A method and device for dynamically matching ship post demand information

Abstract

The application discloses a kind of dynamic matching method and device of ship post demand information, the method includes: obtaining ship post demand information set and crew ability information set;The ship post demand information set includes the ability demand information sub-set of each ship post;The ability demand information sub-set includes the demand value of all ability indexes on ship post;The crew ability information set includes the ability index information sub-set of each crew;The ability index information sub-set includes the ability value of all ability indexes on crew;The ship post demand information set and crew ability information set are jointly clustered and matched processing, and obtain post crew clustering matching pair set;Based on the post crew clustering matching pair set, crew post ability distribution processing is carried out, and the crew post distribution information set is obtained.

Description

Technical Field

[0001] This invention relates to the fields of industrial data processing, strategy optimization, and data processing technology, specifically to a method and apparatus for dynamically matching ship job demand information. Background Technology

[0002] Against the backdrop of the rapid development of the shipping industry, the safety and efficiency of ship operations place extremely high demands on the suitability of crew members for their positions. Ship positions are highly specialized and unique, with different positions not only having clear requirements for crew members' basic physiological indicators (such as height, weight, and vision), but also strict suitability standards for core competency indicators such as professional skills, legal knowledge, and emergency response capabilities. Any mismatch between personnel and positions can directly lead to navigational safety risks and even cause significant personal and property losses.

[0003] Currently, the matching of ship positions with crew members largely relies on manual experience, lacking systematic theoretical support and scientific technical means. Specifically, existing matching methods have the following shortcomings: First, the consideration of competency indicators is incomplete, often focusing on explicit indicators such as professional skills while neglecting implicit indicators that are equally crucial for navigational safety, such as safety awareness, psychological resilience, and language skills, leading to one-sided matching results. Second, the classification of positions and crew members lacks scientific rigor, often relying on subjective manual division of position types and crew levels, failing to accurately reflect the differences in requirements for different positions and the competency characteristics of crew members. Third, the matching process is inefficient; when faced with a large amount of job requirements and crew information, manual matching is not only time-consuming but also prone to insufficient accuracy due to subjective judgment bias. Fourth, the few existing matching methods that attempt to introduce algorithms mostly use single clustering algorithms for classification, resulting in poor stability of clustering results and lacking a precise matching mechanism between job categories and crew categories, making it difficult to adapt to the dynamic changes in job requirements and crew information during ship operations.

[0004] Furthermore, existing technologies for acquiring and quantifying crew competency information are fragmented and lack unified standards for constructing information sets, making it difficult to effectively integrate competency information from different sources for matching analysis. Simultaneously, the matched job assignments lack a dynamic adjustment mechanism, failing to optimize the allocation results in a timely manner based on changes in crew competency or job requirements, further restricting the rationality and flexibility of ship human resource allocation. Therefore, there is an urgent need for a method that can achieve dynamic and accurate matching between ship job requirements and crew competency to address the numerous problems existing in current technologies. Summary of the Invention

[0005] This invention primarily addresses the problem of how to achieve dynamic and accurate matching between ship job requirements and crew capabilities. This invention discloses a method and apparatus for dynamically matching ship job requirement information.

[0006] In a first aspect, this invention discloses a method for dynamically matching ship job requirement information, comprising:

[0007] S1, obtain a set of information on ship job requirements and a set of information on crew capabilities;

[0008] S2, perform joint clustering matching on the set of ship job requirements information and the set of crew competence information to obtain a set of job crew clustering matching pairs;

[0009] S3. Based on the set of crew member clustering and matching pairs, perform crew member job competency allocation processing to obtain a set of crew member job allocation information.

[0010] The joint clustering and matching process of the set of ship job requirements information and the set of crew competence information yields a set of job-crew clustering matching pairs, including:

[0011] S21, Perform a first clustering process on the set of ship job requirements information to obtain a set of job category information;

[0012] S22, perform a second clustering process on the crew capability information set to obtain a crew category information set;

[0013] S23, perform category matching processing on the job category information set and the crew member category information set to obtain a set of job and crew member cluster matching pairs.

[0014] The first clustering process is performed on the set of ship job requirements information to obtain a set of job category information, including:

[0015] S211, using a preset set of k-means clustering algorithms, the set of ship job requirement information is clustered to obtain a first cluster information set; the first cluster information set includes the clustering results of each clustering algorithm in the set of k-means clustering algorithms; the clustering results include the index of each category and the data vector contained therein; the data vector is obtained by vectorizing the subset of capability requirement information; the index of each category in the clustering results is obtained by numbering the number of quantity vectors contained in the category from high to low;

[0016] S212, perform clustering result fusion processing on the first clustering information set to obtain a job category information set.

[0017] The process of fusing the clustering results of the first clustering information set to obtain a job category information set includes:

[0018] S2121, Perform statistical calculations on the first clustering information set to obtain a category statistical matrix;

[0019] The expression for calculating the element in the i-th row and j-th column of the category statistics matrix is:

[0020] ,

[0021] in, Let be the element in the i-th row and j-th column of the category statistics matrix, and K be the total number of clustering algorithms included in the preset k-means clustering algorithm set. For comparison functions, The index of the category to which the data vector belongs after the k-th clustering algorithm clusters the subset of capability requirement information for the i-th ship position is determined. The index of the category to which the data vector belongs after the k-th clustering algorithm clusters the data vector corresponding to the subset of capability requirement information for the j-th ship position;

[0022] S2122, Perform normalization transformation on the category statistical matrix to obtain a normalized statistical matrix;

[0023] S2123, perform bias elimination processing on the normalized statistical matrix to obtain the statistical feature matrix;

[0024] S2124, Perform eigenvalue decomposition on the statistical feature matrix to obtain a set of eigenvectors;

[0025] S2125, Perform DBSCAN clustering on the feature vector set to obtain the clustering result for each feature vector;

[0026] S2126, Confirm the job category information of the ship positions that have the same serial number as each feature vector, which is the clustering result of the feature vector;

[0027] S2127, using the job category information of all ship positions, a set of job category information is constructed.

[0028] The expression for the normalization transformation is:

[0029] ,

[0030] in, Let be the element in the i-th row and j-th column of the normalized statistical matrix. This represents the vector in the i-th row of the category statistics matrix. This represents the vector of the j-th column of the category statistics matrix. Take the maximum value of the vector;

[0031] The expression for the bias elimination process is:

[0032] ,

[0033] in, It is a diagonal matrix. , Let represent the element in the i-th row and i-th column of the diagonal matrix. This is a statistical characteristic matrix.

[0034] The process of performing category matching on the job category information set and the crew member category information set to obtain a set of job and crew member cluster matching pairs includes:

[0035] S221, calculate the center point of all ship positions contained in each job category information set to obtain the center vector of the job category information;

[0036] S222, calculate the center point of all data vectors corresponding to crew members included in each category of the crew member category information set to obtain the center vector of the category information;

[0037] S223, For the center vector of each job category information, calculate the center vector of the category information that has the minimum Euclidean distance to the center vector, and determine the category information as the matching category information of the job category information;

[0038] S224, Perform job allocation processing on each job category information and its corresponding matching category information to obtain the crew allocation result of the job category information;

[0039] S225. Using the crew allocation results of all job category information, construct a set of job crew cluster matching pairs.

[0040] The process of assigning positions to each job category and its corresponding matching category to obtain the crew allocation results for the job category information includes:

[0041] S2241, Set the job number to be assigned, g=1;

[0042] S2242, Based on the subset of capability requirement information of the job category information sorted by g, all crew members included in the matching category information are assigned to obtain the crew member information corresponding to the job.

[0043] S2243, delete the crew member corresponding to the crew member information from the matching category information, and increase the job position number g to be assigned by 1;

[0044] S2244, determine whether g is greater than the total number of all positions included in the job category information, and obtain the first discrimination result;

[0045] If the first determination result is negative, execute S2242;

[0046] If the first determination result is yes, the crew position allocation is completed, and the crew allocation result of the position category information is constructed by using the crew information corresponding to all positions in the position category information.

[0047] A second aspect of the present invention discloses a dynamic matching device for ship job requirement information, the device comprising:

[0048] Memory containing executable program code;

[0049] A processor coupled to the memory;

[0050] The processor calls the executable program code stored in the memory to execute the dynamic matching method for ship job requirements information.

[0051] In a third aspect, the present invention discloses a computer-readable storage medium storing computer instructions, which, when invoked by a computer, are used to execute the dynamic matching method for ship job requirement information.

[0052] In a fourth aspect of this invention, an information data processing terminal is disclosed, which is used to implement the dynamic matching method for ship job requirement information.

[0053] The beneficial effects of this invention are as follows:

[0054] This invention constructs a comprehensive set of ship job requirements information and a set of crew competency information, achieving a holistic consideration of job requirements and crew competency, effectively improving the comprehensiveness and accuracy of the matching results. This invention clarifies that the job requirements information set covers the required values ​​for each job across multiple dimensions of competency indicators, including height, weight, professional skills, vision, familiarity with regulations, emergency response capabilities, safety awareness, psychological resilience, and language skills. Simultaneously, the crew competency information set includes the competency values ​​of each crew member in the aforementioned indicators. Compared to existing matching methods that only focus on some explicit indicators, this invention more completely reflects the characteristics of job requirements and the overall picture of crew competency, providing a solid data foundation for accurate matching.

[0055] This invention significantly improves the scientific rigor and accuracy of job and crew classification by employing a joint clustering matching strategy, combining a multi-algorithm fusion clustering approach with a scientific category matching mechanism. First, it uses a set of six k-means clustering algorithms to perform the first clustering process on the ship's job requirement information set. Then, it fuses the clustering results through statistical calculations, normalization transformation, bias elimination, eigenvalue decomposition, and DBSCAN clustering, effectively avoiding classification bias caused by a single clustering algorithm and improving the stability and reliability of job category division. Second, it uses the PCA method for the crew capability information set, achieving data dimensionality reduction while retaining key capability information, thus improving clustering efficiency. Finally, by calculating the center vectors of job categories and crew categories and determining the matching category based on Euclidean distance, it ensures the accuracy of category-level matching, providing a scientific basis for subsequent job allocation.

[0056] This invention achieves dynamic and precise matching between positions and crew members through a refined crew competency allocation process, effectively improving human resource allocation efficiency. Based on category matching, this invention ensures that each position is matched with the most suitable crew member through steps such as sequential allocation, optimal matching degree calculation, and removal of already assigned crew members, while avoiding duplicate allocation. Position ranking can be prioritized based on distance from the central vector. The matching degree calculation comprehensively considers the mean, variance, and differences of various dimensions of the data vector, accurately quantifying the suitability between crew members and positions. Compared to subjective human judgment or simple indicator comparison, the allocation results are more scientific. The entire matching process is fully automated, from information acquisition and clustering classification to allocation adjustment, significantly reducing matching time. It is particularly suitable for matching scenarios with large-scale job requirements and crew information, significantly improving matching efficiency. Attached Figure Description

[0057] Figure 1 This is a flowchart illustrating the implementation of the method of the present invention. Detailed Implementation

[0058] To better understand the content of this invention, an embodiment is provided here.

[0059] Figure 1 This is a flowchart illustrating the implementation of the method of the present invention.

[0060] In a first aspect, this invention discloses a method for dynamically matching ship job requirement information, comprising:

[0061] S1, obtain a set of information on ship job requirements and a set of information on crew capabilities;

[0062] S2, perform joint clustering matching on the set of ship job requirements information and the set of crew competence information to obtain a set of job crew clustering matching pairs;

[0063] S3. Based on the set of crew member clustering and matching pairs, perform crew member job competency allocation processing to obtain a set of crew member job allocation information.

[0064] The aforementioned set of ship job requirement information can be obtained by retrieving the requirement information from the user manuals or databases of the corresponding equipment for each ship job.

[0065] The set of ship job requirement information includes a subset of capability requirement information for each ship job; the subset of capability requirement information includes the requirement values ​​for all capability indicators for each ship job.

[0066] The set of crew capability information includes a subset of capability index information for each crew member; the subset of capability index information includes the capability values ​​of the crew member across all capability indicators.

[0067] The aforementioned ability indicators include height, weight, profession, vision, familiarity with laws and regulations, emergency response capability, safety awareness, psychological stress tolerance, and language ability.

[0068] The crew members' familiarity with regulations, emergency response capabilities, safety awareness, psychological resilience, and language skills can be obtained through quantitative testing or by searching relevant crew member databases.

[0069] The joint clustering and matching process of the set of ship job requirements information and the set of crew competence information yields a set of job-crew clustering matching pairs, including:

[0070] S21, Perform a first clustering process on the set of ship job requirements information to obtain a set of job category information;

[0071] S22, perform a second clustering process on the crew capability information set to obtain a crew category information set;

[0072] S23, perform category matching processing on the job category information set and the crew member category information set to obtain a set of job and crew member cluster matching pairs.

[0073] The second clustering process can be performed using the PCA method.

[0074] The first clustering process is performed on the set of ship job requirements information to obtain a set of job category information, including:

[0075] S211, using a preset set of k-means clustering algorithms, the set of ship job requirement information is clustered to obtain a first cluster information set; the first cluster information set includes the clustering results of each clustering algorithm; the clustering results include the index and data vector of each category; the data vector is obtained by vectorizing the subset of capability requirement information; the index of each category in the clustering results is obtained by numbering the total number of quantity vectors contained in the category from high to low;

[0076] S212, perform clustering result fusion processing on the first clustering information set to obtain a job category information set;

[0077] The category with the largest total number of included quantity vectors is numbered 1;

[0078] The clustering process for the set of ship job requirements information or crew competency information is achieved by representing each subset of ship job competency requirements information or crew competency index information as a corresponding data vector and then clustering the data vectors.

[0079] The vectorization representation of the subset of capability requirement information or the subset of capability indicator information involves numerically encoding the text data in the subset of capability requirement information or the subset of capability indicator information to obtain the corresponding elements in the vector, and using all the data in the subset of capability requirement information to represent it as a data vector.

[0080] The preset set of k-means clustering algorithms includes six clustering algorithms: k-centroid algorithm, k-median algorithm, k-modes algorithm, fuzzy k-means algorithm, and kernel k-means algorithm.

[0081] The first clustering information set includes the clustering results of each clustering algorithm; the clustering results include the index and data vector of each category; the data vector is obtained by vectorizing the capability requirement information subset; the index of each category in the clustering results is obtained by numbering the number of quantity vectors contained in the category from high to low.

[0082] The crew category information set includes the category information described in the data vector corresponding to the capability indicator information subset of each crew member.

[0083] The process of fusing the clustering results of the first clustering information set to obtain a job category information set includes:

[0084] S2121, Perform statistical calculations on the first clustering information set to obtain a category statistical matrix;

[0085] The expression for calculating the element in the i-th row and j-th column of the category statistics matrix is:

[0086] ,

[0087] in, Let be the element in the i-th row and j-th column of the category statistics matrix, and K be the total number of clustering algorithms included in the preset k-means clustering algorithm set. For comparison functions, The index of the category to which the data vector belongs after the k-th clustering algorithm clusters the subset of capability requirement information for the i-th ship position is determined. The index of the category to which the data vector belongs after the k-th clustering algorithm clusters the data vector corresponding to the subset of capability requirement information for the j-th ship position;

[0088] The formula used for calculating the category statistical matrix effectively integrates the clustering results of different algorithms by statistically averaging the consistency of category assignments of job data vectors under various clustering algorithms, reducing the impact of the randomness of a single algorithm on the classification results. This formula compares whether different jobs are classified into the same category in each algorithm and takes the average, preserving the information from each algorithm's judgment of job similarity while mitigating the extreme results of individual algorithms through mean calculation. This results in a category statistical matrix that more objectively reflects the true similarity between jobs, laying a reliable foundation for subsequent cluster fusion.

[0089] S2122, Perform normalization transformation on the category statistical matrix to obtain a normalized statistical matrix;

[0090] S2123, perform bias elimination processing on the normalized statistical matrix to obtain the statistical feature matrix;

[0091] S2124, Perform eigenvalue decomposition on the statistical feature matrix to obtain a set of eigenvectors;

[0092] S2125, Perform DBSCAN clustering on the feature vector set to obtain the clustering result for each feature vector;

[0093] S2126, Confirm the job category information of the ship positions with the same serial number as the feature vector, which is the clustering result of the feature vector;

[0094] S2127, Using the job category information of all ship positions, a job category information set is constructed. The job category information set includes several ship positions.

[0095] The expression for the normalization transformation is:

[0096] ,

[0097] in, Let be the element in the i-th row and j-th column of the normalized statistical matrix. This represents the vector in the i-th row of the category statistics matrix. This represents the vector of the j-th column of the category statistics matrix. Take the maximum value of the vector;

[0098] The formula used for normalization scaling eliminates the magnitude difference in clustering frequency between different job positions by scaling the elements of the category statistics matrix against the products of the corresponding row and column maximum values. This ensures that all elements in the matrix are on a directly comparable, uniform scale. This process avoids the numerical dominance problem caused by some job positions being frequently grouped into the same category in most algorithms, ensuring that the similarity statistics between each job pair accurately reflect their relative importance after normalization, thus improving the accuracy of subsequent matrix processing.

[0099] The expression for the bias elimination process is:

[0100] ,

[0101] in, It is a diagonal matrix. , Let represent the element in the i-th row and i-th column of the diagonal matrix. This is a statistical characteristic matrix.

[0102] This represents the summation of elements in each row of matrix A based on their column indices. The comparison function has the following expression:

[0103]

[0104] The expression for the eigenvalue decomposition is:

[0105]

[0106] The set of eigenvalues ​​is represented as follows: .

[0107] The formula used for bias elimination, by subtracting the diagonal matrix formed by summing the elements of each row from the normalized statistical matrix, effectively removes the interference of the clustering characteristics of the job positions themselves on similarity judgment. This process eliminates the bias caused by individual jobs being more easily grouped with other jobs in clusters, making the resulting statistical feature matrix more focused on reflecting the relative similarity between jobs, rather than the absolute clustering tendency of the jobs themselves, thus providing a cleaner data foundation for subsequent feature extraction.

[0108] The process of performing category matching on the job category information set and the crew member category information set to obtain a set of job and crew member cluster matching pairs includes:

[0109] S221, calculate the center point of all ship positions contained in each job category information set to obtain the center vector of the job category information;

[0110] S222, calculate the center point of all the data vectors of all the crew members contained in each category of the crew member category information set to obtain the center vector of the category information;

[0111] S223, For the center vector of each job category information, calculate the center vector of the category information with the minimum Euclidean distance from the center vector, and determine the category information as the matching category information of the job category information; a matching category information includes several crew members;

[0112] S224, Perform job allocation processing on each job category information and its corresponding matching category information to obtain the crew allocation result of the job category information;

[0113] S225. Using the crew allocation results of all job category information, construct a set of job crew cluster matching pairs.

[0114] The process of assigning positions to each job category and its corresponding matching category to obtain the crew allocation results for the job category information includes:

[0115] S2241, Set the job number to be assigned, g=1;

[0116] S2242, Based on the subset of capability requirement information of the job category information sorted by g, all crew members included in the matching category information are assigned to obtain the crew member information corresponding to the job.

[0117] S2243, delete the crew member corresponding to the crew member information from the matching category information, and increase the job position number g to be assigned by 1;

[0118] S2244, determine whether g is greater than the total number of all positions included in the job category information, and obtain the first discrimination result;

[0119] S2245, if the first determination result is negative, execute S2242;

[0120] If the first determination result is yes, the crew position allocation is completed, and the crew allocation result of the position category information is constructed by using the crew information corresponding to all positions in the position category information;

[0121] The job number in the job category information can be generated randomly, or sorted from smallest to largest according to its distance from the center vector.

[0122] The calculation of the center vector of category information with the minimum Euclidean distance to the center vector is performed by calculating the Euclidean distance between the center vector based on job category information and the center vector of each category information, and selecting the center vector of category information with the minimum Euclidean distance.

[0123] The calculation of the center point can be achieved using algorithms that calculate the geometric or arithmetic centers of multiple vectors.

[0124] The process involves allocating all crew members included in the matching category information to the subset of job requirements sorted as g in the job category information, thereby obtaining the crew member information corresponding to the job, including:

[0125] The matching degree is calculated between the data vector corresponding to the subset of competency requirements information of the job category information ranked g and the data vector corresponding to the subset of competency index information of each crew member contained in the matching category information, so as to obtain the matching information of each crew member.

[0126] The information of the crew member with the highest matching value is determined, and this information corresponds to the crew member information for the position.

[0127] The crew information includes crew identification information or crew number, etc.

[0128] The expression for calculating the matching degree is:

[0129] ,

[0130] in, For crew matching information, and These are the mean values ​​of the data vectors corresponding to the subset of competency requirements information for positions ranked g in the job category information, and the mean values ​​of the data vectors corresponding to the subset of crew competency indicator information. and These are the variances of the data vectors corresponding to the subset of competency requirements information for positions ranked g in the job category information, and the variances of the data vectors corresponding to the subset of crew competency index information. and These are the i-th item of the data vector corresponding to the subset of competency requirements information for job positions sorted as g in the job category information, and the i-th item of the data vector corresponding to the subset of crew competency index information, respectively, where N1 is the length of the data vector.

[0131] The matching degree calculation formula comprehensively quantifies the degree of compatibility between job positions and crew members by considering the mean difference, variance difference, and specific differences in each dimension of the data vectors. This formula not only focuses on the similarity of the overall data distribution (reflected by the mean and variance) but also takes into account the matching of specific competency indicators (reflected by the differences in each dimension), avoiding the bias caused by judging from a single dimension. Furthermore, by transforming the differences through a reasonable function, the matching degree numerical value can intuitively reflect the level of compatibility, providing a quantitative standard for the accurate matching of job positions and crew members.

[0132] In all embodiments of the present invention, the variables involved in all computational expressions or mathematical functions have been dimensionlessized before computation.

[0133] In all embodiments of the present invention, the values ​​of the independent variables in the input of all computational expressions or mathematical functions meet the reasonable requirements of the input range of the computational expressions or mathematical functions, and can ensure that the computational expressions or mathematical functions can be calculated smoothly without violating physical laws or mathematical rules.

[0134] A second aspect of the present invention discloses a dynamic matching device for ship job requirement information, the device comprising:

[0135] Memory containing executable program code;

[0136] A processor coupled to the memory;

[0137] The processor calls the executable program code stored in the memory to execute the dynamic matching method for ship job requirements information.

[0138] In a third aspect, the present invention discloses a computer-readable storage medium storing computer instructions, which, when invoked by a computer, are used to execute the dynamic matching method for ship job requirement information.

[0139] In a fourth aspect of this invention, an information data processing terminal is disclosed, which is used to implement the dynamic matching method for ship job requirement information.

[0140] The above description is merely an embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of the present invention should be included within the scope of the claims of the present invention.

Claims

1. A method for dynamically matching ship job requirement information, characterized in that, include: S1, obtain a set of information on ship job requirements and a set of information on crew capabilities; The set of ship job requirement information includes a subset of capability requirement information for each ship job; the subset of capability requirement information includes the requirement values ​​for each ship job across all capability indicators; the set of crew capability information includes a subset of capability indicator information for each crew member; the subset of capability indicator information includes the capability values ​​for each crew member across all capability indicators. S2, perform joint clustering matching on the set of ship job requirements information and the set of crew competence information to obtain a set of job-crew clustering matching pairs, including: S21, perform a first clustering process on the set of ship job requirement information to obtain a set of job category information, including: S211, using a preset set of k-means clustering algorithms, the set of ship job requirement information is clustered to obtain a first cluster information set; the first cluster information set includes the clustering results of each clustering algorithm in the set of k-means clustering algorithms; the clustering results include the index of each category and the data vector contained therein; the data vector is obtained by vectorizing the subset of capability requirement information; the index of each category in the clustering results is obtained by numbering the number of quantity vectors contained in the category from high to low; S212, perform clustering result fusion processing on the first clustering information set to obtain a job category information set; S22, perform a second clustering process on the set of crew capability information to obtain a set of crew category information, including: S221, calculate the center point of all ship positions contained in each job category information set to obtain the center vector of the job category information; S222, calculate the center point of all data vectors corresponding to crew members included in each category of the crew member category information set to obtain the center vector of the category information; S223, For the center vector of each job category information, calculate the center vector of the category information that has the minimum Euclidean distance to the center vector, and determine the category information as the matching category information of the job category information; S224, Perform job allocation processing on each job category information and its corresponding matching category information to obtain the crew allocation result of the job category information; S225, using the crew allocation results of all job category information, construct a set of job crew cluster matching pairs; S23, perform category matching processing on the job category information set and the crew category information set to obtain a set of job and crew cluster matching pairs; S3. Based on the set of crew member clustering and matching pairs, perform crew member job competency allocation processing to obtain a set of crew member job allocation information.

2. The dynamic matching method for ship job requirement information as described in claim 1, characterized in that, The process of fusing the clustering results of the first clustering information set to obtain a job category information set includes: S2121, Perform statistical calculations on the first clustering information set to obtain a category statistical matrix; The expression for calculating the element in the i-th row and j-th column of the category statistics matrix is: ， in, Let be the element in the i-th row and j-th column of the category statistics matrix, and K be the total number of clustering algorithms included in the preset k-means clustering algorithm set. For comparison functions, The index of the category to which the data vector belongs after the k-th clustering algorithm clusters the subset of capability requirement information for the i-th ship position is determined. The index of the category to which the data vector belongs after the k-th clustering algorithm clusters the data vector corresponding to the subset of capability requirement information for the j-th ship position; S2122, Perform normalization transformation on the category statistical matrix to obtain a normalized statistical matrix; S2123, perform bias elimination processing on the normalized statistical matrix to obtain the statistical feature matrix; S2124, Perform eigenvalue decomposition on the statistical feature matrix to obtain a set of eigenvectors; S2125, Perform DBSCAN clustering on the feature vector set to obtain the clustering result for each feature vector; S2126, Confirm the job category information of the ship positions that have the same serial number as each feature vector, which is the clustering result of the feature vector; S2127, using the job category information of all ship positions, a set of job category information is constructed.

3. The dynamic matching method for ship job requirement information as described in claim 2, characterized in that, The expression for the normalization transformation is: ， in, Let be the element in the i-th row and j-th column of the normalized statistical matrix. This represents the vector in the i-th row of the category statistics matrix. This represents the vector of the j-th column of the category statistics matrix. Take the maximum value of the vector; The expression for the bias elimination process is: ， in, It is a diagonal matrix. , Let represent the element in the i-th row and i-th column of the diagonal matrix. This is a statistical characteristic matrix.

4. The dynamic matching method for ship job requirement information as described in claim 1, characterized in that, The process of assigning positions to each job category and its corresponding matching category to obtain the crew allocation results for the job category information includes: S2241, Set the job number to be assigned to g=1; S2242, Based on the subset of capability requirement information of the job category information sorted by g, all crew members included in the matching category information are assigned to obtain the crew member information corresponding to the job. S2243, delete the crew member corresponding to the crew member information from the matching category information, and increase the job position number g to be assigned by 1; S2244, determine whether g is greater than the total number of all positions included in the job category information, and obtain the first discrimination result; If the first determination result is negative, execute S2242; If the first determination result is yes, the crew position allocation is completed, and the crew allocation result of the position category information is constructed by using the crew information corresponding to all positions in the position category information.

5. A dynamic matching device for ship job requirement information, characterized in that, The device includes: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the dynamic matching method for ship job requirement information as described in any one of claims 1 to 4.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions, which, when invoked by a computer, are used to execute the dynamic matching method for ship job requirement information as described in any one of claims 1 to 4.

Images