A multidisciplinary team of talented people discovers methods, devices and computer equipment

Abstract

The present application relates to a method, device and computer equipment for discovering a multi-disciplinary talent team. The method includes: obtaining a talent association matrix and a research direction matrix; calculating the association expectation between talents according to the talent affiliation indicator matrix and the team relationship matrix, and constructing a first objective function according to the talent association matrix and the association expectation; calculating the degree of association between talents and teams in research directions according to the research direction matrix and the research direction indicator matrix, and constructing a second objective function according to the talent affiliation indicator matrix and the degree of association; calculating the degree of similarity between research directions between teams according to the research direction indicator matrix, and constructing a third objective function according to the team relationship matrix and the degree of similarity; constructing and solving the overall goal according to the above objective function to obtain the optimal talent affiliation indicator matrix; and planning the talent team according to the optimal talent affiliation indicator matrix and the overlapping division strategy. The use of this method can improve the efficiency of cooperation between teams.

Description

Technical Field

[0001] The present application relates to the field of talent matching technology, and in particular to a method, device and computer equipment for discovering a multi-disciplinary talent team. Background Art

[0002] With the high degree of integration and rapid development of various fields in modern society, traditional single-discipline professionals can no longer meet the diverse needs in complex task environments. Especially in collaborative work and innovation tasks in multiple fields, the knowledge system and capabilities of a single field can no longer adapt to the increasingly complex and dynamic challenges. Therefore, how to form an interdisciplinary and multi-field collaborative team has become the key to solving this problem.

[0003] At present, most existing technologies are based on talent management and team building methods in specific disciplines, and often rely on professional knowledge in a certain field to select and organize team members. However, this method has some obvious shortcomings. First, the traditional team building method fails to effectively consider the interdisciplinary capabilities of members and the interactive relationship between teams, resulting in the failure to fully tap the knowledge and skills synergy potential between team members. Secondly, existing talent discovery and recommendation methods usually rely on a single source of information, such as the talent's disciplinary background or work experience, and lack the comprehensive use of multiple information sources, which can easily cause incomplete or biased information, thereby affecting the overall team building effect.

[0004] In addition, in the process of building an interdisciplinary team, how to accurately identify and assign roles to talents, how to quantify the synergy between teams, and how to ensure coordination and consistency among team members are still important challenges facing current technology. Due to the lack of effective technical means to solve these problems, existing methods often find it difficult to provide flexible, efficient, and accurate solutions for building interdisciplinary talent teams in practical applications.

[0005] Therefore, how to use more advanced data analysis and model optimization technologies to address the shortcomings of existing team building methods and improve the overall collaborative efficiency and innovation capabilities of interdisciplinary teams has become a technical problem that needs to be solved urgently. Summary of the invention

[0006] Based on this, it is necessary to provide a method, device and computer equipment for discovering a multi-disciplinary talent team to address the above-mentioned technical issues.

[0007] A method for discovering a multi-disciplinary talent team, the method comprising:

[0008] Obtaining a talent association matrix, a research direction matrix, a talent affiliation indication matrix, a team relationship matrix and a research direction indication matrix; the talent association matrix includes the academic association relationships between talents; the research direction matrix includes the research directions of talents;

[0009] Calculating the expected correlation between talents according to the talent affiliation indicator matrix and the team relationship matrix, and constructing a first objective function according to the consistency between the talent correlation matrix and the expected correlation between talents;

[0010] Calculating the degree of correlation between the talent and the talent team in terms of research direction according to the research direction matrix and the research direction indication matrix, and constructing a second objective function according to the consistency between the talent affiliation indication matrix and the degree of correlation between the talent and the talent team in terms of research direction;

[0011] The similarity of the research directions between the teams is calculated according to the research direction indicator matrix, and the third objective function is constructed according to the consistency between the team relationship matrix and the similarity of the research directions between the talent teams;

[0012] The overall goal is constructed according to the first objective function, the second objective function and the third objective function, and the update rule is obtained by performing three-factor non-negative matrix decomposition on the talent association matrix and the research direction matrix in the overall goal, and the overall goal is iteratively solved to obtain the optimal talent affiliation indicator matrix and the optimal team relationship matrix; the optimal team relationship matrix is ​​used to quantify the closeness between talent teams;

[0013] Multi-disciplinary talent team planning is carried out according to the optimal talent affiliation indication matrix and the pre-set overlapping division strategy.

[0014] A multi-disciplinary talent team discovery device, comprising:

[0015] A parameter acquisition module is used to acquire a talent association matrix, a research direction matrix, a talent affiliation indication matrix, a team relationship matrix and a research direction indication matrix; the talent association matrix includes the academic association relationships between talents; the research direction matrix includes the research directions of talents;

[0016] A first objective function construction module is used to calculate the expected association between talents according to the talent affiliation indicator matrix and the team relationship matrix, and to construct a first objective function according to the consistency between the talent association matrix and the expected association between the talents;

[0017] A second objective function construction module is used to calculate the correlation between the talent and the talent team in the research direction according to the research direction matrix and the research direction indication matrix, and to construct the second objective function according to the consistency between the talent affiliation indication matrix and the correlation between the talent and the talent team in the research direction;

[0018] A third objective function construction module is used to calculate the similarity of the research directions between teams according to the research direction indication matrix, and to construct a third objective function according to the consistency between the team relationship matrix and the similarity of the research directions between the talent teams;

[0019] A solution module is used to construct an overall goal according to the first objective function, the second objective function and the third objective function, obtain an update rule by performing three-factor non-negative matrix decomposition on the talent association matrix and the research direction matrix in the overall goal, and iteratively solve the overall goal to obtain an optimal talent affiliation indicator matrix and an optimal team relationship matrix; the optimal team relationship matrix is ​​used to quantify the closeness between talent teams;

[0020] The talent team planning module is used to carry out multi-disciplinary and cross-disciplinary talent team planning based on the optimal talent affiliation indication matrix and the pre-set overlapping division strategy.

[0021] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0022] Obtaining a talent association matrix, a research direction matrix, a talent affiliation indication matrix, a team relationship matrix and a research direction indication matrix; the talent association matrix includes the academic association relationships between talents; the research direction matrix includes the research directions of talents;

[0023] Calculating the expected correlation between talents according to the talent affiliation indicator matrix and the team relationship matrix, and constructing a first objective function according to the consistency between the talent correlation matrix and the expected correlation between talents;

[0024] Calculating the degree of correlation between the talent and the talent team in terms of research direction according to the research direction matrix and the research direction indication matrix, and constructing a second objective function according to the consistency between the talent affiliation indication matrix and the degree of correlation between the talent and the talent team in terms of research direction;

[0025] The similarity of the research directions between the teams is calculated according to the research direction indicator matrix, and the third objective function is constructed according to the consistency between the team relationship matrix and the similarity of the research directions between the talent teams;

[0026] The overall goal is constructed according to the first objective function, the second objective function and the third objective function, and the update rule is obtained by performing three-factor non-negative matrix decomposition on the talent association matrix and the research direction matrix in the overall goal, and the overall goal is iteratively solved to obtain the optimal talent affiliation indicator matrix and the optimal team relationship matrix; the optimal team relationship matrix is ​​used to quantify the closeness between talent teams;

[0027] Multi-disciplinary talent team planning is carried out according to the optimal talent affiliation indication matrix and the pre-set overlapping division strategy.

[0028] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the following steps:

[0029] Obtaining a talent association matrix, a research direction matrix, a talent affiliation indication matrix, a team relationship matrix and a research direction indication matrix; the talent association matrix includes the academic association relationships between talents; the research direction matrix includes the research directions of talents;

[0030] Calculating the expected correlation between talents according to the talent affiliation indicator matrix and the team relationship matrix, and constructing a first objective function according to the consistency between the talent correlation matrix and the expected correlation between talents;

[0031] Calculating the degree of correlation between the talent and the talent team in terms of research direction according to the research direction matrix and the research direction indication matrix, and constructing a second objective function according to the consistency between the talent affiliation indication matrix and the degree of correlation between the talent and the talent team in terms of research direction;

[0032] The similarity of the research directions between the teams is calculated according to the research direction indicator matrix, and the third objective function is constructed according to the consistency between the team relationship matrix and the similarity of the research directions between the talent teams;

[0033] The overall goal is constructed according to the first objective function, the second objective function and the third objective function, and the update rule is obtained by performing three-factor non-negative matrix decomposition on the talent association matrix and the research direction matrix in the overall goal, and the overall goal is iteratively solved to obtain the optimal talent affiliation indicator matrix and the optimal team relationship matrix; the optimal team relationship matrix is ​​used to quantify the closeness between talent teams;

[0034] Multi-disciplinary talent team planning is carried out according to the optimal talent affiliation indication matrix and the pre-set overlapping division strategy.

[0035] The above-mentioned multi-disciplinary talent team discovery method, device and computer equipment can effectively make up for the information shortage problem caused by a single information source by obtaining the talent association matrix and the research direction matrix, introduce the talent affiliation indicator matrix and the team relationship matrix, and can quantify the probability that the talent belongs to the team and the degree of association between the teams, construct the first objective function, make the association expectation between talents in the same team as consistent as possible with the actual talent association matrix, thereby reflecting the internal collaborative relationship of the team, introduce the research direction indicator matrix, and ensure the high matching degree between the talent research direction and the team research direction by constructing the second objective function, while unifying the team structure implied by the talent association information and the research direction information, enhancing the rationality of planning, constructing the third objective function, making the similarity between teams as consistent as possible with the team relationship matrix, optimizing the collaborative design between teams, and finally, using the three-factor non-negative matrix decomposition technology, jointly optimizing the objective function, so that the affiliation indicator matrix and the team relationship matrix can accurately extract the potential team structure and quantify the closeness between teams, and through the overlapping division strategy, enable talents to belong to multiple teams at the same time, meeting the multi-disciplinary cross-cutting needs. The embodiment of the present invention can improve the efficiency of inter-team collaboration, innovation ability and adaptability, and provide strong support for the optimization and resource allocation of scientific research teams. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 A schematic diagram of a process for discovering a multi-disciplinary talent team in one embodiment;

[0037] Figure 2 A structural block diagram of a device for discovering a multi-disciplinary talent team in one embodiment;

[0038] Figure 3 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0039] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0040] In one embodiment, Figure 1 As shown, a method for discovering a multi-disciplinary talent team is provided, including the following steps:

[0041] Step 102, obtaining a talent association matrix, a research direction matrix, a talent affiliation indication matrix, a team relationship matrix and a research direction indication matrix.

[0042] First, construct the talent association matrix ,in is the number of talents, if talents and If there is an academic relationship between ,otherwise Similarly, construct the research direction matrix , is the number of research directions. The research direction is ,but ,otherwise .

[0043] For the discovery of multi-disciplinary talent teams, the closeness of the association between different teams has important practical significance. The present invention uses three-factor non-negative matrix decomposition to not only obtain accurate talent team discovery results, but also obtain the closeness of the relationship between talent teams. , Representing talents Belong to the team The probability of. At the same time, the team relationship matrix is ​​introduced , Representation Team and The degree of relationship between the two. Introducing the research direction indicator matrix , Representation Team Have research direction probability.

[0044] Step 104, calculating the expected associations between talents based on the talent affiliation indicator matrix and the team relationship matrix, and constructing a first objective function based on the consistency between the talent association matrix and the expected associations between talents.

[0045] Talent and The expectation of the relationship between The probability that two people in the same team are related is higher, and the expectation that talents are related should be consistent with the talent connection matrix Be as consistent as possible.

[0046] Based on the introduced talent affiliation indicator matrix and team relationship matrix, the optimization objective function can be constructed as follows:

[0047] (1)

[0048] in, and They represent the elements in the talent affiliation indicator matrix and the team relationship matrix respectively.

[0049] Step 106, calculating the correlation between the talents and the talent team in the research direction according to the research direction matrix and the research direction indication matrix, and constructing the second objective function according to the consistency between the talent affiliation indication matrix and the correlation between the talents and the talent team in the research direction.

[0050] If the individual's research direction is similar to that of the team, then the probability that the talent belongs to the team is relatively high. Since the team information implied by the talent association relationship should be consistent with that implied by the research direction, the following objective function is constructed:

[0051] (2)

[0052] in, and They represent the elements in the talent affiliation indication matrix and the research direction indication matrix respectively.

[0053] Step 108, calculating the similarity of the research directions between teams according to the research direction indication matrix, and constructing a third objective function according to the consistency between the team relationship matrix and the similarity of the research directions between the talent teams.

[0054] Not only are the team structures implied by the talent association information and the research direction information consistent, but the implicit relationships between teams are also consistent. and Similarities between Team Relationship Matrix It should be as consistent as possible, so the following regularization term is constructed:

[0055] (3)

[0056] Step 110, constructing the overall goal according to the first objective function, the second objective function and the third objective function, obtaining the update rules by performing three-factor non-negative matrix decomposition on the talent association matrix and the research direction matrix in the overall goal, and iteratively solving the overall goal to obtain the optimal talent affiliation indicator matrix and the optimal team relationship matrix.

[0057] The optimal team relationship matrix is ​​used to quantify the closeness between talent teams.

[0058] Based on the above, the final objective function is as follows:

[0059] (4)

[0060] in, and is the weight parameter of the regularization direction.

[0061] The following update rules can be obtained:

[0062] (5)

[0063] (6)

[0064] (7)

[0065] Through iterative optimization solution, we can get the talent affiliation indicator matrix Relationship matrix with talent team .

[0066] Step 112, planning of a multi-disciplinary talent team is performed based on the optimal talent affiliation indicator matrix and a pre-set overlapping division strategy.

[0067] For multi-dimensional cross-team division, a talent can belong to multiple teams. The present invention proposes an overlapping division strategy to find the label of the team to which the talent belongs. First, a soft division threshold is set. If the talent's affiliation is greater than the threshold, the node is divided into the team. The soft division principle is as follows:

[0068] ;

[0069] Diverse cross-talent team discovery is achieved by finding team labels corresponding to affiliations greater than an overlapping partition threshold.

[0070] In the above multi-disciplinary talent team discovery method, by obtaining the talent association matrix and the research direction matrix, the problem of insufficient information caused by a single information source can be effectively compensated. The introduction of the talent affiliation indicator matrix and the team relationship matrix can quantify the probability that the talent belongs to the team and the degree of association between the teams. The first objective function is constructed to make the association expectation between talents in the same team as consistent as possible with the actual talent association matrix, thereby reflecting the internal collaborative relationship of the team. The research direction indicator matrix is ​​introduced, and by constructing the second objective function, the high matching degree between the talent research direction and the team research direction is ensured. At the same time, the team structure implied by the talent association information and the research direction information is unified to enhance the rationality of planning. The third objective function is constructed to make the similarity between teams as consistent as possible with the team relationship matrix, and optimize the collaborative design between teams. Finally, the three-factor non-negative matrix decomposition technology is used to jointly optimize the objective function, so that the affiliation indicator matrix and the team relationship matrix can accurately extract the potential team structure and quantify the closeness between teams. Through the overlapping division strategy, talents can belong to multiple teams at the same time to meet the needs of multi-disciplinary cross-border. The embodiment of the present invention can improve the efficiency of inter-team collaboration, innovation ability and adaptability, and provide strong support for the optimization and resource allocation of scientific research teams.

[0071] In one embodiment, planning a multi-disciplinary talent team based on the optimal talent affiliation indicator matrix and a preset overlapping division strategy includes: traversing each talent, obtaining the current talent according to the optimal talent affiliation indicator matrix Talent team If the membership degree satisfies the soft division principle, the talent Divide into talent team , and the team Tagged as Talent Team labels are used to plan multi-disciplinary talent teams based on the team labels of each talent.

[0072] In one embodiment, the soft partitioning principle is:

[0073] ;

[0074] in, For talents Belong to the talent team , For talents Talent team The membership degree of For talents The average membership degree of For talents The maximum membership degree of is the soft segmentation threshold.

[0075] In one embodiment, constructing the first objective function according to the consistency between the talent association matrix and the expected association between talents includes: constructing the first objective function according to the consistency between the talent association matrix and the expected association between talents as follows:

[0076] ;

[0077] in, is the talent association matrix, is the talent affiliation indicator matrix, For the team relationship matrix, is the square of the F norm of the matrix. The elements in the talent association matrix and the talent affiliation indicator matrix are all greater than or equal to 0. For the number of talents, The number of talent teams, is the matrix transpose operation.

[0078] In one embodiment, constructing the second objective function according to the consistency of the talent affiliation indication matrix and the degree of association between the talent and the talent team in the research direction includes: constructing the second objective function according to the consistency of the talent affiliation indication matrix and the degree of association between the talent and the talent team in the research direction as follows:

[0079] ;

[0080] in, is the research direction matrix, is the research direction indicator matrix, is the talent affiliation indicator matrix, is the square of the F norm of the matrix. The elements in the research direction indicator matrix and the talent affiliation indicator matrix are both greater than or equal to 0. For the number of talents, The number of talent teams, is the number of research directions, is the matrix transpose operation.

[0081] In one embodiment, constructing the third objective function according to the consistency of the team relationship matrix and the similarity of the research directions between the teams includes: constructing the third objective function according to the consistency of the team relationship matrix and the similarity of the research directions between the teams as follows:

[0082] ;

[0083] in, is the research direction indicator matrix, For the team relationship matrix, is the square of the matrix F norm, is the matrix transpose operation.

[0084] In one embodiment, constructing the overall goal according to the first objective function, the second objective function, and the third objective function includes: constructing the overall goal according to the first objective function, the second objective function, and the third objective function as follows:

[0085] ;

[0086] in, is the talent association matrix, is the talent affiliation indicator matrix, is the research direction indicator matrix, is the research direction matrix, For the team relationship matrix, is the square of the matrix F norm, and is the weight parameter of the regularization direction, is the transposition operation of the matrix. The elements in the team relationship matrix, research direction indicator matrix, and talent affiliation indicator matrix are all greater than or equal to 0. For the number of talents, The number of talent teams, is the number of research directions.

[0087] It should be understood that although Figure 1 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 1 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0088] In one embodiment, Figure 2 As shown, a multi-disciplinary talent team discovery device is provided, comprising:

[0089] Parameter acquisition module 202, used to acquire a talent association matrix, a research direction matrix, a talent affiliation indication matrix, a team relationship matrix and a research direction indication matrix; the talent association matrix includes the academic association relationships between talents; the research direction matrix includes the research directions of talents;

[0090] A first objective function construction module 204 is used to calculate the expected association between talents according to the talent affiliation indicator matrix and the team relationship matrix, and to construct a first objective function according to the consistency between the talent association matrix and the expected association between talents;

[0091] A second objective function construction module 206 is used to calculate the correlation between the talent and the talent team in the research direction according to the research direction matrix and the research direction indication matrix, and to construct a second objective function according to the consistency between the talent affiliation indication matrix and the correlation between the talent and the talent team in the research direction;

[0092] A third objective function construction module 208 is used to calculate the similarity of the research directions between teams according to the research direction indication matrix, and to construct a third objective function according to the consistency between the team relationship matrix and the similarity of the research directions between the talent teams;

[0093] The solution module 210 is used to construct the overall goal according to the first objective function, the second objective function and the third objective function, obtain the update rule by performing three-factor non-negative matrix decomposition on the talent association matrix and the research direction matrix in the overall goal, and iteratively solve the overall goal to obtain the optimal talent affiliation indicator matrix and the optimal team relationship matrix; the optimal team relationship matrix is ​​used to quantify the closeness between talent teams;

[0094] The talent team planning module 212 is used to plan a multi-disciplinary talent team based on the optimal talent affiliation indication matrix and a preset overlapping division strategy.

[0095] For the specific definition of the multi-disciplinary talent team discovery device, please refer to the definition of the multi-disciplinary talent team discovery method above, which will not be repeated here. Each module in the above-mentioned multi-disciplinary talent team discovery device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0096] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 3 As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for discovering a multi-disciplinary talent team is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a button, trackball or touchpad set on the computer device housing, or an external keyboard, touchpad or mouse, etc.

[0097] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0098] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method in the above embodiment when executing the computer program.

[0099] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method in the above embodiment are implemented.

[0100] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0101] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0102] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for discovering a multi-disciplinary talent team, characterized in that: The method comprises: Obtaining a talent association matrix, a research direction matrix, a talent affiliation indication matrix, a team relationship matrix and a research direction indication matrix; the talent association matrix includes the academic association relationships between talents; the research direction matrix includes the research directions of talents; Calculating the expected correlation between talents according to the talent affiliation indicator matrix and the team relationship matrix, and constructing a first objective function according to the consistency between the talent correlation matrix and the expected correlation between talents; Calculating the degree of correlation between the talent and the talent team in terms of research direction according to the research direction matrix and the research direction indication matrix, and constructing a second objective function according to the consistency between the talent affiliation indication matrix and the degree of correlation between the talent and the talent team in terms of research direction; The similarity of the research directions between the teams is calculated according to the research direction indicator matrix, and the third objective function is constructed according to the consistency between the team relationship matrix and the similarity of the research directions between the talent teams; The overall goal is constructed according to the first objective function, the second objective function and the third objective function, and the update rule is obtained by performing three-factor non-negative matrix decomposition on the talent association matrix and the research direction matrix in the overall goal, and the overall goal is iteratively solved to obtain the optimal talent affiliation indicator matrix and the optimal team relationship matrix; the optimal team relationship matrix is ​​used to quantify the closeness between talent teams; Carry out multi-disciplinary and cross-disciplinary talent team planning according to the optimal talent affiliation indicator matrix and the pre-set overlapping division strategy; The planning of a multi-disciplinary talent team based on the optimal talent affiliation indicator matrix and the pre-set overlapping division strategy includes: Traverse each talent and get the current talent according to the optimal talent affiliation indicator matrix Talent team If the membership degree satisfies the soft division principle, the talent Divide into talent team , and the team Tagged as Talent Team labels of talents are used to plan multi-disciplinary talent teams according to the team labels of each talent. The soft partitioning principle is: ； in, For talents Belong to the talent team , For talents Talent team The membership degree of For talents The average membership degree of For talents The maximum membership degree of is the soft segmentation threshold.

2. The method according to claim 1, characterized in that The step of constructing the first objective function according to the consistency between the talent association matrix and the expected association between the talents includes: The first objective function is constructed based on the consistency between the talent association matrix and the expected association between the talents: ； in, is the talent association matrix, is the talent affiliation indicator matrix, For the team relationship matrix, is the square of the F norm of the matrix. The elements in the talent association matrix and the talent affiliation indicator matrix are all greater than or equal to 0. For the number of talents, The number of talent teams, is the matrix transpose operation.

3. The method according to claim 1, characterized in that The second objective function is constructed according to the consistency of the talent affiliation indicator matrix and the degree of association between the talent and the talent team in terms of research direction, including: The second objective function is constructed based on the talent affiliation indicator matrix and the consistency of the degree of association between the talent and the talent team in terms of research direction: ； in, is the research direction matrix, is the research direction indicator matrix, is the talent affiliation indicator matrix, is the square of the F norm of the matrix. The elements in the research direction indicator matrix and the talent affiliation indicator matrix are both greater than or equal to 0. For the number of talents, The number of talent teams, is the number of research directions, is the matrix transpose operation.

4. The method according to claim 1, characterized in that: The third objective function is constructed based on the consistency between the team relationship matrix and the similarity of the research directions between the teams, including: The third objective function is constructed based on the consistency between the team relationship matrix and the similarity of the research directions between the teams: ； in, is the research direction indicator matrix, For the team relationship matrix, is the square of the matrix F norm, is the matrix transpose operation.

5. The method according to claim 1, characterized in that Constructing the overall goal according to the first objective function, the second objective function and the third objective function includes: The overall goal is constructed based on the first objective function, the second objective function and the third objective function: ； in, is the talent association matrix, is the talent affiliation indicator matrix, is the research direction indicator matrix, is the research direction matrix, For the team relationship matrix, is the square of the matrix F norm, and is the weight parameter of the regularization direction, is the transposition operation of the matrix. The elements in the team relationship matrix, research direction indicator matrix, and talent affiliation indicator matrix are all greater than or equal to 0. For the number of talents, The number of talent teams, is the number of research directions.

6. A multi-disciplinary talent team discovery device, characterized in that: The device comprises: A parameter acquisition module is used to acquire a talent association matrix, a research direction matrix, a talent affiliation indication matrix, a team relationship matrix and a research direction indication matrix; the talent association matrix includes the academic association relationships between talents; the research direction matrix includes the research directions of talents; A first objective function construction module is used to calculate the expected association between talents according to the talent affiliation indicator matrix and the team relationship matrix, and to construct a first objective function according to the consistency between the talent association matrix and the expected association between the talents; A second objective function construction module is used to calculate the correlation between the talent and the talent team in the research direction according to the research direction matrix and the research direction indication matrix, and to construct the second objective function according to the consistency between the talent affiliation indication matrix and the correlation between the talent and the talent team in the research direction; A third objective function construction module is used to calculate the similarity of the research directions between teams according to the research direction indication matrix, and to construct a third objective function according to the consistency between the team relationship matrix and the similarity of the research directions between the talent teams; A solution module is used to construct an overall goal according to the first objective function, the second objective function and the third objective function, obtain an update rule by performing three-factor non-negative matrix decomposition on the talent association matrix and the research direction matrix in the overall goal, and iteratively solve the overall goal to obtain an optimal talent affiliation indicator matrix and an optimal team relationship matrix; the optimal team relationship matrix is ​​used to quantify the closeness between talent teams; A talent team planning module, used to plan a multi-disciplinary talent team according to the optimal talent affiliation indication matrix and a preset overlapping division strategy; The talent team planning module is also used to traverse each talent and obtain the current talent according to the optimal talent affiliation indicator matrix. Talent team If the membership degree satisfies the soft division principle, the talent Divide into talent team , and the team Tagged as Talent Team labels of talents are used to plan multi-disciplinary talent teams according to the team labels of each talent. The soft partitioning principle is: ； in, For talents Belong to the talent team , For talents Talent team The membership degree of For talents The average membership degree of For talents The maximum membership degree of is the soft segmentation threshold.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

Images