diversity of the assessment content and avoiding poor quality assessment content due to a lack of data or information silos.

[0131] It should be noted that the above description of process 300 is merely for illustration and explanation, and does not limit the scope of the invention. Those skilled in the art can make various modifications and changes to process 300 under the guidance of this specification. However, these modifications and changes are still within the scope of the invention.

[0132] Figure 4 This is an exemplary flowchart illustrating the determination of a reference worker according to some embodiments of the present invention. Figure 4 As shown, process 400 includes the following steps. In some embodiments, process 400 may be executed by a first determining module 220.

[0133] Step S410: Obtain the target company's company information and the project information it undertakes through the second terminal device.

[0134] Company information reflects the characteristics of a company. For example, company information includes the company's business scope, management system (including attendance system, performance system, promotion system, etc.), and worker job distribution (the categories of jobs included in the company, the number of workers in each job, the proportion of workers in each job to the total number of workers, etc.).

[0135] The company's business scope refers to the types of projects or engineering works that the company can undertake. For example, the company's business scope includes the types of projects that the company can undertake (e.g., building construction projects, bridge construction projects, tunnel excavation projects, etc.).

[0136] Project information reflects the engineering characteristics of the projects the company has undertaken, including their construction difficulty and quality requirements. Project information includes the project type (e.g., building construction, bridge construction, tunnel excavation), project duration, and quality requirements. The project duration refers to the total estimated time from project commencement to acceptance; for example, it could be 1 year, 2 years, or 5 years. Quality requirements refer to the quality standards that the constructed buildings (e.g., buildings in a building construction project) must meet, such as seismic resistance, wind resistance, and fire resistance ratings.

[0137] In some embodiments, the first determining module 220 may obtain the target company's company information and the project information it undertakes through a second terminal device (e.g., a terminal device used by the target company).

[0138] In some embodiments, the first determining module 220 may obtain the target company's company information through the target company's internal management system (e.g., human resources management system, financial management system, etc.) or a third-party recruitment platform.

[0139] In some embodiments, the first determining module 220 may obtain project information undertaken by the target company through the internal management system (e.g., project management system, etc.) or government system (e.g., government procurement network, government bidding information platform, construction market supervision platform) used by the target company.

[0140] Step S420: Based on the target position, company information, and project information, construct a virtual work scenario.

[0141] A virtual work scenario refers to a virtual construction environment that closely resembles the actual work environment of the target position. A virtual construction scenario can comprehensively represent the target position, company information, and project information.

[0142] In some embodiments, the virtual work scenario can be described in text form. For example, the virtual work scenario could be "The project is a bridge construction project, located in a mountainous area at an altitude of 300 meters, the job is welder, the project duration is 3 years, the bridge can withstand an earthquake of magnitude 8, ...".

[0143] In some embodiments, virtual work scenarios can be represented in vector form. In some embodiments, virtual work scenarios are based on virtual scenario vector representations.

[0144] Virtual scenario vectors are a digital, structured representation of positional features used to describe the work scenario of a specific job in a specific company environment and project conditions.

[0145] In some embodiments, the first determining module 220 can determine the virtual scenario vector in a variety of ways based on the target position, company information, and project information.

[0146] In some embodiments, before constructing a virtual work scenario based on target job position, company information, and project information (e.g., before determining the virtual scenario vector), the processing device 110 or the first determining module 220 may filter the project information and use the filtered target project information to construct the virtual work scenario.

[0147] In some embodiments, the first determining module 220 can filter target project information from the project information undertaken by the target company to construct a virtual work scenario. A target project refers to a project involving jobs corresponding to the target position. For example, taking a scaffolder as the target position, if project A involves jobs such as concrete pouring, scaffolding, and electrician, then project A is a target project, and its project information can be used to construct a virtual work scenario; if project B involves jobs such as concrete pouring, welding, and electrician (excluding scaffolding), then project B is not a target project, and its project information is not used to construct a virtual work scenario.

[0148] In some embodiments, the first determining module 220 uses a trained first vector determining model to process the target job, company information, and project information to obtain a virtual scenario vector. The first vector determining model is a machine learning model (e.g., BERT (Bidirectional Encoder Representations from Transformers) model), and the input of the first vector determining model includes the target job, company information, and target project information, while the output includes the virtual scenario vector.

[0149] In some embodiments, the first determining module 220 determines a job vector based on the job description information of the target job; determines a company vector based on the proportion of workers, number of job types, rectification frequency, and payroll timeliness in the company information; determines a project vector based on the project type, project cycle, and current project stage in the project information; and determines a virtual scenario vector based on the job vector, company vector, and project vector.

[0150] Job description information refers to data that describes the characteristics of the target job. For example, job description information could be "The job is scaffolder, with 26 working days per month, 8 working hours per day, and a monthly salary of 15,000 yuan, etc."

[0151] A job vector is a numerical vector obtained by encoding the job description information of a target job.

[0152] In some embodiments, the first determining module 220 can use the feature values ​​corresponding to each field appearing in the job description information as the values ​​of each element of the company vector to obtain the job vector. For example, taking the job description information as "job type is scaffolder, working 26 days a month, working hours 8 hours a day, monthly salary is 15,000 yuan / month" as an example, the job vector is (2, 26, 8, 15,000), where the feature values ​​corresponding to the number of working days per month, the number of working hours per day, and the monthly salary are equal to their values. The feature values ​​corresponding to the target job can be preset. For example, the feature value of a concrete pourer is preset to 1, the feature value of a scaffolder is preset to 2, the feature value of a welder is preset to 3, and so on.

[0153] In some embodiments, the first determining module 220 processes the job description information of the target job using a trained second vector determining model to obtain a job vector. The second vector determining model is a machine learning model (e.g., BERT model, Sentence-BERT model, etc.), and its input includes job description information, while its output includes the job vector. In some embodiments, the first vector determining model and the second vector determining model can be the same model.

[0154] The worker percentage refers to the proportion of the target company's total number of workers (the total number of workers in each position / job type) to the total number of employees in the target company. For example, if the target company employs 800 workers and 200 non-worker personnel (e.g., HR personnel, finance personnel, sales personnel, etc.), then the worker percentage is 800 / (200+800)×100%=80%.

[0155] The number of job types reflects the number of different job types a worker can do. For example, if the target company employs 25 concrete pourers, 22 scaffolders, 13 welders, 10 electricians, and 30 carpenters, then the number of job types is 5.

[0156] In some embodiments, the first determining module 220 can normalize the value of the number of job types to obtain a normalized number of job types. For example, the normalization process can be to divide the value of the number of job types by the total number of all job types in the construction industry, and use the result as the normalized number of job types. For example, if the number of job types is 5 and the total number of job types for all workers in the construction industry is 50, then the normalized number of job types is 5 / 50 = 0.1.

[0157] The rectification frequency refers to the total number of times all workers of the target company are required to make rectifications within a unit of time (e.g., daily or weekly). For example, if the target company required 3 workers to make rectifications in the past month (calculated as 30 days), and the number of rectifications for the three workers were 2, 3, and 4 respectively, then the rectification frequency is (2+3+4) / 30=0.3.

[0158] In some embodiments, the first determining module 220 can normalize the value of the rectification frequency to obtain a normalized rectification frequency. For example, the normalization process can be the result of performing a mathematical operation (e.g., taking the logarithm) on the rectification frequency as the normalized rectification frequency. For example, taking a rectification frequency of 0.3 as an example, the normalized rectification frequency can be ln(0.3+1)=ln1.3.

[0159] Timeliness of payroll reflects the probability that the target company pays salaries on time. For example, if the target company pays salaries 40 times in total, of which 39 times are timely (payroll date is before the planned payroll date) and 1 time is delayed (payroll date is after the planned payroll date), then the timeliness of payroll is 39 / 40 × 100% = 97.5%.

[0160] A company vector is a numerical vector obtained by encoding the company information of a target company.

[0161] In some embodiments, the first determining module 220 can use the worker percentage, the normalized number of job types, the normalized rectification frequency, and the timeliness of payroll as the values ​​of each element of the company vector to obtain the company vector. For example, taking a worker percentage of 80%, a normalized number of job types of 0.1, a normalized rectification frequency of ln1.3, and a payroll timeliness of 97.5% as an example, the company vector would be (0.8, 0.1, ln1.3, 0.975).

[0162] The current project stage refers to the current construction phase of the target project. For example, if the target project is a building construction project, the current project stage could be the foundation stage, the construction stage, or the topping-out stage.

[0163] A project vector is a numerical vector obtained by encoding the project information (also known as the target project information) of a target project.

[0164] In some embodiments, the first determining module 220 determines the feature values ​​corresponding to the project type, project cycle, and current project stage by looking up a preset table, and uses the feature values ​​corresponding to the project type, project cycle, and current project stage as the values ​​of each element of the project vector to obtain the company vector.

[0165] The preset table contains characteristic values ​​corresponding to various project types, project durations, and project stages. For example, the characteristic values ​​for building construction projects, bridge construction projects, and tunnel excavation projects are 1, 2, and 3, respectively. Similarly, the characteristic values ​​for project durations of 1 year, 2 years, and 3 years are 1, 2, and 3, respectively.

[0166] For example, assuming the project type, project cycle, and feature values ​​corresponding to the current project stage of the target project are 2, 5, and 4 respectively, the resulting project vector is (2, 5, 4).

[0167] In some embodiments, the first determining module 220 can encode the project type, project cycle, and current project stage using the same or different encoding methods, and merge the encoding results to obtain a project vector. For example, the processing device 110 or the first determining module 220 can perform one-hot encoding on the project type, segmented encoding on the project cycle, and compressed encoding on the current project stage, assuming the obtained encoding results are a1a2a3…a k b1b2b3…b m c1c2c3…c n The resulting item vector after merging is (a1a2a3…a ... k b1b2b3…b m ,c1c2c3…c n ).

[0168] In some embodiments, the first determining module 220 fuses the job vector, company vector, and project vector to obtain a virtual scenario vector.

[0169] In some embodiments, the fusion method can be concatenation fusion, that is, concatenating the first and last elements of the job vector, company vector, and project vector to obtain a virtual scenario vector. For example, taking the job vector as (2, 26, 8, 15000), the company vector as (0.8, 0.1, ln1.3, 0.975), and the project vector as (2, 5, 4), the virtual scenario vector obtained by concatenation fusion is (2, 26, 8, 15000, 0.8, 0.1, ln1.3, 0.975, 2, 5, 4).

[0170] In some embodiments, the fusion method may be weighted fusion.

[0171] The weighted fusion includes: performing mathematical operations (e.g., multiplication) on the values ​​of each element of the job vector, company vector, and project vector based on their corresponding weight values, and using the result as the new element values ​​to obtain job vectors, company vectors, and project vectors with new element values; and concatenating and fusing the new job vectors, company vectors, and project vectors to obtain a virtual scenario vector.

[0172] For example, assuming the job vector is (2, 26, 8, 15000), the weight value of the job vector is 0.5, and the mathematical operation method is multiplication, then the job vector with new element values ​​is (2×0.5, 26×0.5, 8×0.5, 15000×0.5), which is (1, 13, 4, 7500).

[0173] In some embodiments, the first determining module 220 processes the job vector, company vector, and project vector using a trained third vector determining model to obtain a virtual scenario vector. The third vector determining model is a machine learning model (e.g., a neural network model). The input to the third vector determining model includes the job vector, company vector, and project vector; alternatively, the input includes job description information of the target job, company information of the target company, and information of the target project. The output of the third vector determining model includes the virtual scenario vector.

[0174] In some embodiments, the first determining module 220 may also determine the virtual working scenario through other methods, which are not limited here.

[0175] In some embodiments of the present invention, by converting target positions, company information, and project information into computable vectors, the data type barrier is broken, providing data support for screening reference workers through virtual work scenarios.

[0176] Step S430: Based on the virtual work scenario, select one or more reference workers from the worker database.

[0177] In some embodiments, the first determining module 220 acquires historical task data of at least one worker (i.e., at least one existing worker) from the worker database; determines the similarity (hereinafter referred to as the second similarity) between the historical task data of the at least one worker and the current task data of the job applicant (including the target position, company information of the target company, and project information of the target project); and determines the existing worker corresponding to the historical task data with a second similarity higher than the second similarity threshold as a reference worker. The second similarity threshold can be preset, for example, the second similarity threshold can be preset to 0.8, 0.85, or 0.9, etc.

[0178] Historical task data refers to data related to tasks that current workers have participated in. For example, historical task data may include the companies that current workers have previously worked for, the types of projects they have participated in, and their previous job positions.

[0179] The second similarity reflects the degree of similarity between the historical task data of existing workers and the current task data of job applicants. Understandably, the value range of the second similarity between the historical task data of existing workers and the current task data of job applicants is [0,1].

[0180] In some embodiments, the first determining module 220 processes the historical task data of at least one existing worker and the current task data of a job applicant using a trained first similarity model to obtain a second similarity. The first similarity model is a machine learning model (e.g., a deep learning model), and the input of the first similarity model includes the historical task data of the existing worker and the current task data of the job applicant, and the output includes the second similarity between the historical task data of the existing worker and the current task data of the job applicant.

[0181] In some embodiments, the first determining module 220 may communicate with the storage device (e.g., storage device 130) where the worker database is located (e.g., via network 120) to obtain data (e.g., worker information) from the worker database.

[0182] It should be noted that the aforementioned at least one worker (at least one existing worker) can be all existing workers in the worker database, or it can be a portion of the existing workers in the worker database (to save computational costs). For example, the aforementioned at least one worker can be randomly selected from all existing workers in the worker database, or selected according to preset criteria, and its number can be 20% of the total number of existing workers in the worker database.

[0183] Furthermore, it should be noted that since each existing worker may have worked for multiple companies, participated in multiple projects, or held multiple positions, each existing worker may have multiple historical task data entries (each historical task data entry corresponds to a different company, project, or position). Based on this, when determining the second similarity between each existing worker's historical task data and the applicant's current task data, the second similarity between each of the worker's historical task data entries and the applicant's current task data can be determined separately, resulting in multiple second similarity scores. The maximum or average of these multiple second similarity scores is then selected as the second similarity between the existing worker's historical task data and the applicant's current task data.

[0184] In some embodiments, the first determining module 220 retrieves historical task data of at least one worker from the worker database; for each of the at least one worker, it determines the historical scenario vector corresponding to the historical task data; and based on the similarity between the historical scenario vector and the virtual scenario vector (hereinafter referred to as the third similarity), it determines one or more reference workers.

[0185] Similar to virtual scenario vectors, historical scenario vectors are also a type of digital, structured representation of positional features. Historical scenario vectors are used to describe the historical work scenarios of current workers in a specific historical position within a historical company environment and under historical project conditions.

[0186] In some embodiments, the first determining module 220 may determine the historical scenario vector based on historical task data (including the companies where current workers have historically worked, the types of projects they have historically participated in, and their historical positions) using a method similar to that used to determine the virtual scenario vector. Further explanation of the methods for determining the virtual scenario vector can be found in the preceding description.

[0187] The third similarity reflects the degree of similarity between the historical scenario vector and the virtual scenario vector. Understandably, the value range of the third similarity between the historical scenario vector and the virtual scenario vector is [0,1].

[0188] In some embodiments, the first determining module 220 can calculate the vector distance between the historical scenario vector and the virtual scenario vector, and determine a third similarity based on the vector distance. For example, the third similarity can be negatively correlated with the vector distance (e.g., the third similarity is inversely proportional to the vector distance). The vector distance can be calculated using Euclidean distance, Manhattan distance, cosine distance, Chebyshev distance, etc., and is not limited thereto.

[0189] In some embodiments, the first determining module 220 processes the historical scenario vector and the virtual scenario vector using a trained second similarity model to obtain a third similarity between the historical scenario vector and the virtual scenario vector. The second similarity model is a machine learning model (e.g., a deep learning model), and its input includes the historical scenario vector and the virtual scenario vector, while its output includes the third similarity between the historical scenario vector and the virtual scenario vector.

[0190] In some embodiments, the first determining module 220 may determine existing workers corresponding to historical scenario vectors whose third similarity to virtual scenario vectors is higher than a third similarity threshold as reference workers. The third similarity threshold may be preset, for example, the third similarity threshold may be preset to 0.8, 0.85, or 0.9, etc.

[0191] It should be noted that, as mentioned above, each existing worker may have multiple historical task data entries. Therefore, each existing worker may also have multiple historical scenario vectors (each historical scenario vector corresponds to one historical task data entry). Based on this, when determining the third similarity between each existing worker's historical scenario vector and virtual scenario vector, the third similarity between each of the worker's historical scenario vectors and virtual scenario vectors can be determined separately, resulting in multiple third similarities. The maximum or average of these multiple third similarities is then selected as the third similarity between the existing worker's historical scenario vector and virtual scenario vector.

[0192] In some embodiments of the present invention, by constructing virtual work scenarios to match industry-wide data, even when the target company has no historical rectification information (first-time recruitment or insufficient rectification information), it is still possible to find historical rectification information with similar scenarios in the industry-wide data through similar scenarios, ensuring that suitable reference workers can be identified.

[0193] Figure 5 This is an exemplary flowchart illustrating the generation of assessment content according to some embodiments of the present invention. Figure 5 As shown, process 500 includes the following steps. In some embodiments, process 500 may be executed by a second determining module 230.

[0194] Step S510: Based on the virtual scenario vector, determine the target rectification task from the historical rectification information of one or more reference workers. For an explanation of the virtual scenario vector and historical rectification information, please refer to [link to relevant documentation]. Figures 2 to 4 The relevant description in the document.

[0195] The target rectification task refers to the rectification task determined from the historical rectification information of one or more reference workers, and is used to determine the assessment content.

[0196] In some embodiments, the second determining module 230 can determine the rectification task corresponding to the target position as the target rectification task. For example, taking the target position as a scaffolder (the scaffolder's job is to erect scaffolding) as an example, assuming that the rectification tasks in the historical rectification information of one or more reference workers include "adjusting the scaffolding position", "adjusting the spacing of the rebar tying", and "adjusting the thickness of the waterproof layer", then the target rectification task is the rectification task involving scaffolding, that is, "adjusting the scaffolding position".

[0197] In some embodiments, the second determining module 230 can select reference workers whose historical scenario vectors meet the second preset conditions as target workers based on the third similarity between the virtual scenario vector and the historical scenario vectors of each reference worker, and use the rectification tasks in the historical rectification information of the target worker as target rectification tasks. The second preset conditions include that the third similarity between the historical scenario vector and the virtual scenario vector is greater than a fourth similarity threshold. The fourth similarity threshold is greater than the third similarity threshold. For example, assuming there are three reference workers, namely workers A, B, and C, whose third similarities between their historical scenario vectors and virtual scenario vectors are 0.9, 0.88, and 0.96, respectively, and taking the fourth similarity threshold of 0.95 as an example, then worker C's historical scenario vector meets the second preset conditions, and the second determining module 230 can determine worker C as the target worker and use the rectification tasks in worker C's historical rectification information as target rectification tasks.

[0198] Step S520: Generate assessment content based on the information records of the target rectification tasks.

[0199] Information records refer to the recorded data on the details of the rectification issues corresponding to the rectification tasks. For example, taking the rectification issue as "incorrect spacing of rebar tying", the corresponding information record could be "the spacing of the rebar does not meet the spacing requirements (e.g., the spacing error is greater than the preset error threshold). The preliminary analysis suggests that the cause may be that the workers did not carefully read the design drawings when tying the rebars, or were not proficient in using the measuring tools, or did not set the positioning clamps as required."

[0200] For example, based on the information recorded in the previous example, the generated assessment content can be a practical assessment, specifically including: giving the applicants steel bars, design drawings, measuring tools, etc., and having the applicants tie the steel bars, and the examiners determining the assessment results based on whether the tying results conform to the design drawings.

[0201] In some embodiments of the present invention, target rectification tasks are filtered by virtual scenario vectors, making the assessment content closer to the problems most likely to occur in similar environments. This allows for the assessment of the actual problems that applicants are most likely to encounter, thereby improving the relevance and effectiveness of recruitment assessments.

[0202] The basic concepts have been described above. It is clear that the detailed disclosure above is merely illustrative and does not constitute a limitation of the present invention. Although not explicitly stated herein, various modifications, improvements, and corrections may be made to the present invention by those skilled in the art. Such modifications, improvements, and corrections are suggested in this invention and therefore remain within the spirit and scope of the exemplary embodiments of the present invention.

[0203] Furthermore, unless expressly stated in the claims, the order of processing elements and sequences, the use of numbers and letters, or other names described in this invention are not intended to limit the order of the processes and methods of this invention. Although the foregoing disclosure has discussed some embodiments of the invention that are currently considered useful through various examples, it should be understood that such details are for illustrative purposes only, and the appended claims are not limited to the disclosed embodiments; rather, the claims are intended to cover all modifications and equivalent combinations that conform to the spirit and scope of this invention. For example, while the system components described above can be implemented using hardware devices, they can also be implemented solely through software solutions, such as installing the described system on existing servers or mobile devices.

[0204] Similarly, it should be noted that, in order to simplify the description of this invention and thus aid in the understanding of one or more embodiments, the foregoing description of this invention sometimes combines multiple features into a single embodiment, drawing, or description thereof. However, this method of disclosure does not imply that the subject matter of the invention requires more features than those mentioned in the claims. In fact, the embodiments contain fewer features than all the features of a single embodiment disclosed above.

[0205] In some embodiments, numbers describing the quantity of components and attributes are used. It should be understood that such numbers used in the description of embodiments are sometimes modified by the terms "approximately," "approximately," or "generally." Unless otherwise stated, "approximately," "approximately," or "generally" indicates that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may be changed depending on the characteristics required by individual embodiments. Although the numerical ranges and parameters used to confirm the breadth of their range in some embodiments of the invention are approximate values, in specific embodiments, such values ​​are set as precisely as feasible.

[0206] For each patent, patent application, patent application publication, and other material, such as articles, books, specifications, publications, and documents, referenced in this invention, the entire contents of which are incorporated herein by reference. This excludes historical application documents that are inconsistent with or conflict with the content of this invention, as well as documents that limit the broadest scope of the claims of this invention (currently or subsequently appended to this specification). It should be noted that if there is any inconsistency or conflict between the descriptions, definitions, and / or terminology used in the appended materials of this specification and the content of this invention, the descriptions, definitions, and / or terminology used in this specification shall prevail.

[0207] Finally, it should be understood that the embodiments described in this invention are merely illustrative of the principles of the invention. Other modifications may also fall within the scope of this invention. Therefore, alternative configurations of the invention are considered as examples and not limitations, and are consistent with the teachings of the invention. Accordingly, the invention is not limited to the embodiments explicitly described and illustrated in this specification.

Claims

1. A construction worker recruiting method characterized by comprising: The method is applied to a recruitment platform, comprising: obtaining an operation instruction of a user through the recruitment platform; in response to the operation instruction, calling, through a first terminal device, a job application information of a job applicant, the job application information comprising a target post of a target company to which the job applicant applies and resume information of the job applicant; calling a first determination module to determine one or more reference workers based on the target post and / or the resume information; calling a second determination module to determine an examination content of the job applicant based on historical rectification information of the one or more reference workers.

2. The method of claim 1, wherein, The determination of the one or more reference workers comprises: determining a plurality of candidate workers from the target company based on the resume information; determining one or more candidate workers having work experience of the target post in the plurality of candidate workers as the one or more reference workers based on the target post.

3. The method of claim 1, wherein, The determination of the one or more reference workers comprises: obtaining company information and project information of the target company through a second terminal device; constructing a virtual work scenario based on the target post, the company information and the project information; screening the one or more reference workers from a worker database based on the virtual work scenario.

4. The method of claim 3, wherein, The construction of the virtual work scenario based on the target post, the company information and the project information comprises: determining a post vector based on post description information of the target post; determining a company vector based on worker proportion, number of types of work, rectification frequency, and timeliness of salary payment in the company information; determining a project vector based on project type, project cycle, and current project stage in the project information; determining a virtual scenario vector based on the post vector, the company vector and the project vector, the virtual scenario vector being used to represent the virtual work scenario.

5. The method of claim 4, wherein, The screening of the one or more reference workers from the worker database based on the virtual work scenario comprises: calling historical task data of at least one worker from the worker database; determining a historical scenario vector corresponding to the historical task data for each worker in the at least one worker; determining the one or more reference workers based on similarity between the historical scenario vector and the virtual scenario vector.

6. The method according to claim 4 or 5, characterized in that, The determination of the examination content of the job applicant comprises: determining a target rectification task from historical rectification information of the one or more reference workers based on the virtual scenario vector; generating the examination content based on information records of the target rectification task.

7. The method of claim 1, wherein, The method further comprises: obtaining an examination result of the job applicant through the first terminal device; in response to the examination result representing passing of the examination, generating training content for the job applicant.

8. The method of claim 7, wherein, The generation of the training content for the job applicant comprises: in response to the examination result belonging to a first segment, generating the training content based on historical rectification information of the one or more reference workers.

9. The method of claim 7, wherein, The generation of the training content for the job applicant comprises: In response to the examination result belonging to the second segment, a first group and a second group are determined from the one or more reference workers, the first group including workers with an overall rectification rate lower than a threshold, and the second group including workers without a history of rectification in work experience of a target job type corresponding to the target post and with an overall rectification rate higher than the threshold; Based on the difference between the historical rectification information corresponding to the first group and the second group, the training content is generated.

10. A construction worker recruiting system characterized by comprising: Comprise: An acquisition module is configured to, in response to an operation instruction of a user, call, through a first terminal device, employment information of an employment applicant, the employment information including a target post of a target company to which the employment applicant applies and resume information of the employment applicant; A first determination module is configured to determine one or more reference workers based on the target post and / or the resume information; A second determination module is configured to determine examination content of the employment applicant based on historical rectification information of the one or more reference workers.

Images