Intelligent screening method for qualification certificates for building industry communication project bidding

Abstract

The invention relates to the technical field of building communication engineering, in particular to an intelligent screening method of qualification certificates for building industry communication project bidding, comprising the following steps: acquiring project information text of a target communication project in the building industry; extracting capability information of each qualification certificate object conforming to the required qualification certificate type from a qualification certificate database; the qualification object of each qualification type corresponds to an employee of one post role; constructing a three-dimensional matching graph according to the post roles of the to-be-dispatched team of the target communication project and the capability information of the qualification object; based on the three-dimensional matching atlas, performing semantic matching scoring on the qualification object and the project information text of the target communication project by adopting a pre-constructed multi-channel neural semantic matching network to obtain a semantic matching score of each qualification object; and determining an optimal team combination and a final qualification certificate set according to the post role of the to-be-distributed team of the target communication project and the semantic matching score of each qualification certificate object.

Description

Technical Field

[0001] The present invention relates to the technical field of building communication engineering, and in particular to an intelligent screening method for qualification certificates used for bidding for communication projects in the construction industry. Background Art

[0002] During the bidding process in the construction and communications industry, contractors are typically required to assemble a qualified team based on project requirements and submit relevant qualification documentation to pass the bidding and tendering qualification review and scoring process. This process also requires varying requirements for personnel qualifications, experience, and regional adaptability, depending on factors such as project size, structure, construction period, geographic environment, and technical limitations.

[0003] Existing qualification certificate screening methods rely primarily on manual experience and retrieval. Bidders or project managers must repeatedly compare job requirements with the certificate holder's competency information, including majors, titles, validity periods, and historical project experience, from multiple qualification certificate databases. This process is not only inefficient but also prone to mismatches due to subjective judgment or incomplete information, impacting bid quality and success rates.

[0004] In recent years, while some companies have begun to introduce simple information systems to manage qualification certificate information, most systems are still limited to static screening criteria, such as keyword filtering based on qualification certificate type and certificate status. These systems lack the semantic understanding and intelligent matching capabilities between job responsibilities and the capabilities of certificate holders, making it difficult to meet the comprehensive assessment needs of job fit and team collaboration in complex communication projects within the construction industry. Furthermore, current technologies generally fail to effectively utilize environmental information from certificate holders' past projects (such as geographic region, climate conditions, and construction type) to assist in assessing their adaptability to the current project. Furthermore, they lack analysis of historical collaboration relationships and time-based coordination among team members. Summary of the Invention

[0005] In view of the above-mentioned shortcomings and deficiencies of the prior art, the present invention provides an intelligent screening method for registration certificates for bidding in the construction industry.

[0006] In order to achieve the above objectives, the main technical solutions adopted by the present invention include:

[0007] An embodiment of the present invention provides a method for intelligently screening qualification certificates for bidding on communication projects in the construction industry, comprising the following steps:

[0008] S1. Obtain project information text of a target communication project in the construction industry;

[0009] The project information text includes: an introduction to the communications project, total cost, construction period requirements, geographical and climatic characteristics, construction technology limitations, required qualification types, and the job role composition of the proposed team;

[0010] S2. extracting competency information of each qualification certificate subject that meets the required qualification certificate type from the qualification certificate database;

[0011] The qualification certificate object of each qualification certificate type corresponds to a practitioner of a job role;

[0012] The competency information of each qualification certificate subject includes: qualification certificate professional category, professional title level, current qualification certificate status, conflict information of ongoing communication projects, validity period of qualification certificate, historical communication project experience, types of construction technology mastered, special operation license qualification, service area experience, and environmental information of each past communication project participated in;

[0013] S3. Construct a three-dimensional matching map based on the job roles of the proposed team for the target communications project and the capability information of the qualification certificate recipients;

[0014] S4. Based on the three-dimensional matching graph, a pre-built multi-channel neural semantic matching network is used to perform semantic matching scores on the project information text of the qualification certificate object and the target communication project, thereby obtaining a semantic matching score for each qualification certificate object;

[0015] S5. Determine the optimal team combination and the final qualification certificate set based on the job roles of the proposed team for the target communication project and the semantic matching score of each qualification certificate object.

[0016] Preferably, the S3 specifically includes:

[0017] S31. Extracting job task keywords and competency requirements for each position role of the proposed team for the target communications project based on publicly available job responsibilities in the national professional qualification standards database, and constructing a job responsibility description vector corresponding to each position role of the proposed team for the target communications project.

[0018] The national occupational qualification standard database is a collection of public occupational qualification standard information resources issued by the Ministry of Human Resources and Social Security or the competent industry department, and the national occupational qualification standard database includes the text of the job descriptions of all job roles in the field of communications construction;

[0019] S32. Extract capability information of each qualification certificate object that meets the required qualification certificate type from the qualification certificate database and construct a qualification certificate capability vector;

[0020] The qualification certificate capability vector is a multi-dimensional vector representation generated by structured processing of the capability information of the qualification certificate object;

[0021] S33. Use natural language processing technology to perform semantic analysis on the job description vector and the qualification certificate capability vector, and map them into a unified vector space to form a three-dimensional matching graph with a three-dimensional structure;

[0022] The three-dimensional structure includes: a job role name dimension, a job responsibility description vector dimension, and a qualification certificate capability vector dimension.

[0023] Preferably, the S31 specifically includes:

[0024] S311. Obtain a list of positions for each proposed position role in the target communication project;

[0025] S312. Extract the text of the job description corresponding to each job role from the national occupational qualification standard database;

[0026] S313. Perform natural language preprocessing on the text of the job description to extract job task keywords and ability requirements corresponding to the job role;

[0027] S314. Convert the extracted keywords and capability requirements into a vector representation of unified dimensions, and construct a job responsibility description vector corresponding to each job role of the proposed team for the target communication project.

[0028] Preferably, the S33 specifically includes:

[0029] S331. Based on a preset unified semantic vector space model, the job description vector and the qualification certificate capability vector are mapped into semantic vector representations respectively;

[0030] S332. Construct a three-dimensional structural matching map based on the mapped vectors in three dimensions: job role name dimension, job responsibility description vector dimension, and qualification certificate capability vector dimension, to represent the semantic matching relationship between each job responsibility and qualification certificate capability.

[0031] Preferably, the pre-built multi-channel neural semantic matching network includes:

[0032] The task intention channel uses an attention mechanism network to perform attention processing on the job description vector and output a task matching score.

[0033] The capability feature channel is used to perform nonlinear transformation on the qualification certificate capability vector using a deep feedforward neural network and output the capability matching score;

[0034] The environmental experience channel is used to input the target project environmental information and the historical project environmental information of the qualification certificate object into the graph neural network, extract environmental features and output the environmental matching score;

[0035] The project environment information comes from the fields of geographical climate characteristics and construction technology limitations in the project information text;

[0036] The historical project environment information of the qualification certificate subject is derived from the environment information of the communication projects in which the subject has participated in the past, including geographical area, construction conditions, and construction type;

[0037] The three channels output task matching scores, ability matching scores, and environment matching scores respectively, which are input into the weighted fusion module to calculate the semantic matching scores and output the semantic matching scores between job roles and qualification certificate objects.

[0038] Preferably, the weighted fusion module uses formula (1) to calculate the matching score to obtain the semantic matching score between the job role and the qualification certificate object; the formula (1) is:

[0039] Score(i,j)=α·Ftask(i,j)+β·Fability(i,j)+γ·Fenv(i,j);

[0040] Score(i, j) represents the semantic matching score between the i-th job role and the j-th qualification certificate object;

[0041] Ftask(i, j) represents the task matching score between the i-th job role and the j-th qualification certificate object output by the task intention channel;

[0042] Fability(i, j) represents the ability matching score between the i-th job role and the j-th qualification certificate object output by the ability feature channel;

[0043] Fenv(i, j) represents the environmental experience matching score between the i-th job role and the j-th qualification certificate object output by the environmental experience channel;

[0044] α is the first weight coefficient; β is the second weight coefficient; γ is the third weight coefficient; wherein 1=α+β+γ.

[0045] Preferably, the task matching score is the semantic similarity value between the job description vector and the qualification certificate capability vector measured by cosine similarity;

[0046] The ability matching score is a weighted score of the matching degree of the qualification certificate vector of the qualification certificate object in the corresponding fields according to the specific requirements of the job role for the qualification certificate type, professional title level, and mastered construction technology;

[0047] The environmental matching score is obtained by embedding the historical communication project environmental information of the qualification certificate object and the environmental information of the target communication project into a graph structure through a graph neural network, and calculating the embedding vector similarity value between the two.

[0048] Preferably, the S5 specifically includes:

[0049] S51. For each of the positions in the proposed team, select M qualification certificate objects in descending order of semantic matching scores with the position role to form a candidate set for the position role, and use the semantic matching scores between the M qualification certificate objects and the position role as the position adaptation score.

[0050] For each candidate set of job role personnel, divide all the object groups in the candidate set of job role personnel into groups;

[0051] The number of people in each target group is the number of candidates with the required qualifications for the position roles in the proposed team;

[0052] Randomly select a group of candidates from each position role candidate set and combine them to obtain a complete team candidate combination set;

[0053] S52. Obtain a comprehensive score for each complete set of candidate team combinations using a preset comprehensive evaluation function;

[0054] The comprehensive evaluation score is formed by weighting the following indicators: the candidate team's single-person job adaptation score, team schedule coordination index, multi-position conflict avoidance coefficient, and team historical cooperation tacit understanding;

[0055] S53. Based on the comprehensive score of each complete team candidate combination set, determine the optimal team combination and the final qualification certificate set;

[0056] Among them, the optimal team combination is the complete team candidate combination set with the highest comprehensive score; the final qualification certificate set includes the qualification certificates corresponding to each qualification certificate object in the optimal team combination.

[0057] Preferably, the comprehensive evaluation function is:

[0058]

[0059] MatchScore k is the job suitability score of the kth qualification certificate candidate in the complete team candidate combination set;

[0060] XT is the team duration synergy index in the complete team candidate combination set;

[0061] GB is the multi-position conflict avoidance coefficient in the complete team candidate combination set;

[0062] LS is the historical cooperation tacit understanding of the team in the complete set of team candidate combinations;

[0063] δ is the preset first coefficient; ε is the preset second coefficient; θ is the preset third coefficient; ω is the preset fourth coefficient.

[0064] Preferably, the calculation method of the team duration coordination index includes:

[0065] For each certificated subject in the complete team candidate combination set, obtain the estimated end time and reserved buffer period of the current communication project, and determine the time window for entry. Map the time window for entry of each certificated subject to a Gaussian time distribution function. Perform overlap integral calculation on the entry time distribution function between all certificated subjects to obtain the synergy overlap of each pair of certificated subjects. Take the average of the synergy overlap of all pairs of certificated subjects as the team's duration synergy index.

[0066] The calculation method of the conflict avoidance coefficient includes:

[0067] Check whether each certificate holder in the complete team candidate set has already held multiple roles in other ongoing communications projects. If so, record it as a conflict. Check whether each certificate holder in the complete team candidate set has multiple mutually exclusive roles in the target communications project. If so, record it as a conflict. Divide the actual number of conflicts by the theoretical maximum number of conflicts and take the inverse proportion to obtain the conflict avoidance coefficient, which ranges from 0 to 1.

[0068] The calculation method of the cooperative tacit understanding includes:

[0069] Extract the joint participation records of any two qualification certificate objects in historical communication projects in the complete team candidate combination set; count the historical number of cooperation between each pair of qualification certificate objects and the number of successful cooperation; calculate the cooperation success rate of each pair of qualification certificate objects, and take the weighted average according to the historical number of cooperation to obtain the cooperation tacit understanding of the entire candidate team.

[0070] The beneficial effects of the present invention are:

[0071] The intelligent screening method for qualification certificates used in bidding for communication projects in the construction industry adopts the technical path of structured extraction of project information text, construction of capability vectors of qualification certificate objects, three-dimensional matching graph modeling and multi-channel neural semantic matching network scoring mechanism. Compared with the existing technology, it can accurately evaluate the matching relationship between job responsibilities and qualification certificate objects from multiple dimensions based on the specific needs of the communication project, and automatically generate a highly matched personnel combination plan, achieving the beneficial effects of improving personnel selection efficiency, increasing team matching accuracy, and enhancing the intelligent level of personnel deployment.

[0072] Furthermore, since the job responsibilities description in the national occupational qualification standard database is used to construct the job responsibilities description vector, and natural language processing technology is combined to perform semantic analysis of job tasks and capability information, compared with the traditional method based on keyword matching or static screening rules, it can more accurately understand the semantic relationship between the actual job requirements and the capabilities of the qualification certificate object, achieving the beneficial effect of enhancing the intelligent analysis capability of human-job adaptation and reducing manual judgment errors.

[0073] In addition, by constructing a multi-channel neural semantic matching network and embedding task intention features, ability features and environmental experience features into different channels for semantic modeling, compared with the existing single indicator evaluation method, it can evaluate the job competence, process experience matching and environmental adaptability of the qualification certificate subjects in multiple dimensions, achieving the beneficial effect of improving the multi-source credibility of the scoring results and the business adaptability of the matching results.

[0074] Furthermore, in the process of team combination optimization, by introducing comprehensive evaluation indicators such as job adaptation score, construction period coordination index, multi-job conflict avoidance coefficient and team historical cooperation tacit understanding, compared with traditional manual combination or random recommendation methods, it can systematically evaluate the overall stability and construction coordination of the team, achieving the beneficial effect of improving the rationality of team combination. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 This is a flow chart of an intelligent screening method for qualification certificates used in bidding for communication projects in the construction industry according to the present invention. DETAILED DESCRIPTION

[0076] In order to better explain the present invention and facilitate understanding, the present invention is described in detail below through specific implementation methods in conjunction with the accompanying drawings.

[0077] To better understand the above technical solutions, exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments described herein. Instead, these embodiments are provided to enable a clearer and more thorough understanding of the present invention and to fully convey the scope of the present invention to those skilled in the art.

[0078] Example 1

[0079] See also Figure 1 This embodiment provides a method for intelligently screening qualifications for bidding on communication projects in the construction industry, comprising the following steps:

[0080] S1. Obtain project information text of a target communication project in the construction industry;

[0081] The project information text includes: an introduction to the communications project, total cost, construction period requirements, geographical and climatic characteristics, construction technology limitations, required qualification types, and the job role composition of the proposed team;

[0082] For example, the communication project content in the project information text in this embodiment is introduced: such as the introduction of the content of a specific intelligent building control system communication project; the total construction cost: about 200 million yuan; the construction period requirement: 18 months; the geographical and climatic characteristics: the humid and hot climate of the southern coastal areas; the construction technology limitations: for example, the communication distance limitation (the bus length is limited, and relaying is required if it exceeds the limit); the wiring environment limitation (wiring is difficult in high humidity and high temperature places); the required qualification certificate types: first-class construction engineer certificate (mechanical and electrical engineering), registered electrical engineer certificate, weak current engineer certificate, safety officer registration certificate, quality inspector registration certificate, etc.; the job role composition of the team to be dispatched: first-class construction engineer, registered electrical engineer, weak current engineer, safety officer, quality inspector, etc., a total of 5 types of job roles.

[0083] S2. extracting competency information of each qualification certificate subject that meets the required qualification certificate type from the qualification certificate database;

[0084] The qualification certificate object of each qualification certificate type corresponds to a practitioner of a job role;

[0085] The competency information of each qualification certificate holder includes: qualification certificate professional category (such as communications engineering, etc.), professional title level (such as intermediate, senior or first, second level), current qualification certificate status (whether it is valid), conflict information of ongoing communications projects (whether there is a time conflict), qualification certificate validity period (remaining validity period), historical communications project experience (type and scale of communications projects participated in by the certificate holder), types of construction technology mastered, special operation license qualifications, service area experience (whether the certificate holder has similar project experience in the southern region), and environmental information of each past communications project participated in by the certificate holder (such as geographical location, construction conditions, climate characteristics, etc.);

[0086] In this embodiment, the qualification certificate object is the holder of the qualification certificate.

[0087] In this embodiment, the qualification database pre-stores capability information for multiple qualification candidates for each qualification type in the construction industry communications field. For example, assuming there are five qualification types in the construction industry communications field, the qualification database stores capability information for multiple qualification candidates for each qualification type. For example, if the qualification type is registered electrical engineer, the qualification database stores capability information for seven individuals holding the registered electrical engineer qualification.

[0088] S3. Constructing a three-dimensional matching graph based on the job roles of the proposed team for the target communication project and the capability information of the qualification certificate recipients. In this embodiment, S3 specifically includes:

[0089] S31. Extract the job task keywords and competency requirements for each role in the proposed team of the target communications project based on the publicly available job descriptions in the national occupational qualification standards database (or a pre-set database), and construct a job description vector corresponding to each role in the proposed team of the target communications project. The national occupational qualification standards database is a collection of publicly available occupational qualification standards information resources published by the Ministry of Human Resources and Social Security or the relevant industry authorities. The national occupational qualification standards database (or a pre-set database) includes the text of job descriptions for all roles in the communications construction field. Specifically, S31 includes: S311. Obtain a list of positions for each proposed role in the target communications project; assuming the list includes: first-class construction engineer, registered electrical engineer, weak current engineer, safety officer, quality inspector, etc. S312. Extract the text of the job description corresponding to each role from the national occupational qualification standards database. For example, a first-class construction engineer (for mechanical and electrical engineering) is positioned as a project manager or project manager, primarily responsible for the overall organization and management of the project. Main responsibilities: Be fully responsible for the construction organization and implementation of the project, ensuring that indicators such as construction period, quality, safety, and cost are met. Organize the preparation and review of project construction organization design plans. Responsible for communicating and coordinating resources with the client, supervisor, subcontractor, and other parties. Coordinate the work of various disciplines, including civil engineering, electrical engineering, communications, low-voltage electrical engineering, and equipment installation. Ensure construction complies with specifications and standards and fulfill safety production responsibilities. Liaise with the client and relevant parties, and participate in project acceptance and final handover. Registered Electrical Engineer (Power Supply and Distribution / Generation, Transmission, and Transformation) serves as an expert in electrical system design and review, participating in system planning, design calculations, and technical review. Key Responsibilities: Responsible for the design of building power supply and distribution systems, equipment power systems, emergency power supply, lighting, and grounding. Review construction and design drawings to confirm compliance with national standards and actual usage requirements. Conduct power load calculations, cable selection, and distribution cabinet selection. Responsible for designing electrical interface linkages with intelligent building systems, communications systems, fire protection systems, and other systems. Provide technical consultation and support during the project construction, commissioning, and acceptance phases. If the project includes a substation or UPS or EPS room, these must be reviewed and confirmed by this person. S313. Perform natural language preprocessing on the text of the job description to extract the job task keywords and capability requirements corresponding to the job role; S314. Convert the extracted keywords and capability requirements into a vector representation of unified dimension to construct the job description vector corresponding to each job role of the proposed team of the target communication project. (Use natural language preprocessing technology (such as word segmentation, keyword extraction, syntactic analysis, etc.) to extract job task keywords (such as "construction coordination", "on-site guidance") and capability requirements. Then, convert these keywords and capability requirements into a unified vector expression (such as BERT embedding vector) to form the job description vector for each job role).

[0090] S32. Extract capability information of each qualification certificate object that meets the required qualification certificate type from the qualification certificate database and construct a qualification certificate capability vector; the qualification certificate capability vector is a multidimensional vector representation generated by structured processing of the capability information of the qualification certificate object;

[0091] For example, Electrical Engineer A holds the title of Senior Engineer and is familiar with 10kV power distribution system and emergency power supply design. He has seven years of experience in public building projects and has participated in the design of electrical systems for several large hospitals and airports. He excels at optimizing UPS and EPS system configurations and possesses strong cross-disciplinary coordination skills. Weak-current Engineer B, a certified Information Systems Project Manager, is proficient in integrated cabling, security monitoring, intelligent lighting, and building automation system integration. He has five years of experience in intelligent building and rail transit projects and has led the implementation of smart campus systems in multiple hot and humid climate zones. These information fields are structured and embedded into a unified semantic vector, forming a capability vector representation of the corresponding qualification object.

[0092] S33. Use natural language processing technology to perform semantic analysis on the job description vector and the qualification certificate capability vector, and map them into a unified vector space to form a three-dimensional matching graph with a three-dimensional structure; the three-dimensional structure includes: the job role name dimension, the job description vector dimension, and the qualification certificate capability vector dimension. S33 specifically includes: S331. Based on a preset unified semantic vector space model, map the job description vector and the qualification certificate capability vector into semantic vector representations;

[0093] S332. Construct a three-dimensional structural matching map based on the mapped vectors in three dimensions: job role name dimension, job responsibility description vector dimension, and qualification certificate capability vector dimension, to represent the semantic matching relationship between each job responsibility and qualification certificate capability.

[0094] Taking the position of "electrical engineer" as an example, the job responsibility vector is semantically mapped and organized with the ability vectors of multiple candidate qualification certificate objects such as electrical engineer A and electrical engineer B, and based on the following three dimensions: job role name dimension (such as "electrical engineer"), job responsibility description vector dimension, and qualification certificate ability vector dimension; a structured three-dimensional matching map is constructed for subsequent semantic matching score calculation.

[0095] In this embodiment, in the process of constructing a three-dimensional matching map between job roles and qualification certificate objects, the method of automatically extracting job responsibilities from the national occupational qualification standard database and combining natural language processing technology to semantically vectorize job tasks and ability requirements can more accurately express the job's responsibilities and ability characteristics compared to the traditional method of relying on manual reading of job requirements or static field screening. Furthermore, by simultaneously mapping the job responsibilities description vector and the qualification certificate ability vector to a unified semantic vector space, and constructing a semantic matching map according to the three dimensions of job role name, job responsibilities and qualification certificate ability, it can fully express the semantic relationship between jobs and people from multiple angles and levels, achieving the beneficial effect of improving the accuracy of job adaptation judgment and enhancing matching intelligence. In addition, the three-dimensional matching map serves as the basic input structure for subsequent neural network model scoring, which helps to improve the context modeling quality of semantic matching and reduce the risk of shallow keyword mismatching, thereby achieving high-quality and efficient intelligent screening recommendations.

[0096] S4. Based on the three-dimensional matching graph, a pre-built multi-channel neural semantic matching network is used to perform semantic matching scores on the project information text of the qualification certificate object and the target communication project, thereby obtaining a semantic matching score for each qualification certificate object;

[0097] In this embodiment, the pre-built multi-channel neural semantic matching network includes:

[0098] The task intention channel uses an attention mechanism network to perform attention processing on the job description vector and output a task matching score.

[0099] For example, a job description vector contains word meanings. A Transformer-based attention mechanism (such as Self-Attention) is used to encode this job description vector, extracting the semantic components representing the core work tasks and generating a task preference feature vector. The competency vector for a qualification certificate object contains similar semantic representations. The attention-weighted similarity between the two is calculated, and the task matching score for the qualification certificate object to the job description is output, for example, 0.89.

[0100] The capability feature channel is used to perform nonlinear transformation on the qualification certificate capability vector using a deep feedforward neural network and output the capability matching score;

[0101] For example, the competency vector of qualification certificate object A is input into a multi-layer feedforward neural network (e.g., an MLP). The network automatically performs nonlinear transformations on dimensions such as "professional title level = senior" and "qualification certificate status = valid" to extract competency features. The target competency requirements for a job role include tags such as "experience in communications projects." The network aligns the mapping result with the target requirements at the job level and outputs a competency match score, for example, 0.93.

[0102] The environmental experience channel is used to input the target project environmental information and the historical project environmental information of the qualification certificate object into the graph neural network, extract environmental features and output the environmental matching score;

[0103] For example, project environmental information can be extracted from the project text, including: geographical climate characteristics: southern coastal hot and humid area; fields for construction technology restrictions, etc.

[0104] The historical project environment information of Qualification Certificate Subject A includes: having participated in communications projects in three southern cities; and being familiar with wet working conditions.

[0105] The "target project environment" and "historical project environment" are represented as environment feature nodes respectively, and modeled with a graph structure. After being embedded through a graph neural network (such as GraphSAGE), the similarity of the graph vectors of the two is calculated, and the environment matching score is output, for example, 0.85.

[0106] The project environment information comes from the fields of geographical climate characteristics and construction technology limitations in the project information text;

[0107] The historical project environment information of the qualification certificate subject is derived from the environment information of the projects in which the subject has participated in the past, including geographical area, communication type, etc.

[0108] The three channels output task matching scores, ability matching scores, and environment matching scores respectively, which are input into the weighted fusion module to calculate the semantic matching scores and output the semantic matching scores between job roles and qualification certificate objects.

[0109] The weighted fusion module uses formula (1) to calculate the matching score and obtain the semantic matching score between the job role and the qualification certificate object; the formula (1) is:

[0110] Score(i,j)=α·Ftask(i,j)+β·Fability(i,j)+γ·Fenv(i,j);

[0111] Score(i, j) represents the semantic matching score between the i-th job role and the j-th qualification certificate object; Ftask(i, j) represents the task matching score between the i-th job role and the j-th qualification certificate object output by the task intention channel; Fability(i, j) represents the ability matching score between the i-th job role and the j-th qualification certificate object output by the ability feature channel; Fenv(i, j) represents the environment experience matching score between the i-th job role and the j-th qualification certificate object output by the environment experience channel; α is the first weight coefficient; β is the second weight coefficient; γ is the third weight coefficient; where 1 = α + β + γ.

[0112] The above three channels output respectively: task matching score: 0.89; ability matching score: 0.93; environment matching score: 0.85; input weighted fusion module, according to the preset weights (such as α = 0.4, β = 0.4, γ = 0.2), calculate the comprehensive semantic matching score: 0.4×0.89+0.4×0.93+0.2×0.85=0.896; that is, the comprehensive semantic matching score between the qualification certificate object and the "electrical engineer" position is 0.896.

[0113] The task matching score is the semantic similarity value between the job description vector and the qualification certificate capability vector measured by cosine similarity; the capability matching score is a weighted score of the matching degree of the qualification certificate object's qualification certificate capability vector in the corresponding fields based on the specific requirements of the job role for the qualification certificate category, professional title level, and mastered construction technology; the environment matching score is the graph structure embedding of the historical project environment information of the qualification certificate object and the environment information of the target communication project through a graph neural network, and the calculated embedding vector similarity value between the two.

[0114] In this embodiment, a comprehensive semantic matching score is formed through weight fusion through multi-channel output, so that the matching result is more in line with the actual comprehensive considerations of the project, and the intelligence level and decision-making efficiency of the team screening are improved.

[0115] S5. Determine the optimal team combination and the final qualification certificate set based on the job roles of the proposed team for the target communication project and the semantic matching score of each qualification certificate object.

[0116] In this embodiment, S5 specifically includes:

[0117] S51. For each of the positions in the proposed team, select M qualification certificate objects in descending order of semantic matching scores with the position role to form a candidate set for the position role, and use the semantic matching scores between the M qualification certificate objects and the position role as the position adaptation score.

[0118] For each candidate set of job role personnel, divide all the object groups in the candidate set of job role personnel into groups;

[0119] The number of people in each target group is the number of candidates with the required qualifications for the position roles in the proposed team;

[0120] Randomly select a group of candidates from each position role candidate set and combine them to obtain a complete team candidate combination set;

[0121] That is to say, in each candidate set of job roles, the candidates are divided into multiple object groups according to the number of people required for the job; one combination is selected from each group corresponding to each job to form a complete team candidate set. Assume that the target project needs to assign 5 job roles, namely: first-class construction engineer (1 person), registered electrical engineer (1 person), weak current engineer (1 person), safety officer (2 people), quality officer (1 person). The intelligent screening method for qualifications is based on the matching score results of step S4. For each job role, the top M (such as M=5) qualification certificate objects with semantic matching scores are extracted as candidates to form the following job candidate set:

[0122] The top five qualification certificates (numbers) with the best matching scores for first-class construction engineers are: A1, A2, A3, A4, and A5; the top five qualification certificates (numbers) with the best matching scores for registered electrical engineers are: B1, B2, B3, B4, and B5; the top five qualification certificates (numbers) with the best matching scores for weak current engineers are: C1, C2, C3, C4, and C5; the top five qualification certificates (numbers) with the best matching scores for safety officers (two persons required) are: D1, D2, D3, D4, and D5; the top five qualification certificates (numbers) with the best matching scores for quality inspectors are: E1, E2, E3, E4, and E5;

[0123] Then, combine the people in these candidate sets:

[0124] For the safety officer position (requires 2 people), enumerate all non-repeating two-person combinations from the 5 people to form 10 groups; for each other position (requires 1 person), each person is considered a group, for a total of 5 groups;

[0125] Randomly select a team member from the target group of each position (or use an exhaustive or heuristic algorithm), and merge the selected teams of all positions to form a complete team combination.

[0126] For example, including the following members:

[0127] First-class construction engineer: A3; registered electrical engineer: B1; weak current engineer: C4; safety officer: D2, D5; quality inspector: E1; this combination is one of the items in a "complete team candidate combination set".

[0128] S52. Obtain a comprehensive score for each complete set of candidate team combinations using a preset comprehensive evaluation function;

[0129] The comprehensive evaluation score is formed by weighting the following indicators: the candidate team's single-person job adaptation score, team schedule coordination index, multi-position conflict avoidance coefficient, and team historical cooperation tacit understanding;

[0130] The comprehensive evaluation function is:

[0131]

[0132] MatchScore k is the job adaptation score of the kth qualification certificate candidate in the complete team candidate combination set. For example, assume that the complete team candidate combination set is A3, a first-class construction engineer; B1, a registered electrical engineer; C4, a weak current engineer; D2 and D5, a safety officer; and E1, a quality inspector. The job adaptation scores of all members in the combination are averaged; for example: A3 = 0.91, B1 = 0.87, C4 = 0.90, D2 = 0.82, D5 = 0.84, and E1 = 0.86.

[0133] Average job fit score = (0.91 + 0.87 + 0.90 + 0.82 + 0.84 + 0.86) / 6 = 0.866;

[0134] XT is the team duration synergy index in the complete team candidate combination set;

[0135] GB is the multi-position conflict avoidance coefficient in the complete team candidate combination set;

[0136] LS is the historical cooperation tacit understanding of the team in the complete set of team candidate combinations;

[0137] δ is a preset first coefficient (in this embodiment, δ=0.4); ε is a preset second coefficient (in this embodiment, ε=0.2); θ is a preset third coefficient (in this embodiment, θ=0.2); ω is a preset fourth coefficient (in this embodiment, ω=0.2).

[0138] The calculation method of the team duration coordination index includes:

[0139] For each qualification certificate subject in the complete team candidate combination set, obtain the estimated end time and reserved buffer period of the current communication project, and determine the time window for entry. Map the time window for entry of each qualification certificate subject to a Gaussian time distribution function. Perform overlap integral calculation on the time distribution function for entry of all qualification certificate subjects to obtain the synergy overlap of each pair of qualification certificate subjects. Take the average of the synergy overlap of all pairs of qualification certificate subjects as the team's duration synergy index.

[0140] For example, a candidate team consists of five certified members. The estimated end time and reserved buffer period for each member's current communications project are extracted from the database to determine their available time windows. These time windows are converted to a Gaussian distribution, and the degree of overlap between the time windows of all members is calculated. A higher degree of overlap indicates that all members are more likely to arrive at the construction site at the same time. Finally, the average overlap between all pairs of members is taken as the construction schedule coordination index for that team. A higher value indicates a more coordinated team arrival.

[0141] The calculation method of the conflict avoidance coefficient includes:

[0142] Check whether each qualified individual in the complete candidate team set already holds multiple roles in other ongoing communications projects. If so, record this as a conflict. Check whether each qualified individual in the complete candidate team set holds multiple mutually exclusive roles in the target communications project. If so, record this as a conflict. Divide the actual number of conflicts by the theoretical maximum number of conflicts, and take the inverse of the proportionality to obtain the conflict avoidance coefficient, which ranges from 0 to 1. For a candidate team, check whether any of its members already hold multiple roles in other construction projects, or have been assigned multiple mutually exclusive roles in the target communications project (e.g., serving as both a safety officer and a construction worker). If a conflict exists, record it as 1, and calculate the conflict avoidance coefficient based on the actual number of conflicts and the theoretical maximum number of conflicts. A higher coefficient indicates a more reasonable distribution of team member responsibilities and fewer conflicts.

[0143] The calculation method of the cooperative tacit understanding includes:

[0144] Extract the joint participation records of any two qualification certificate objects in historical communication projects in the complete team candidate combination set; count the historical number of cooperation between each pair of qualification certificate objects and the number of successful cooperation; calculate the cooperation success rate of each pair of qualification certificate objects, and take the weighted average according to the historical number of cooperation to obtain the cooperation tacit understanding of the entire candidate team.

[0145] For example, the system calculates whether candidate team members have previously worked on the same project and successfully completed it. For example, if two members have collaborated multiple times and have a high completion rate, this indicates a strong level of rapport. The system then calculates and weights the collaboration records of all pairs of members and averages them to determine the overall team rapport. A higher value indicates a stronger foundation for teamwork.

[0146] S53. Based on the comprehensive score of each complete team candidate combination set, determine the optimal team combination and the final qualification certificate set;

[0147] Among them, the optimal team combination is the complete team candidate combination set with the highest comprehensive score; the final qualification certificate set includes the qualification certificates corresponding to each qualification certificate object in the optimal team combination.

[0148] In this embodiment, during the team combination optimization process, by introducing comprehensive evaluation indicators such as job adaptation score, construction period coordination index, multi-job conflict avoidance coefficient and team historical cooperation tacit understanding, compared with traditional manual combination or random recommendation methods, it can systematically evaluate the overall stability and construction coordination of the team, thereby achieving the beneficial effects of improving the rationality of team combination and reducing construction risks and management costs.

[0149] Example 2

[0150] This embodiment provides a method for intelligently screening qualification certificates for bidding on communication projects in the construction industry, comprising the following steps:

[0151] Step 1: Obtain project information text of a target communication project in the construction industry; the project information text includes: an introduction to the communication project content, total cost, construction period requirements, geographical and climatic characteristics, construction technology limitations, required qualification certificate types, and job role composition information of the proposed team.

[0152] Step 2: Extract the competency information of each qualification certificate subject that meets the required qualification certificate type from the qualification certificate database;

[0153] The competency information of each qualification certificate subject includes: qualification certificate professional category, professional title level, current qualification certificate status, conflict information of ongoing communication projects, validity period of qualification certificate, historical communication project experience, types of construction technology mastered, special operation license qualification, service area experience, and environmental information of each past communication project participated in;

[0154] Step 3: Construct a three-dimensional matching map, including:

[0155] The job description texts of different job roles are extracted from the national occupational qualification standard database, and keywords are extracted. Then, natural language processing technology (such as BERT embedding) is used to vectorize the job tasks and ability requirements to obtain the job job description vector.

[0156] Based on the fields of the capability information of the qualification certificate object, structural processing is performed to construct a qualification certificate capability vector.

[0157] Construct a three-dimensional matching graph, including: the job role name dimension: such as "registered electrical engineer"; the job description vector dimension: representing the ability requirements of the role through semantic embedding; the qualification certificate ability vector dimension: representing the comprehensive ability of the qualification certificate holder;

[0158] Under the three-dimensional structure, the job description vector of the "Registered Electrical Engineer" position is semantically matched with the capability vectors of multiple candidate qualification certificate objects to construct a three-dimensional matching graph.

[0159] Step 4: Based on the three-dimensional matching graph, a pre-built multi-channel neural semantic matching network is used to match and score the project information text of the qualification certificate object and the target communication project to obtain a semantic matching score for each qualification certificate object;

[0160] In this embodiment, the preset multi-channel neural network includes the following three channels:

[0161] Task intent channel: This channel uses an attention mechanism to extract the job description vector of the position of "Registered Electrical Engineer" and outputs a task matching score.

[0162] Specifically, the task intent pipeline incorporates an attention mechanism to weightedly extract key information from the job description vector. This attention mechanism, based on a pre-set task intent query vector, calculates the similarity between the query vector and each word vector to determine the importance weight of each word. This weighted summation of the original word vectors yields a task intent vector representing the core intent of the job responsibilities. This task intent vector fully captures the key functional requirements and task objectives of the target position.

[0163] After obtaining the task intent vector, the channel calculates similarity between it and the vector of the candidate matching object (such as the qualification certificate capability vector), and uses matching functions such as cosine similarity, vector dot product, or multi-layer perceptron to output the task matching score. This score is used to measure the degree of semantic match between the candidate object and the job task requirements, providing input for subsequent multi-channel fusion.

[0164] Through the above method, the task intent channel can extract semantic features related to the core tasks of the position from the unstructured job description vector, and realize efficient task intent modeling and matching evaluation, thereby improving the accuracy and intelligence level of the overall multi-channel matching network in talent recommendation or job matching scenarios.

[0165] Competence characteristic channel: Extract the key competence dimensions of the qualification certificate object (such as "professional title = senior"); combine it with the competence requirements set by the position, and perform in-depth matching, with a score of 0.91, for example.

[0166] Environmental experience channel: used to input the target project environmental information and the historical project environmental information of the qualification certificate object into the graph neural network, extract environmental features and output the environmental matching score.

[0167] The scoring results of the three channels will be comprehensively weighted to generate the final semantic matching score for each qualification certificate subject for each job role in the target project.

[0168] Step 5: Determine the optimal team combination and final qualification set based on the job roles of the proposed team for the target communications project and the semantic matching scores of each qualification object, including:

[0169] Step 5-1: For each of the positions in the proposed team, select M qualification certificate objects in descending order of semantic matching scores with the position role, and combine them into a candidate set for the position role. The semantic matching scores between the M qualification certificate objects and the position role are respectively used as the position adaptation scores;

[0170] For each candidate set of job role personnel, divide all the object groups in the candidate set of job role personnel into groups;

[0171] The number of people in each target group is the number of candidates with the required qualifications for the position roles in the proposed team;

[0172] Randomly select a group of candidates from each position role candidate set and combine them to obtain a complete team candidate combination set;

[0173] For example, assume that the required job roles are: 1 first-class construction engineer, 1 registered electrical engineer, 1 weak current engineer, 2 safety officers, and 1 quality inspector.

[0174] Then, the semantic matching scores between each qualification certificate object and each position are obtained respectively.

[0175] For each job role, select the top M = 5 qualification certificate objects (job candidate set) with the highest semantic matching degree.

[0176] The candidate set for first-class construction engineer: A1 (semantic matching score 0.94), A2 (semantic matching score 0.92), A3 (semantic matching score 0.90), A4 (semantic matching score 0.88), A5 (semantic matching score 0.85).

[0177] Registered electrical engineer candidate set: B1 (semantic matching score 0.93), B2 (semantic matching score 0.91), B3 (semantic matching score 0.89), B4 (semantic matching score 0.87), B5 (semantic matching score 0.85).

[0178] Candidate set of weak current engineers: C1 (semantic matching score 0.96), C2 (semantic matching score 0.95), C3 (semantic matching score 0.93), C4 (semantic matching score 0.90), C5 (semantic matching score 0.88).

[0179] Safety officer candidate set (2 persons required): D1 (semantic matching score 0.91), D2 (semantic matching score 0.90), D3 (semantic matching score 0.88), D4 (semantic matching score 0.87), D5 (semantic matching score 0.86).

[0180] The candidate set of quality inspectors: E1 (semantic matching score 0.89), E2 (semantic matching score 0.87), E3 (semantic matching score 0.86), E4 (semantic matching score 0.85), E5 (semantic matching score 0.84).

[0181] Generate object groups according to the required number of people: First-class construction engineer (1 person): group = [A1], [A2], [A3], [A4], [A5] (5 groups in total); Registered electrical engineer (1 person): group = [B1], [B2], [B3], [B4], [B5]; Weak current engineer (1 person): group = [C1], [C2], [C3], [C4], [C5]; Safety officer (2 people): select 2 people from 5 people, for a total of C (5, 2) = 10 groups, for example: [D1+D2], [D1+D3], [D1+D4], [D1+D5], [D2+D3], [D2+D4], [D2+D5], [D3+D4], [D3+D5], [D4+D5]. Quality inspector (1 person): Group = [E1], [E2], [E3], [E4], [E5].

[0182] Randomly select a group from each role group to form a complete team candidate combination. For example, the following combination is selected: Class A Construction Engineer A3, Registered Electrical Engineer B1, Weak Current Engineer C2, Safety Officers D1+D4, Quality Inspector E2. This combination is a candidate in the "Complete Team Candidate Set."

[0183] Step 5-2: Use the preset comprehensive evaluation function to obtain the comprehensive score of each complete team candidate combination set;

[0184] The comprehensive evaluation score is formed by weighting the following indicators: the candidate team's single-person job adaptation score, team schedule coordination index, multi-position conflict avoidance coefficient, and team historical cooperation tacit understanding;

[0185] The comprehensive evaluation function is:

[0186]

[0187] MatchScore k is the job suitability score of the kth qualification certificate candidate in the complete team candidate combination set;

[0188] XT is the team duration synergy index in the complete team candidate combination set;

[0189] GB is the multi-position conflict avoidance coefficient in the complete team candidate combination set;

[0190] LS is the historical cooperation tacit understanding of the team in the complete set of team candidate combinations;

[0191] δ is a preset first coefficient; ε is a preset second coefficient; θ is a preset third coefficient; ω is a preset fourth coefficient. In this embodiment, the preset first coefficient δ is 0.1, the preset second coefficient ε is 0.2, the preset third coefficient θ is 0.3, and the preset fourth coefficient ω is 0.4.

[0192] The calculation method of the team duration coordination index includes:

[0193] For each qualification certificate object in the complete team candidate combination set, obtain the estimated end time of the current project and the reserved buffer period to determine its available entry time window; map the available entry time window of each qualification certificate object to a Gaussian time distribution function; perform overlap integral calculation on the entry time distribution function between all qualification certificate objects to obtain the synergy overlap of each pair of qualification certificate objects; take the average of the synergy overlap of all pairs of qualification certificate objects as the team's construction period synergy index;

[0194] For example, to obtain the available entry time window for each qualification certificate object, for example:

[0195] A3: 2025 / 06 / 01-2025 / 09 / 01 (center date = July 1);

[0196] B1: 2025 / 06 / 15-2025 / 08 / 15 (center date = July 15);

[0197] C2: 2025 / 05 / 20-2025 / 08 / 20 (center date = July 1);

[0198] D1: 2025 / 06 / 01-2025 / 07 / 30 (center date = June 30);

[0199] D4: 2025 / 06 / 05-2025 / 08 / 05 (center date = July 5);

[0200] E2: 2025 / 06 / 20-2025 / 08 / 10 (center date = July 15);

[0201] Convert each time window to a Gaussian distribution function, assuming the variance is uniformly 10 days.

[0202] Calculate the time distribution overlap integral of each pair of members to obtain the degree of co-coincidence, for example:

[0203] A3 and B1: 0.85;

[0204] A3 and C2: 0.95;

[0205] A3 and D1: 0.90;

[0206] A3 and D4: 0.92;

[0207] A3 and E2: 0.88;

[0208] ... (a total of C(6,2)=15 pairs of combinations);

[0209] Take the average of all overlaps, assuming: average overlap = 0.89.

[0210] The calculation method of the conflict avoidance coefficient includes:

[0211] Check whether each certificate holder in the complete team candidate set has already held multiple roles in other ongoing projects. If so, record it as a conflict. Check whether each certificate holder in the complete team candidate set has multiple mutually exclusive roles in this project. If so, record it as a conflict. Divide the actual number of conflicts by the theoretical maximum number of conflicts and take the inverse proportion to obtain the conflict avoidance coefficient, which ranges from 0 to 1.

[0212] For example, check if a candidate has held multiple roles in other projects:

[0213] Assume that D1 holds two positions in two other projects at the same time (recorded as one conflict);

[0214] The rest do not have multiple positions;

[0215] Check whether the same person holds multiple positions in the current combination (none found);

[0216] Total number of conflicts = 1; Theoretical maximum number of conflicts = 6 (each person can have a maximum of 1 conflict);

[0217] Conflict avoidance coefficient = 1-1 / 6 = 0.833.

[0218] The calculation method of the cooperative tacit understanding includes:

[0219] Extract the joint participation records of any two qualification certificate objects in historical communication projects in the complete team candidate combination set; count the historical number of cooperation between each pair of qualification certificate objects and the number of successful cooperation; calculate the cooperation success rate of each pair of qualification certificate objects, and take the weighted average according to the historical number of cooperation to obtain the cooperation tacit understanding of the entire candidate team.

[0220] For example, let's query the historical cooperation records between people:

[0221] A3 and B1: 3 collaborations, 2 successful → 66.7% success rate;

[0222] A3 and C2: 2 times, 2 successes → 100%;

[0223] A3 and D1: no cooperation record;

[0224] B1 and D1: Cooperation once, failure → 0%;

[0225] C2 and D4: Cooperation once, success → 100%;

[0226] D1 and D4: 2 collaborations, 1 success → 50%;

[0227] E2 and other members: Collaboration varies from 1 to 2 times, with an average success rate of about 70%;

[0228] The weighted average of the cooperation success rates of all member pairs is used to obtain the team's cooperation tacit understanding:

[0229] (66.7+100+0+0+100+50+70+…) / logarithm of cooperation ≈ 0.71.

[0230] Step 5-3: Based on the comprehensive scores of each complete team candidate combination set, determine the optimal team combination and the final qualification certificate set;

[0231] Among them, the optimal team combination is the complete team candidate combination set with the highest comprehensive score; the final qualification certificate set includes the qualification certificates corresponding to each qualification certificate object in the optimal team combination.

[0232] The method of this embodiment not only selects the person with the strongest individual ability, but also selects a combination with a high overall job adaptation score, strong teamwork tacit understanding, and good coordination of entry time. It introduces a multi-job conflict avoidance coefficient to prevent one person from appearing in multiple project teams at the same time; the team's historical cooperation tacit understanding index takes into account actual cooperation experience to improve overall cooperation efficiency.

[0233] In the description of the present invention, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of the technical features indicated. Therefore, a feature specified as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, "plurality" means two or more, unless otherwise specifically defined.

[0234] In the present invention, unless otherwise expressly specified or limited, the terms "mounted," "connected," "connect," "fixed," etc. should be understood broadly. For example, they may refer to fixed connection, detachable connection, or integration; mechanical connection or electrical connection; direct connection or indirect connection through an intermediate medium; and internal communication between two components or interaction between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on specific circumstances.

[0235] In the present invention, unless otherwise expressly specified or limited, when a first feature is "above" or "below" a second feature, it may mean that the first and second features are in direct contact, or that the first and second features are in indirect contact through an intermediate medium. Furthermore, when a first feature is "above," "above," or "above" a second feature, it may mean that the first feature is directly above or obliquely above the second feature, or simply means that the first feature is at a higher level than the second feature. When a first feature is "below," "below," or "below" a second feature, it may mean that the first feature is directly below or obliquely below the second feature, or simply means that the first feature is at a lower level than the second feature.

[0236] In the description of this specification, the terms "one embodiment", "some embodiments", "embodiments", "examples", "specific examples" or "some examples" refer to the specific features, structures, materials or characteristics described in conjunction with the embodiment or example and included in at least one embodiment or example of the present invention. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine different embodiments or examples described in this specification and features of different embodiments or examples, unless they are mutually inconsistent.

[0237] Although the embodiments of the present invention have been shown and described above, it will be understood that the above embodiments are illustrative and are not to be construed as limitations on the present invention. A person skilled in the art may alter, modify, replace and modify the above embodiments within the scope of the present invention.

Claims

1. An intelligent screening method for qualification certificates for bidding on communication projects in the construction industry, characterized in that: The following steps are involved: S1. Obtain project information text of a target communication project in the construction industry; The project information text includes: an introduction to the communications project, total cost, construction period requirements, geographical and climatic characteristics, construction technology limitations, required qualification types, and the job role composition of the proposed team; S2. extracting competency information of each qualification certificate subject that meets the required qualification certificate type from the qualification certificate database; The qualification certificate object of each qualification certificate type corresponds to a practitioner of a job role; The competency information of each qualification certificate subject includes: qualification certificate professional category, professional title level, current qualification certificate status, conflict information of ongoing communication projects, validity period of qualification certificate, historical communication project experience, types of construction technology mastered, special operation license qualification, service area experience, and environmental information of each past communication project participated in; S3. Construct a three-dimensional matching map based on the job roles of the proposed team for the target communications project and the capability information of the qualification certificate recipients; S4. Based on the three-dimensional matching graph, a pre-built multi-channel neural semantic matching network is used to perform semantic matching scores on the project information text of the qualification certificate object and the target communication project, thereby obtaining a semantic matching score for each qualification certificate object; S5. Determine the optimal team combination and the final qualification certificate set based on the job roles of the proposed team for the target communication project and the semantic matching score of each qualification certificate object.

2. The intelligent screening method for qualification certificates for construction industry communication project bidding according to claim 1 is characterized in that: The S3 specifically includes: S31. Extracting job task keywords and competency requirements for each position role of the proposed team for the target communications project based on publicly available job responsibilities in the national professional qualification standards database, and constructing a job responsibility description vector corresponding to each position role of the proposed team for the target communications project. The national occupational qualification standard database is a collection of public occupational qualification standard information resources issued by the Ministry of Human Resources and Social Security or the competent industry department, and the national occupational qualification standard database includes the text of the job descriptions of all job roles in the field of communications construction; S32. Extract capability information of each qualification certificate object that meets the required qualification certificate type from the qualification certificate database and construct a qualification certificate capability vector; The qualification certificate capability vector is a multi-dimensional vector representation generated by structured processing of the capability information of the qualification certificate object; S33. Use natural language processing technology to perform semantic analysis on the job description vector and the qualification certificate capability vector, and map them into a unified vector space to form a three-dimensional matching graph with a three-dimensional structure; The three-dimensional structure includes: a job role name dimension, a job responsibility description vector dimension, and a qualification certificate capability vector dimension.

3. The intelligent screening method for qualification certificates for construction industry communication project bidding according to claim 2 is characterized in that: The S31 specifically includes: S311. Obtain a list of positions for each proposed position role in the target communication project; S312. Extract the text of the job description corresponding to each job role from the national occupational qualification standard database; S313. Perform natural language preprocessing on the text of the job description to extract job task keywords and ability requirements corresponding to the job role; S314. Convert the extracted keywords and capability requirements into a vector representation of unified dimensions, and construct a job responsibility description vector corresponding to each job role of the proposed team for the target communication project.

4. The intelligent screening method for qualification certificates for construction industry communication project bidding according to claim 3 is characterized in that: The S33 specifically includes: S331. Based on a preset unified semantic vector space model, the job description vector and the qualification certificate capability vector are mapped into semantic vector representations respectively; S332. Construct a three-dimensional structural matching map based on the mapped vectors in three dimensions: job role name dimension, job responsibility description vector dimension, and qualification certificate capability vector dimension, to represent the semantic matching relationship between each job responsibility and qualification certificate capability.

5. The intelligent screening method for qualification certificates for construction industry communication project bidding according to claim 4 is characterized in that: Pre-built multi-channel neural semantic matching networks include: The task intention channel uses an attention mechanism network to perform attention processing on the job description vector and output a task matching score. The capability feature channel is used to perform nonlinear transformation on the qualification certificate capability vector using a deep feedforward neural network and output the capability matching score; The environmental experience channel is used to input the target project environmental information and the historical project environmental information of the qualification certificate object into the graph neural network, extract environmental features and output the environmental matching score; The project environment information comes from the fields of geographical climate characteristics and construction technology limitations in the project information text; The historical project environment information of the qualification certificate subject is derived from the environment information of the communication projects in which the subject has participated in the past, including geographical area, construction conditions, and construction type; The three channels output task matching scores, ability matching scores, and environment matching scores respectively, which are input into the weighted fusion module to calculate the semantic matching scores and output the semantic matching scores between job roles and qualification certificate objects.

6. The intelligent screening method for qualification certificates for construction industry communication project bidding according to claim 5 is characterized in that: The weighted fusion module uses formula (1) to calculate the matching score and obtain the semantic matching score between the job role and the qualification certificate object; the formula (1) is: Score(i,j)=α·Ftask(i,j)+β·Fability(i,j)+γ·Fenv(i,j); Score(i, j) represents the semantic matching score between the i-th job role and the j-th qualification certificate object; Ftask(i, j) represents the task matching score between the i-th job role and the j-th qualification certificate object output by the task intention channel; Fability(i, j) represents the ability matching score between the i-th job role and the j-th qualification certificate object output by the ability feature channel; Fenv(i, j) represents the environmental experience matching score between the i-th job role and the j-th qualification certificate object output by the environmental experience channel; α is the first weight coefficient; β is the second weight coefficient; γ is the third weight coefficient; wherein 1=α+β+γ.

7. The intelligent screening method for qualification certificates for construction industry communication project bidding according to claim 6 is characterized in that: The task matching score is the semantic similarity between the job description vector and the qualification certificate capability vector measured by cosine similarity; The ability matching score is a weighted score of the matching degree of the qualification certificate vector of the qualification certificate object in the corresponding fields according to the specific requirements of the job role for the qualification certificate type, professional title level, and mastered construction technology; The environmental matching score is obtained by embedding the historical communication project environmental information of the qualification certificate object and the environmental information of the target communication project into a graph structure through a graph neural network, and calculating the embedding vector similarity value between the two.

8. The intelligent screening method for qualification certificates for construction industry communication project bidding according to claim 7 is characterized in that: The S5 specifically includes: S51. For each of the positions in the proposed team, select M qualification certificate objects in descending order of semantic matching scores with the position role to form a candidate set for the position role, and use the semantic matching scores between the M qualification certificate objects and the position role as the position adaptation score. For each candidate set of job role personnel, divide all the object groups in the candidate set of job role personnel into groups; The number of people in each target group is the number of candidates with the required qualifications for the position roles in the proposed team; Randomly select a group of candidates from each position role candidate set and combine them to obtain a complete team candidate combination set; S52. Obtain a comprehensive score for each complete set of candidate team combinations using a preset comprehensive evaluation function; The comprehensive evaluation score is formed by weighting the following indicators: the candidate team's single-person job adaptation score, team schedule coordination index, multi-position conflict avoidance coefficient, and team historical cooperation tacit understanding; S53. Based on the comprehensive score of each complete team candidate combination set, determine the optimal team combination and the final qualification certificate set; Among them, the optimal team combination is the complete team candidate combination set with the highest comprehensive score; the final qualification certificate set includes the qualification certificates corresponding to each qualification certificate object in the optimal team combination.

9. The intelligent screening method for qualification certificates for construction industry communication project bidding according to claim 6 is characterized in that: The comprehensive evaluation function is: MatchScore k is the job suitability score of the kth qualification certificate candidate in the complete team candidate combination set; XT is the team duration synergy index in the complete team candidate combination set; GB is the multi-position conflict avoidance coefficient in the complete team candidate combination set; LS is the historical cooperation tacit understanding of the team in the complete set of team candidate combinations; δ is the preset first coefficient; ε is the preset second coefficient; θ is the preset third coefficient; ω is the preset fourth coefficient.

10. The intelligent screening method for qualification certificates for construction industry communication project bidding according to claim 9, characterized in that: The calculation method of the team duration coordination index includes: For each certificated subject in the complete team candidate combination set, obtain the estimated end time and reserved buffer period of the current communication project, and determine the time window for entry. Map the time window for entry of each certificated subject to a Gaussian time distribution function. Perform overlap integral calculation on the entry time distribution function between all certificated subjects to obtain the synergy overlap of each pair of certificated subjects. Take the average of the synergy overlap of all pairs of certificated subjects as the team's duration synergy index. The calculation method of the conflict avoidance coefficient includes: Check whether each certificate holder in the complete team candidate set has already held multiple roles in other ongoing communications projects. If so, record it as a conflict. Check whether each certificate holder in the complete team candidate set has multiple mutually exclusive roles in the target communications project. If so, record it as a conflict. Divide the actual number of conflicts by the theoretical maximum number of conflicts and take the inverse proportion to obtain the conflict avoidance coefficient, which ranges from 0 to 1. The calculation method of the cooperative tacit understanding includes: Extract the joint participation records of any two qualification certificate objects in historical communication projects in the complete team candidate combination set; count the historical number of cooperation between each pair of qualification certificate objects and the number of successful cooperation; calculate the cooperation success rate of each pair of qualification certificate objects, and take the weighted average according to the historical number of cooperation to obtain the cooperation tacit understanding of the entire candidate team.

Images