Project cluster expert matching method and device, equipment, medium and program product

Abstract

The invention provides an expert matching method and device for a project cluster, equipment, a medium and a program product, and relates to the technical field of software information. According to the method, project text information of a target project cluster and expert information corresponding to an expert set are obtained, and based on a predefined field keyword dictionary, according to the project text information and the expert information, a first relevance vector of a project and a field and a second relevance vector of an expert and the field are determined; according to the first correlation vector and the second correlation vector, generating an initial matching cost matrix of the project and the expert, and further performing optimization processing on the initial matching cost matrix to obtain a target matching cost matrix of the project and the expert for representing an expert matching result corresponding to each project in the target project cluster; the specific matching requirement of each project in the target project cluster is met, the expert matching accuracy is improved, and meanwhile the expert matching coverage rate of the target project cluster is improved.

Description

Technical Field

[0001] This application relates to the field of software information technology, and particularly to an expert matching method, device, equipment, medium and program product for a project cluster. Background Art

[0002] In the synchronous evaluation stage of multiple technical projects, it is crucial to screen a group of highly specialized and knowledgeable experts to participate in the evaluation. Matching suitable experts for each project is crucial for providing a comprehensive and accurate evaluation of complex technical projects. However, as the scale and complexity of the evaluation increase, the process of manually matching experts with projects becomes inefficient and error-prone. Therefore, a systematic, algorithm-based method must be adopted to ensure that projects can be evaluated by the most relevant experts in a timely manner.

[0003] In the related art, a combinatorial optimization method is adopted to match experts for a project cluster. However, the inventor's research found that this method has the problem of low coverage rate of the expert set for the project cluster. Summary of the Invention

[0004] This application provides an expert matching method, device, equipment, medium and program product for a project cluster, aiming to solve the problem of low coverage rate of the expert set for the project cluster in the related art.

[0005] In a first aspect, this application provides an expert matching method for a project cluster, including: obtaining the project text information of a target project cluster and the expert information corresponding to an expert set; based on a predefined domain keyword dictionary, determining a first relevance vector between the project and the domain and a second relevance vector between the expert and the domain according to the project text information and the expert information; generating an initial matching cost matrix between the project and the expert according to the first relevance vector and the second relevance vector; performing an optimization process on the initial matching cost matrix to obtain a target matching cost matrix between the project and the expert, where the target matching cost matrix is used to represent the expert matching result corresponding to each project in the target project cluster.

[0006] In a possible implementation manner, based on a predefined domain keyword dictionary, determining a first relevance vector between the project and the domain and a second relevance vector between the expert and the domain according to the project text information and the expert information includes: based on the predefined domain keyword dictionary, generating a first domain relevance matrix corresponding to the target project cluster according to the project text information, and determining the first relevance vector according to the first domain relevance matrix; based on the predefined domain keyword dictionary, generating a second domain relevance matrix corresponding to the expert according to the expert information, and determining the second relevance vector according to the second domain relevance matrix.

[0007] In a possible implementation manner, generating an initial matching cost matrix of items and experts according to a first correlation vector and a second correlation vector includes: for each item in the target item cluster and each expert in the expert set, determining the similarity between the item and the expert according to the first correlation vector corresponding to the item and the second correlation vector corresponding to the expert; determining the matching cost between the item and the expert according to the similarity; and generating an initial matching cost matrix according to the matching cost.

[0008] In a possible implementation manner, after optimizing the initial matching cost matrix to obtain a target matching cost matrix of items and experts, it further includes: determining the number of first items corresponding to the covered items in the target item cluster according to the target matching cost matrix; calculating the percentage of the number of first items in the number of second items, and determining the percentage as the expert matching coverage rate, where the number of second items is the total number of items corresponding to the target item cluster.

[0009] In a possible implementation manner, determining the number of first items corresponding to the covered items in the target item cluster according to the target matching cost matrix includes: for each item in the target item cluster, determining the minimum matching cost corresponding to the item from the target matching cost matrix; determining whether the minimum matching cost is less than a preset threshold; if it is less than, marking the item as covered; and determining the number of all items marked as covered as the number of first items.

[0010] In a possible implementation manner, the expert matching method for the item cluster further includes: if the minimum matching cost is greater than or equal to the preset threshold, marking the item as not covered.

[0011] In a second aspect, the present application provides an expert matching device for an item cluster, including:

[0012] An acquisition module, configured to acquire the item text information of the target item cluster and the expert information corresponding to the expert set;

[0013] A determination module, configured to determine a first correlation vector between an item and a field and a second correlation vector between an expert and a field based on a predefined domain keyword dictionary according to the item text information and the expert information;

[0014] A generation module, configured to generate an initial matching cost matrix of items and experts according to the first correlation vector and the second correlation vector;

[0015] A processing module, configured to optimize the initial matching cost matrix to obtain a target matching cost matrix of items and experts, where the target matching cost matrix is used to represent the expert matching result corresponding to each item in the target item cluster.

[0016] In a possible implementation, the determination module is specifically configured to: based on a predefined domain keyword dictionary, generate a first domain correlation matrix corresponding to the target project cluster according to the project text information, and determine a first correlation vector according to the first domain correlation matrix; based on the predefined domain keyword dictionary, generate a second domain correlation matrix corresponding to the expert according to the expert information, and determine a second correlation vector according to the second domain correlation matrix.

[0017] In a possible implementation, the generation module is specifically configured to: for each project in the target project cluster and each expert in the expert set, determine the similarity between the project and the expert according to the first correlation vector corresponding to the project and the second correlation vector corresponding to the expert; determine the matching cost between the project and the expert according to the similarity; generate an initial matching cost matrix according to the matching cost.

[0018] In a possible implementation, the expert matching device for the project cluster further includes a calculation module (not shown). After optimizing the initial matching cost matrix to obtain the target matching cost matrix of the project and the expert, the calculation module is configured to: determine the first project quantity corresponding to the covered projects in the target project cluster according to the target matching cost matrix; calculate the percentage of the first project quantity in the second project quantity, and determine the percentage as the expert matching coverage rate, where the second project quantity is the total number of projects corresponding to the target project cluster.

[0019] In a possible implementation, the expert matching device for the project cluster further includes a marking module (not shown). The marking module is configured to: for each project in the target project cluster, determine the minimum matching cost corresponding to the project from the target matching cost matrix; determine whether the minimum matching cost is less than a preset threshold; if less, mark the project as covered; determine the quantity of all projects marked as covered as the first project quantity.

[0020] In a possible implementation, the marking module is further configured to: when the minimum matching cost is greater than or equal to the preset threshold, mark the project as not covered.

[0021] In a third aspect, the present application provides an electronic device, including a processor and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the method provided in the first aspect as described above.

[0022] In a fourth aspect, the present application provides a computer-readable storage medium, in which computer-executable instructions are stored, and when the computer-executable instructions are executed by a processor, they are used to implement the method provided in the first aspect as described above.

[0023] In a fifth aspect, the present application provides a computer program product, including a computer program which, when executed by a processor, implements the method provided in the first aspect as described above.

[0024] The method, apparatus, device, medium and program product for expert matching in a project cluster provided by the present application obtain the project text information of a target project cluster and the expert information corresponding to an expert set. Based on a predefined domain keyword dictionary, according to the project text information and the expert information, a first relevance vector between the project and the domain and a second relevance vector between the expert and the domain are determined. And according to the first relevance vector and the second relevance vector, an initial matching cost matrix between the project and the expert is generated. Further, the initial matching cost matrix is optimized to obtain a target matching cost matrix between the project and the expert, which is used to represent the expert matching result corresponding to each project in the target project cluster. In the present application, based on the predefined domain keyword dictionary, combining the project text information and the expert information, a first relevance vector between the project and the domain and a second relevance vector between the expert and the domain in a multi-dimensional domain are extracted, and an initial matching cost matrix between the project and the expert is generated based on the first relevance vector and the second relevance vector to meet the specific matching requirements of each project in the target project cluster, improve the accuracy of expert matching. At the same time, further by optimizing the initial matching cost matrix, a target matching cost matrix between the project and the expert is obtained, and the expert matching coverage rate of the target project cluster is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The drawings herein are incorporated into the specification and form a part of the specification, showing embodiments consistent with the present application and, together with the specification, are used to explain the principles of the present application.

[0026] Figure 1 is a schematic flowchart of the method for expert matching in a project cluster provided by an embodiment of the present application Figure 1 ;

[0027] Figure 2 is a schematic flowchart of the method for expert matching in a project cluster provided by an embodiment of the present application Figure 2 ;

[0028] Figure 3 is a schematic flowchart of the method for expert matching in a project cluster provided by an embodiment of the present application Figure 3 ;

[0029] Figure 4 is a schematic structural diagram of the device for expert matching in a project cluster provided by an embodiment of the present application;

[0030] Figure 5 is a schematic structural diagram of an electronic device provided by an embodiment of the present application.

[0031] Through the above-mentioned accompanying drawings, specific embodiments of the present application have been shown, and will be described in more detail hereinafter. These drawings and the written description are not intended to limit the scope of the concept of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by reference to specific embodiments. Detailed Description of the Invention

[0032] Here, exemplary embodiments will be described in detail, and examples are shown in the accompanying drawings. When the following description refers to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0033] In the related art, for the method of matching experts for a project cluster using a greedy algorithm, the matching results lack effectiveness and reliability; for the method of matching experts for a project cluster using a combinatorial optimization method, there are problems that may lead to unmet project requirements or an inability to obtain an optimal solution within a limited time. Especially when the number of projects and experts is large, the matching results are prone to falling into a local optimal solution and unable to achieve a global optimal solution. At the same time, the matching results may ignore the detailed requirements of each project, resulting in unmet matching for individual projects, affecting the accuracy of the final result, and the coverage rate of expert matching is relatively low.

[0034] Based on the problems existing in the related art, in the embodiments of the present application, by based on a predefined domain keyword dictionary, combining project text information and expert information, the correlation vectors between projects and domains in multiple dimensions and the correlation vectors between experts and domains are respectively extracted, and an initial matching cost matrix between projects and experts is generated based on the extracted correlation vectors to meet the specific matching requirements of each project in the project cluster and improve the accuracy of expert matching. At the same time, by further optimizing the initial matching cost matrix, a target matching cost matrix between projects and experts is obtained, and the coverage rate of expert matching for the project cluster is improved.

[0035] First, the application scenario of the embodiments of the present application will be described below.

[0036] The method for matching experts for a project cluster provided in the embodiments of the present application is applicable to scenarios such as scientific and technological project review, technical evaluation, and management strategy formulation that require screening expert members to participate in project evaluation.

[0037] Next, the method for matching experts for a project cluster provided in the embodiments of the present application will be described in detail with reference to specific embodiments.

[0038] Figure 1 Flow schematic of the method for matching experts for a project cluster provided in the embodiments of the present applicationFigure 1 As Figure 1 shown, the specific implementation of the expert matching method for this project cluster may include the following steps:

[0039] S101. Obtain the project text information of the target project cluster and the expert information corresponding to the expert set.

[0040] It can be understood that the target project cluster contains multiple different projects, and the multiple different projects may involve different professional fields.

[0041] Exemplarily, the project text information may be the text information corresponding to each project in the project cluster.

[0042] Exemplarily, the text information corresponding to the project includes, but is not limited to, one or more of the information such as the project title, project field, project summary, and responsible agency.

[0043] It can be understood that in the evaluation of scientific and technological projects, the declarant usually needs to submit a project proposal and a feasibility report, as well as a project summary and research content, etc.

[0044] Exemplarily, the expert information may be the expert background information corresponding to each expert in the expert set, and the expert background information includes, but is not limited to, one or more of the information such as educational background, professional experience, technical achievements, professional title, and field contributions.

[0045] S102. Based on the predefined domain keyword dictionary, determine the first correlation vector between the project and the domain and the second correlation vector between the expert and the domain according to the project text information and the expert information.

[0046] First, the predefined domain keyword dictionary will be described below.

[0047] The predefined domain keyword dictionary is a predefined dictionary that contains specific keywords for each professional field, and the keywords therein are words that can be used to help identify and extract information related to a specific field in text processing.

[0048] Exemplarily, the acquisition sources of the predefined domain keyword dictionary may include, but are not limited to, one or more of the reference materials of standardization institutions, literature research, and expert consultation.

[0049] Exemplarily, the first correlation vector between the project and the domain is the correlation between each project in the target project cluster and the domain, and this first correlation vector is used to represent the characteristics or requirements of the project in different fields.

[0050] In a possible implementation, the first correlation vector between projects and fields can be obtained as follows: for each project in the target project cluster, calculate the correlation between the project and each field, and form the first correlation vector.

[0051] Exemplarily, the second correlation vector between experts and fields, i.e., the correlation between each expert in the expert set and the fields, is used to represent the knowledge or ability of the experts in different fields.

[0052] In a possible implementation, the second correlation vector between experts and fields can be obtained as follows: for each expert in the expert set, calculate the correlation between the expert and each field, and form the second correlation vector.

[0053] S103. Generate an initial matching cost matrix between projects and experts according to the first correlation vector and the second correlation vector.

[0054] Exemplarily, the initial matching cost matrix between projects and experts can be a matrix used to represent and evaluate the matching degree between projects and experts.

[0055] Exemplarily, the matrix structure of the initial matching cost matrix between projects and experts can be a two-dimensional array. Among them, the rows of the matrix represent different projects, the columns of the matrix represent different experts, and each element in the matrix is used to represent the matching cost between a project and an expert.

[0056] Exemplarily, the matching cost can be a specific value used to quantify the matching degree between a project and an expert.

[0057] Exemplarily, the lower the matching cost, the higher the matching degree between the project and the expert.

[0058] In a possible implementation, first, based on the first correlation vector and the second correlation vector, calculate the matching cost between projects and experts, and then traverse all the calculated matching costs between projects and experts, and fill each element in the matrix to generate the initial matching cost matrix between projects and experts.

[0059] S104. Optimize the initial matching cost matrix to obtain the target matching cost matrix between projects and experts, and this target matching cost matrix is used to represent the expert matching results corresponding to each project in the target project cluster.

[0060] Exemplarily, perform global optimization on the initial matching cost matrix to minimize the total matching cost of the target project cluster and experts, so as to realize the optimal expert group screening and high expert matching coverage rate under multiple project clusters.

[0061] In one possible implementation, the initial matching cost matrix is input into a global optimization algorithm for optimization processing to obtain the target matching cost matrix of projects and experts output by the global optimization algorithm.

[0062] Exemplarily, the global optimization algorithm can be an optimization algorithm for solving the assignment problem.

[0063] The implementation steps of the optimization algorithm for solving the assignment problem are explained below.

[0064] Exemplarily, the implementation of the optimization algorithm for solving the assignment problem may include the following steps:

[0065] 1) Input the matching cost matrix of projects and experts into the algorithm. Among them, the initialized matching matrix contains n projects and l experts, and the number of experts required for project review is N.

[0066] Exemplarily, the matching cost matrix of projects and experts can be the initial matching cost matrix of projects and experts.

[0067] 2) Initialize the matching cost matrix.

[0068] Construct an empty matrix as the matching cost matrix, and each element in the matching cost matrix is used to represent the matching relationship between projects and experts.

[0069] 3) Start from the unmatched projects and expand the alternating tree.

[0070] Select an unmatched project and start constructing an alternating tree. Among them, the alternating tree is used to find an augmenting path.

[0071] 4) Mark each project-expert pair, construct an equality subgraph, and find an augmenting path.

[0072] Exemplarily, mark the projects and experts to construct an equality subgraph. Among them, the equality subgraph is a subgraph containing all zero-cost edges.

[0073] Exemplarily, find an augmenting path in the equality subgraph. Among them, the augmenting path is a path from an unmatched project to an unmatched expert, where the matched and unmatched edges alternate.

[0074] 5) Augment the matching cost matrix by flipping the matched and unmatched edges on the augmenting path.

[0075] Exemplarily, in response to finding an augmenting path, increase the size of the matching cost matrix by flipping the matched and unmatched edges on the path.

[0076] 6) Adjust the marking to increase the size of the equality subgraph.

[0077] If an augmenting path cannot be found, adjust the labels to change the size of the equality subgraph.

[0078] Exemplarily, it is achieved by adjusting the minimum values of the uncovered rows and columns.

[0079] 7) Output the list of expert groups required for the project cluster according to the final matching cost matrix.

[0080] When all projects are matched, the matching cost matrix contains the optimal project-expert matching scheme.

[0081] Output the matching cost matrix to obtain the list of expert groups required for each project.

[0082] In the embodiment of the present application, by obtaining the project text information of the target project cluster and the expert information corresponding to the expert set, based on the predefined domain keyword dictionary, according to the project text information and the expert information, determine the first correlation vector between the project and the domain and the second correlation vector between the expert and the domain, and according to the first correlation vector and the second correlation vector, generate the initial matching cost matrix of the project and the expert, and further optimize the initial matching cost matrix to obtain the target matching cost matrix of the project and the expert representing the expert matching result corresponding to each project in the target project cluster. In the embodiment of the present application, by based on the predefined domain keyword dictionary, combining the project text information and the expert information, extract the first correlation vector between the project and the domain and the second correlation vector between the expert and the domain in the multi-dimensional domain, and generate the initial matching cost matrix of the project and the expert based on the first correlation vector and the second correlation vector to meet the specific matching needs of each project in the target project cluster, improve the accuracy of expert matching. At the same time, further optimize the initial matching cost matrix to obtain the target matching cost matrix of the project and the expert, and improve the expert matching coverage rate of the target project cluster.

[0083] Optionally, a specific implementation manner of step S102 for determining the first correlation vector between the project and the domain and the second correlation vector between the expert and the domain based on the predefined domain keyword dictionary, according to the project text information and the expert information, may be: based on the predefined domain keyword dictionary, according to the project text information, generate the first domain correlation matrix corresponding to the target project cluster, and according to the first domain correlation matrix, determine the first correlation vector; based on the predefined domain keyword dictionary, according to the expert information, generate the second domain correlation matrix corresponding to the expert, and according to the second domain correlation matrix, determine the second correlation vector.

[0084] Exemplarily, the matrix structure of the first domain correlation matrix can be a two-dimensional array. Among them, the rows of the matrix represent different domains, and the columns in the matrix represent different items. Each element in the matrix is used to represent the correlation weight coefficient between an item and a domain.

[0085] Exemplarily, the correlation weight coefficient can be a value between 0 and 1.

[0086] Exemplarily, each column in the first domain correlation matrix is determined as the correlation vector of the item represented by this column between the item and the domain, that is, the first correlation vector.

[0087] Exemplarily, each element in the first domain correlation matrix X can be expressed as:

[0088]

[0089] Among them, i represents the domain, that is, there are m different domains in total, and j represents the item, that is, there are n items in total.

[0090] Exemplarily, define the correlation weight coefficient of the j-th column in the first domain correlation matrix to represent the domain correlation vector of item j. The domain correlation vector of item j can be expressed as .

[0091] In a possible implementation, the predefined domain keyword dictionary and item text information are input into the Term Frequency Inverse Document Frequency (TF-IDF) algorithm to automatically determine the correlation between the item and each domain, obtain the first domain correlation matrix output by the TF-IDF algorithm, and determine the correlation weight coefficient corresponding to each column item in the first domain correlation matrix as the first correlation vector between the item and the domain.

[0092] Exemplarily, the matrix structure of the second domain correlation matrix can be a two-dimensional array. Among them, the rows of the matrix represent different domains, and the columns in the matrix represent different experts. Each element in the matrix is used to represent the correlation weight coefficient between an expert and a domain.

[0093] Exemplarily, the correlation weight coefficient can be a value between 0 and 1.

[0094] Exemplarily, each column in the second domain correlation matrix is determined as the correlation vector of the expert represented by this column between the expert and the domain, that is, the second correlation vector.

[0095] Exemplarily, each element in the second domain correlation matrix Y can be expressed as:

[0096]

[0097] Among them, \(i\) represents the field, that is, there are \(m\) different fields in total, and \(k\) represents the item, that is, there are \(l\) experts in total.

[0098] Exemplarily, the correlation weight coefficient of the \(k\)-th column in the second field correlation matrix is defined to represent the field correlation vector of expert \(k\), and the field correlation vector of expert \(k\) can be expressed as .

[0099] In a possible implementation, the predefined field keyword dictionary and expert information are input into the TF-IDF algorithm to automatically determine the correlation between experts and each field, obtain the second field correlation matrix output by the TF-IDF algorithm, and determine the correlation weight coefficient corresponding to each column of experts in the second field correlation matrix as the second correlation vector between the expert and the field.

[0100] Next, in conjunction with Figure 2 the specific implementation of generating the initial matching cost matrix of items and experts according to the first correlation vector and the second correlation vector in step S103 will be described in detail.

[0101] Figure 2 It is a schematic flow of the expert matching method for the project cluster provided by the embodiment of the present application Figure 2 . As Figure 2 shown, the specific implementation of generating the initial matching cost matrix of items and experts according to the first correlation vector and the second correlation vector may include the following steps:

[0102] S201, for each item in the target project cluster and each expert in the expert set, determine the similarity between the item and the expert according to the first correlation vector corresponding to the item and the second correlation vector corresponding to the expert.

[0103] Exemplarily, the similarity between the item and the expert can be the cosine similarity, which is used to represent the matching degree between the item and the expert.

[0104] Exemplarily, the cosine similarity between the item and the expert can be expressed by the following formula:

[0105]

[0106] Among them, represents the cosine similarity.

[0107] It can be understood that when the similarity between the item and the expert is greater than the preset threshold, it indicates that the matching degree between the item and the expert is relatively high.

[0108] Exemplarily, the size of the preset threshold can be 0.8.

[0109] It should be noted that the present application embodiment does not limit the size of the preset threshold for defining the similarity, and it can be determined specifically according to the actual application requirements.

[0110] S202. Determine the matching cost between the project and the expert according to the similarity.

[0111] Exemplarily, the matching cost between the project and the expert can be expressed by the following formula:

[0112]

[0113] Among them, represents the matching cost between the project and the expert.

[0114] It can be understood that when the matching cost between the project and the expert is less than the preset threshold, it indicates that the matching degree between the project and the expert is relatively high.

[0115] Exemplarily, the size of the preset threshold can be 0.2.

[0116] It should be noted that the present application embodiment does not limit the size of the preset threshold for defining the matching cost, and it can be determined specifically according to the actual application requirements.

[0117] S203. Generate an initial matching cost matrix according to the matching cost.

[0118] In a possible implementation manner, traverse all the calculated matching costs between the project and the expert, and fill each element in the matrix to generate the initial matching cost matrix between the project and the expert.

[0119] In the embodiment of the present application, for each project in the target project cluster and each expert in the expert set, by determining the similarity between the project and the expert according to the first correlation vector corresponding to the project and the second correlation vector corresponding to the expert, and determining the matching cost between the project and the expert according to the similarity, and further generating an initial matching cost matrix according to the matching cost, so as to meet the specific matching requirements of each project in the target project cluster and improve the accuracy of expert matching.

[0120] Optionally, in the expert matching method for the project cluster provided by the embodiment of the present application, after optimizing the initial matching cost matrix to obtain the target matching cost matrix between the project and the expert, it further includes calculating the expert matching coverage rate between the project and the expert.

[0121] Next, in combination with Figure 3 A detailed description will be given to the specific implementation manner of calculating the expert matching coverage rate between the project and the expert provided by the embodiment of the present application.

[0122] Figure 3Schematic flow of the expert matching method for the project cluster provided by the embodiment of the present application Figure 3 As Figure 3 shown, after optimizing the initial matching cost matrix to obtain the target matching cost matrix of projects and experts, the expert matching method further includes the following steps:

[0123] S301. Determine the first project quantity corresponding to the covered projects in the target project cluster according to the target matching cost matrix.

[0124] Exemplarily, the covered project means a project that matches at least one expert with a suitable height.

[0125] Exemplarily, in the target matching cost matrix, if the minimum matching cost of one project is less than the preset threshold, it is determined that the project has matched at least one expert with a suitable height, and the project is determined as a covered project.

[0126] Optionally, a possible implementation manner may be: for each project in the target project cluster, determine the minimum matching cost corresponding to the project from the target matching cost matrix; determine whether the minimum matching cost is less than the preset threshold; if less, mark the project as covered; determine the quantity corresponding to all the projects marked as covered as the first project quantity.

[0127] Exemplarily, a possible implementation manner of determining the minimum matching cost corresponding to the project from the target matching cost matrix may be: for each project in the target project cluster, traverse each row in the target matching cost matrix, and determine the minimum value in the row as the minimum matching cost.

[0128] Exemplarily, the preset threshold may be 0.2.

[0129] It should be noted that the embodiment of the present application does not limit the preset threshold for limiting the minimum matching cost, and it can be determined specifically according to actual application requirements.

[0130] Optionally, if the minimum matching cost is greater than or equal to the preset threshold, mark the project as uncovered.

[0131] It can be understood that if a project is uncovered, it means that the project has not matched a review expert with a suitable height.

[0132] S302. Calculate the percentage of the first project quantity in the second project quantity, and determine the percentage as the expert matching coverage rate, where the second project quantity is the total quantity of corresponding projects in the target project cluster.

[0133] Exemplarily, the percentage of the first project quantity in the second project quantity can be expressed by the following formula:

[0134]

[0135] Among them, F represents the expert matching coverage rate, n represents the number of the second items, represents the number of the first items.

[0136] It can be understood that when the required number of experts is less than the number of items included in the target project cluster, there may be a situation where some items cannot be matched with suitable review experts. Such a situation should be avoided as much as possible. Therefore, the higher the expert matching coverage rate, the more items in the target project cluster that can be matched with suitable review experts.

[0137] In the embodiment of the present application, by determining the number of the first items corresponding to the covered items in the target project cluster according to the target matching cost matrix, calculating the percentage of the number of the first items in the number of the second items, and determining the percentage as the expert matching coverage rate, the effectiveness of the expert matching of the target project cluster can be intuitively reflected, which helps to evaluate the matching efficiency of the project.

[0138] In summary, the expert matching method for the project cluster provided by the embodiment of the present application focuses on optimizing the matching effect of the review expert group in the context of multiple project clusters, ensuring efficient solution within a limited time while avoiding falling into a local optimal solution. This method meets the matching requirements of each independent project in the project cluster for experts, and at the same time, improves the global expert matching coverage rate of the project cluster.

[0139] The following is the device embodiment of the present application, which can be used to execute the method embodiment of the present application. For the details not disclosed in the device embodiment of the present application, please refer to the method embodiment of the present application.

[0140] Figure 4 It is a schematic structural diagram of the expert matching device for the project cluster provided by the embodiment of the present application. As Figure 4 shown, the expert matching device 40 for the project cluster includes: an acquisition module 410, a determination module 420, a generation module 430, and a processing module 440.

[0141] Among them, the acquisition module 410 is configured to acquire the project text information of the target project cluster and the expert information corresponding to the expert set;

[0142] The determination module 420 is configured to determine the first correlation vector between the project and the field and the second correlation vector between the expert and the field based on a predefined domain keyword dictionary according to the project text information and the expert information;

[0143] The generation module 430 is configured to generate an initial matching cost matrix between the project and the expert according to the first correlation vector and the second correlation vector;

[0144] A processing module 440 is configured to optimize the initial matching cost matrix to obtain a target matching cost matrix of projects and experts, where the target matching cost matrix is used to represent the expert matching results corresponding to each project in the target project cluster.

[0145] In a possible implementation manner, the determining module 420 is specifically configured to: based on a predefined domain keyword dictionary, generate a first domain correlation matrix corresponding to the target project cluster according to the project text information, and determine a first correlation vector according to the first domain correlation matrix; based on the predefined domain keyword dictionary, generate a second domain correlation matrix corresponding to the expert according to the expert information, and determine a second correlation vector according to the second domain correlation matrix.

[0146] In a possible implementation manner, the generating module 430 is specifically configured to: for each project in the target project cluster and each expert in the expert set, determine the similarity between the project and the expert according to the first correlation vector corresponding to the project and the second correlation vector corresponding to the expert; determine the matching cost between the project and the expert according to the similarity; generate an initial matching cost matrix according to the matching cost.

[0147] In a possible implementation manner, the expert matching device for the project cluster further includes a calculation module (not shown). After optimizing the initial matching cost matrix to obtain a target matching cost matrix of projects and experts, the calculation module is configured to: according to the target matching cost matrix, determine a first project quantity corresponding to the covered projects in the target project cluster; calculate the percentage of the first project quantity in the second project quantity, and determine the percentage as the expert matching coverage rate, where the second project quantity is the total project quantity corresponding to the target project cluster.

[0148] In a possible implementation manner, the expert matching device for the project cluster further includes a marking module (not shown). The marking module is configured to: for each project in the target project cluster, determine the minimum matching cost corresponding to the project from the target matching cost matrix; determine whether the minimum matching cost is less than a preset threshold; if less, mark the project as covered; determine the quantity of all projects marked as covered as the first project quantity.

[0149] In a possible implementation manner, the marking module is further configured to: when the minimum matching cost is greater than or equal to the preset threshold, mark the project as uncovered.

[0150] The expert matching device for the project cluster provided in this embodiment can execute the method provided in the above method embodiment, and its implementation principle and technical effects are similar, which will not be elaborated here in this embodiment.

[0151] Figure 5 It is a schematic structural diagram of an electronic device provided in an embodiment of the present application. AsFigure 5 As shown in Figure 5 , the electronic device 50 provided in this embodiment includes at least one processor 501 and a memory 502. Optionally, the device 50 further includes a communication component 503. Among them, the processor 501, the memory 502, and the communication component 503 are connected through a bus 504.

[0152] In the specific implementation process, at least one processor 501 executes the computer-executable instructions stored in the memory 502, so that at least one processor 501 executes the above-mentioned method.

[0153] For the specific implementation process of the processor 501, reference can be made to the above method embodiment. Its implementation principle and technical effect are similar, and will not be elaborated here in this embodiment.

[0154] In the above embodiment, it should be understood that the processor may be a central processing unit (Central Processing Unit, abbreviated as CPU), or other general-purpose processors, digital signal processors (Digital Signal Processor, abbreviated as DSP), application specific integrated circuits (Application Specific Integrated Circuit, abbreviated as ASIC), etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The steps of the method disclosed in combination with the invention can be directly embodied as being executed by a hardware processor, or executed by a combination of hardware and software modules in the processor.

[0155] The memory may include a high-speed memory (Random Access Memory, abbreviated as RAM), and may also include a non-volatile memory (Non-volatile Memory, abbreviated as NVM), such as at least one disk memory.

[0156] The bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the convenience of representation, the bus in the drawings of this application is not limited to only one bus or one type of bus.

[0157] This embodiment of the present application also provides a computer program product, including a computer program, which implements the above-mentioned method when executed by a processor.

[0158] The embodiments of the present application also provide a computer-readable storage medium, in which computer-executable instructions are stored. When the processor executes the computer-executable instructions, the above-mentioned method is implemented.

[0159] The above-readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, a magnetic disk or an optical disk. The readable storage medium can be any available medium accessible by a general-purpose or special-purpose computer.

[0160] An exemplary readable storage medium is coupled to the processor, enabling the processor to read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can be located in an application-specific integrated circuit (ASIC). Of course, the processor and the readable storage medium can also exist as discrete components in a device.

[0161] The division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection between each other can be through some interfaces, and the indirect coupling or communication connection of devices or units can be in an electrical, mechanical or other forms.

[0162] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0163] In addition, in each embodiment of the present invention, the functional units can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit.

[0164] If a function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in various embodiments of the present invention. The aforementioned storage medium includes: USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs, etc., all kinds of media that can store program codes.

[0165] Those of ordinary skill in the art can understand that all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When this program is executed, it executes the steps including the above method embodiments; and the aforementioned storage medium includes: ROMs, RAMs, magnetic disks, or optical discs, etc., all kinds of media that can store program codes.

[0166] Finally, it should be noted that: After considering the specification and practicing the invention disclosed herein, those skilled in the art will easily think of other implementation manners of the present invention. The present invention is intended to cover any variations, uses, or adaptive changes of the present invention. These variations, uses, or adaptive changes follow the general principles of the present invention and include the common general knowledge or conventional technical means in the technical field not disclosed in the present invention. It is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present invention is only limited by the appended claims.

Claims

1. An expert matching method for project clusters, characterized in that: include: Obtain project text information of the target project cluster and expert information corresponding to the expert set; Based on a predefined domain keyword dictionary, determining a first correlation vector between the project and the domain and a second correlation vector between the expert and the domain according to the project text information and the expert information; generating an initial matching cost matrix between projects and experts according to the first correlation vector and the second correlation vector; The initial matching cost matrix is ​​optimized to obtain a target matching cost matrix between projects and experts, wherein the target matching cost matrix is ​​used to represent the expert matching result corresponding to each project in the target project cluster.

2. The expert matching method according to claim 1, characterized in that: The determining, based on the predefined domain keyword dictionary and according to the project text information and the expert information, a first correlation vector between the project and the domain and a second correlation vector between the expert and the domain comprises: Based on the predefined domain keyword dictionary and according to the project text information, generating a first domain relevance matrix corresponding to the target project cluster, and determining the first relevance vector according to the first domain relevance matrix; Based on the predefined domain keyword dictionary and according to the expert information, a second domain relevance matrix corresponding to the expert is generated, and according to the second domain relevance matrix, the second relevance vector is determined.

3. The expert matching method according to claim 1, characterized in that: The step of generating an initial matching cost matrix between projects and experts according to the first correlation vector and the second correlation vector comprises: For each project in the target project cluster and each expert in the expert set, determine the similarity between the project and the expert according to a first correlation vector corresponding to the project and a second correlation vector corresponding to the expert; Determining a matching cost between the project and the expert according to the similarity; The initial matching cost matrix is ​​generated according to the matching cost.

4. The expert matching method according to any one of claims 1 to 3, characterized in that: After the initial matching cost matrix is ​​optimized to obtain the target matching cost matrix between the project and the expert, the method further includes: Determining, according to the target matching cost matrix, a first number of projects corresponding to covered projects in the target project cluster; The percentage of the first number of items in the second number of items is calculated, and the percentage is determined as the expert matching coverage rate, and the second number of items is the total number of items corresponding to the target item cluster.

5. The expert matching method according to claim 4, characterized in that: The determining, according to the target matching cost matrix, a first number of projects corresponding to the covered projects in the target project cluster includes: For each item in the target item cluster, determining a minimum matching cost corresponding to the item from the target matching cost matrix; Determining whether the minimum matching cost is less than a preset threshold; If it is less, mark the item as covered; The quantity corresponding to all items marked as covered is determined as the first item quantity.

6. The expert matching method according to claim 5, characterized in that: Also includes: If the minimum matching cost is greater than or equal to the preset threshold, the item is marked as uncovered.

7. An expert matching device for project clusters, characterized in that: include: An acquisition module is used to acquire project text information of a target project cluster and expert information corresponding to an expert set; A determination module, configured to determine a first correlation vector between a project and a domain and a second correlation vector between an expert and a domain based on a predefined domain keyword dictionary and according to the project text information and the expert information; A generating module, configured to generate an initial matching cost matrix between projects and experts according to the first correlation vector and the second correlation vector; The processing module is used to optimize the initial matching cost matrix to obtain a target matching cost matrix between projects and experts, wherein the target matching cost matrix is ​​used to represent the expert matching result corresponding to each project in the target project cluster.

8. An electronic device, characterized in that: include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 6 when executed by a processor.

10. A computer program product, characterized in that The invention comprises a computer program, which implements the method according to any one of claims 1 to 6 when being executed by a processor.

Images