Dynamic talent team discovery method, device and equipment based on complex network analysis

Abstract

The invention relates to a dynamic talent team discovery method, device and equipment based on complex network analysis. The method comprises the following steps: acquiring a talent incidence matrix and a research direction matrix; performing similarity sorting on each talent and other talents according to the talent incidence matrix and the research direction matrix to obtain a sorting vector, performing information fusion on the sorting vector, performing singular value decomposition, extracting a first column of a left singular vector, obtaining neighbors of each talent based on similarity, and constructing a talent unification graph; performing non-negative matrix factorization on the talent unified graph to obtain a talent membership indication matrix, calculating the similarity degree of the talents on team attribution according to the talent membership indication matrix, minimizing the difference between the similarity degree of the talent unified graph and the similarity degree of the talents on team attribution, applying historical evolution constraints, and constructing a target function; and performing dynamic talent team division according to the optimal talent membership indication matrix obtained by solving the objective function at each moment. By adopting the method, team division can be adjusted in real time.

Description

Technical Field

[0001] The present application relates to the field of talent matching technology, and in particular to a method, device and equipment for discovering a dynamic talent team based on complex network analysis. Background Art

[0002] To accurately and efficiently locate professional talents that meet the needs and quickly form a systematic professional force, that is, the problem of discovering talent teams has always been an important research topic in the field of human resources. The problem of mining talent teams based on similar research directions has always been a hot topic in this field. In actual scenarios, it is often necessary to gather a group of high-tech professionals with the same research directions or knowledge and skills to support specific work tasks. However, key information such as the research direction of talents usually has time evolution characteristics, which requires real-time mining of talent teams with similar research directions in a dynamically changing environment.

[0003] Traditional human resource research mostly uses qualitative analysis, but in a dynamically changing environment, traditional methods cannot effectively support real-time team discovery. Therefore, artificial intelligence technology provides a more objective, comprehensive perspective and intelligent tools for the discovery research of talent teams. In order to better develop talent teams in the information age, it is necessary to use more quantitative and intelligent means to explore talent teams, thereby providing new decision-making support methods for talent development and management. Based on complex network-related theories and methods, breaking the limitations of intuition or personal experience and providing a more scientific basis for decision-making will open up new directions for the discovery research of talent teams. Summary of the invention

[0004] Based on this, it is necessary to provide a dynamic talent team discovery method, device and equipment based on complex network analysis to address the above technical problems.

[0005] A dynamic talent team discovery method based on complex network analysis, the method comprising:

[0006] Build a talent association matrix based on the academic associations between talents, and build a research direction matrix based on the research directions of talents;

[0007] According to the talent association matrix and the research direction matrix, each talent is sorted by similarity with other talents to obtain a sorting vector, multi-dimensional information fusion is performed on the sorting vector to obtain a sorting matrix, singular value decomposition is performed on the sorting matrix, the first column of the left singular vector is extracted, and the first K talents with the highest similarity are selected as neighbors of the current talent, and a talent unified graph is constructed according to each talent and the corresponding neighbor; the nodes of the talent unified graph represent talents, and the edges represent the association relationship between talents;

[0008] Performing non-negative matrix decomposition on the talent unified graph to obtain a corresponding talent affiliation indicator matrix, calculating the similarity of talents in team affiliation according to the talent affiliation indicator matrix, minimizing the difference between the talent unified graph and the similarity of talents in team affiliation, obtaining a talent team division result, applying historical evolution constraints to the talent team division result, and constructing an objective function;

[0009] The objective function at each moment is optimized and solved to obtain the corresponding optimal talent affiliation indicator matrix, and dynamic talent team division is performed based on the optimal talent affiliation indicator matrix.

[0010] A dynamic talent team discovery device based on complex network analysis, the device comprising:

[0011] The parameter acquisition module is used to construct a talent association matrix based on the academic association relationships between talents, and to construct a research direction matrix based on the research directions of talents;

[0012] A graph construction module is used to sort the similarity of each talent with other talents according to the talent association matrix and the research direction matrix to obtain a sorting vector, perform multi-dimensional information fusion on the sorting vector to obtain a sorting matrix, perform singular value decomposition on the sorting matrix, extract the first column of the left singular vector, select the first K talents with the highest similarity as neighbors of the current talent, and construct a unified talent graph according to each talent and the corresponding neighbor; the nodes of the unified talent graph represent talents, and the edges represent the association relationship between talents;

[0013] A model building module is used to perform non-negative matrix decomposition on the talent unified graph to obtain a corresponding talent affiliation indicator matrix, calculate the similarity of talents in team affiliation according to the talent affiliation indicator matrix, minimize the difference between the talent unified graph and the similarity of talents in team affiliation, obtain the talent team division result, impose historical evolution constraints on the talent team division result, and construct an objective function;

[0014] The model solving module is used to optimize and solve the objective function at each moment, obtain the corresponding optimal talent affiliation indicator matrix, and perform dynamic talent team division based on the optimal talent affiliation indicator matrix.

[0015] 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:

[0016] Build a talent association matrix based on the academic associations between talents, and build a research direction matrix based on the research directions of talents;

[0017] According to the talent association matrix and the research direction matrix, each talent is sorted by similarity with other talents to obtain a sorting vector, multi-dimensional information fusion is performed on the sorting vector to obtain a sorting matrix, singular value decomposition is performed on the sorting matrix, the first column of the left singular vector is extracted, and the first K talents with the highest similarity are selected as neighbors of the current talent, and a talent unified graph is constructed according to each talent and the corresponding neighbor; the nodes of the talent unified graph represent talents, and the edges represent the association relationship between talents;

[0018] Performing non-negative matrix decomposition on the talent unified graph to obtain a corresponding talent affiliation indicator matrix, calculating the similarity of talents in team affiliation according to the talent affiliation indicator matrix, minimizing the difference between the talent unified graph and the similarity of talents in team affiliation, obtaining a talent team division result, applying historical evolution constraints to the talent team division result, and constructing an objective function;

[0019] The objective function at each moment is optimized and solved to obtain the corresponding optimal talent affiliation indicator matrix, and dynamic talent team division is performed based on the optimal talent affiliation indicator matrix.

[0020] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the following steps:

[0021] Build a talent association matrix based on the academic associations between talents, and build a research direction matrix based on the research directions of talents;

[0022] According to the talent association matrix and the research direction matrix, each talent is sorted by similarity with other talents to obtain a sorting vector, multi-dimensional information fusion is performed on the sorting vector to obtain a sorting matrix, singular value decomposition is performed on the sorting matrix, the first column of the left singular vector is extracted, and the first K talents with the highest similarity are selected as neighbors of the current talent, and a talent unified graph is constructed according to each talent and the corresponding neighbor; the nodes of the talent unified graph represent talents, and the edges represent the association relationship between talents;

[0023] Performing non-negative matrix decomposition on the talent unified graph to obtain a corresponding talent affiliation indicator matrix, calculating the similarity of talents in team affiliation according to the talent affiliation indicator matrix, minimizing the difference between the talent unified graph and the similarity of talents in team affiliation, obtaining a talent team division result, applying historical evolution constraints to the talent team division result, and constructing an objective function;

[0024] The objective function at each moment is optimized and solved to obtain the corresponding optimal talent affiliation indicator matrix, and dynamic talent team division is performed based on the optimal talent affiliation indicator matrix.

[0025] The above-mentioned dynamic talent team discovery method, device and equipment based on complex network analysis can effectively integrate multi-dimensional data, improve the analysis accuracy of talent relationships and collaboration potential, and perform non-negative matrix decomposition on the talent unified graph to obtain the corresponding talent affiliation indicator matrix. The similarity of talents in team affiliation is calculated according to the talent affiliation indicator matrix, and the difference between the talent unified graph and the talent in team affiliation is minimized to obtain the talent team division result, and the historical evolution constraint is imposed on the talent team division result, and the objective function is constructed to achieve the smooth evolution of team division in a dynamic environment, effectively support the adjustment of team structure at different time points, and ensure that the team can adapt to the dynamic changes in talent demand. The present invention avoids the deviation of subjective judgment through data-driven quantitative analysis, and ensures that the team division is more scientific and accurate. The embodiment of the present invention can adjust the team division in real time, optimize the allocation of talent resources, and improve the efficiency of team collaboration and innovation ability. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 A flowchart of a dynamic talent team discovery method based on complex network analysis in one embodiment;

[0027] Figure 2 A schematic diagram of experimental results of an algorithm in an embodiment on a real network data set Q;

[0028] Figure 3 A structural block diagram of a dynamic talent team discovery device based on complex network analysis in one embodiment;

[0029] Figure 4 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0030] 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.

[0031] In one embodiment, Figure 1 As shown, a dynamic talent team discovery method based on complex network analysis is provided, including the following steps:

[0032] Step 102, constructing a talent association matrix based on the academic association relationships between talents, and constructing a research direction matrix based on the research directions of the talents.

[0033] Construct the talent association matrix A∈R according to the talent pre-positioning database n×n , where n is the number of talents. If there is an academic relationship between talents i and j, then Aij =1, otherwise A ij = 0. At the same time, construct the talent research direction attribute matrix T∈R n×m , where m is the number of research directions. If the research direction of talent i is j, T ij =1, otherwise T ij = 0. If there is an academic connection between two talents, such as academic citations, it is considered that their research directions are similar. The talent pre-positioning database refers to a pre-established database or system containing talent information, which is used to store and manage various information related to talents, including talents' academic background, research direction, skills, work experience, project participation, etc.

[0034] Step 104, sort the similarity between each talent and other talents according to the talent association matrix and the research direction matrix to obtain a sorting vector, perform multi-dimensional information fusion on the sorting vector to obtain a sorting matrix, perform singular value decomposition on the sorting matrix, extract the first column of the left singular vector, select the first K talents with the highest similarity as the neighbors of the current talent, and construct a unified talent graph based on each talent and the corresponding neighbor.

[0035] Dynamic talent team discovery involves multi-dimensional heterogeneous data (such as academic connections and research directions of talents), and these data may change over time. These characteristics make it difficult for simple linear models or traditional data mining techniques to effectively solve the problem. The present invention is based on complex network analysis methods and realizes dynamic talent team discovery by constructing a multi-dimensional network model. In the present invention, complex network analysis refers to modeling and analyzing various relationships between talents through graph theory methods to identify potential collaboration patterns, optimize team structure and achieve dynamic adjustment.

[0036] By integrating the talent association matrix and the research direction matrix, a unified graph model is constructed, whose input includes the talent set {u1,u2,...,u n}, talent relationship view and research direction view, each view represents part or all of the n talents, the nodes of the talent unified graph represent talents, and the edges represent the associations between talents. For data containing w dimensions, the nodes are recorded as U = {u1,u2,...,u n}, the data of each dimension contains all or part of the nodes, and the information fusion steps are as follows:

[0037] Step 1: For each dimension of data j (1≤j≤w), calculate u i Similarity vector p with other nodes in this dimension ij Specifically, for each talent i and for each view j, compute talent u using the similarity measure provided by that view iand the similarity vector between all other talents in the view.

[0038] Step 2: According to p ij The value of generates a sorted vector containing the other n-1 nodes, denoted as r ij If dimension j does not contain all nodes, the non-existent nodes are assigned an order n′ j +1, n′ j is the number of nodes in this dimension.

[0039] Step 3: Arrange the w sorted vectors into a (n-1)×w sorting matrix R i , and normalize each row. Specifically, the sorting vectors of the two views are stacked as columns to form a sorting matrix, and the columns of the sorting matrix are normalized to unit length, where w = 2.

[0040] Step 4: Calculation SVD (singular value decomposition) of , extract the first column of the left singular vector. Arrange the vector in descending order and select the first k talents as u i neighbor nodes.

[0041] Step 5: After the k nearest neighbor nodes of n users are sorted, a directed unweighted graph is constructed. Each node in the graph represents a talent, and the edge starts from the empty set. j is u i If there are neighbors of node i, an edge from node j is added to the graph, thereby generating a unified talent graph G that integrates talent relationships and research directions.

[0042] By constructing a unified talent graph G that contains talent association information and research direction information, the relationship between talents and the possibility of collaboration can be effectively represented, thereby achieving in-depth mining and analysis of multi-dimensional dynamic data and relationships. This unified graph integrates information from two dimensions, academic association and research field, reflecting the synergy potential and cooperation paths between talents. Through complex network analysis methods, the most promising talent teams can be dynamically identified and discovered, supporting the efficient discovery and optimal allocation of talents. In addition, as time goes by, the relationships and collaboration patterns in the graph will evolve. The method of the present invention can adjust the division of talent teams in real time in a constantly changing environment, thereby providing accurate decision support.

[0043] Step 106, perform non-negative matrix decomposition on the unified talent graph to obtain the corresponding talent affiliation indicator matrix, calculate the similarity of talents in team affiliation based on the talent affiliation indicator matrix, minimize the difference between the unified talent graph and the similarity of talents in team affiliation, obtain the talent team division result, impose historical evolution constraints on the talent team division result, and construct the objective function.

[0044] After obtaining the unified talent graph G that integrates the two types of information at each moment through step 104, it is expected that accurate talent team division can be obtained at each moment, and it is expected that the talent team division will not change drastically in a short period of time. First, the non-negative matrix factorization architecture is used to transform the graph G at each moment into t Decomposed into talent affiliation indicator matrix H t ∈R n×k , H ij,t It represents the probability that talent i belongs to team j at time t, and then the talent team division result at the current moment can be obtained:

[0045]

[0046] For the talent association relationships and research direction information that evolve dynamically over time, the talent team division at continuous moments evolves smoothly. By minimizing the continuous talent team division results, the accuracy of team discovery is improved.

[0047]

[0048] In formula (2), only the case where the number of teams is the same can be processed, because the dimension of the H matrix is ​​the same. In order to better handle the case of team changes, the team evolution constraint at continuous moments is written as:

[0049]

[0050] Finally, the talent division at each moment is combined with the constraints describing the talent team at consecutive moments, and the final objective function of the method is obtained as follows:

[0051]

[0052] Among them, α is the weight coefficient, which is used for the weight ratio of the quality of current talent division and historical evolution.

[0053] In dynamic talent team discovery, the core goal is to dynamically divide talent teams according to the graph structure and ensure that the team division remains stable and evolves smoothly over time. By introducing smooth constraints, the team division is ensured to be stable over time and avoid drastic fluctuations in the team structure. This allows the team affiliation of talents to be dynamically adjusted to adapt to changing research directions and task requirements.

[0054] Step 108, optimizing and solving the objective function at each moment, obtaining the corresponding optimal talent affiliation indicator matrix, and performing dynamic talent team division according to the optimal talent affiliation indicator matrix.

[0055] The final objective function (5) is a non-convex function, so we keep other variables constant and optimize the objective for one variable to obtain the matrix H t The update formula is:

[0056]

[0057] Through iterative optimization, we can get the talent affiliation indicator matrix H at each moment: t , through H ic,t Assign talent i to team c corresponding to the maximum membership degree, and then obtain the talent team division at each moment:

[0058]

[0059] It can be understood that the method of the present invention uses artificial intelligence methods to propose a dynamic talent team real-time discovery method based on complex network analysis in view of the dynamic characteristics of talent development and mobility. It can make full use of historical evolution information and obtain accurate talent team discovery results at continuous moments based on the non-negative matrix decomposition framework, which not only improves the accuracy and real-time performance of talent team division, but also provides scientific and reliable support for talent management decision-making.

[0060] In the above-mentioned dynamic talent team discovery method based on complex network analysis, by constructing a unified talent graph containing talent association information and research direction information, it is possible to effectively integrate multi-dimensional data, improve the analysis accuracy of talent relationships and collaboration potential, perform non-negative matrix decomposition on the unified talent graph, and obtain the corresponding talent affiliation indicator matrix. The similarity of talents in team affiliation is calculated according to the talent affiliation indicator matrix, and the difference between the talent unified graph and the similarity of talents in team affiliation is minimized to obtain the talent team division result, and historical evolution constraints are imposed on the talent team division result, and an objective function is constructed to achieve smooth evolution of team division in a dynamic environment, effectively support the adjustment of team structure at different time points, and ensure that the team can adapt to the dynamic changes in talent demand. The present invention avoids the deviation of subjective judgment through data-driven quantitative analysis, and ensures that the team division is more scientific and accurate. The embodiment of the present invention can adjust the team division in real time, optimize the allocation of talent resources, and improve the efficiency of team collaboration and innovation ability.

[0061] In one embodiment, minimizing the difference between the unified talent graph and the similarity between talents in team affiliation to obtain the talent team division result includes: minimizing the difference between the unified talent graph and the similarity between talents in team affiliation to obtain the talent team division result as follows:

[0062]

[0063] Among them, G t is the unified talent graph at time t, H t is the talent affiliation indicator matrix at time t, H ij,t is the probability that talent i belongs to team j at time t, is the square of the F-norm of the matrix, (·) T is the transpose of the matrix, n is the number of talents, and k is the number of talent teams.

[0064] In one embodiment, the historical evolution constraint is:

[0065]

[0066] Among them, L2 is the historical evolution constraint, H is the talent affiliation indicator matrix, t is the time, (·) T is the transpose of the matrix, is the square of the F-norm of the matrix.

[0067] In one embodiment, the objective function is:

[0068]

[0069] Among them, G t is the unified talent graph at time t, H t is the talent affiliation indicator matrix at time t, H ij,t is the probability that talent i belongs to team j at time t, is the square of the F-norm of the matrix, (·) T is the transpose of the matrix, n is the number of talents, k is the number of talent teams, and α is the weight coefficient.

[0070] In one embodiment, optimizing and solving the objective function at each moment to obtain the corresponding optimal talent affiliation indication matrix includes: optimizing and solving the objective function at each moment according to a preset talent affiliation indication matrix update rule to obtain the corresponding optimal talent affiliation indication matrix.

[0071] In one embodiment, performing dynamic talent team division according to the optimal talent affiliation indication matrix includes: obtaining the probability of talents belonging to different teams according to the optimal talent affiliation indication matrix, and dividing the talents into teams corresponding to the maximum probability values, thereby achieving dynamic talent team division.

[0072] In a specific embodiment, a series of experiments were conducted on two datasets to evaluate the effectiveness of the proposed method DTTD (Dynamic Talent Team Discovery). DTTD was compared with FacetNet and SCI. The FacetNet algorithm is a dynamic community detection method, and the SCI algorithm is a static community detection method. First, a dynamic talent dataset is constructed, which contains researchers from a certain school and covers cross-disciplinary research directions. Since there are similar research fields between disciplines, there may be academic connections between researchers. A talent association network is constructed based on the papers published by each researcher. If two researchers appear in the same article, an edge is added between them. In addition, a research direction attribute network is constructed based on the research direction of each researcher. A time slice is constructed every year, so the entire dataset contains multiple time slices. As new researchers join and some researchers leave, the number of nodes and research direction attributes of the network will change over time. As Figure 2 The schematic diagram of the experimental results of the algorithm on the real network dataset Q shows that DTTD has achieved stable and excellent community talent division performance. The Q value obtained by the proposed method remains at a high level in all steps. This proves that the method we proposed can consistently group talents with similar research directions and close academic interests into the same team at all times.

[0073] In addition, the paper collaborator network DBLP is extracted from dblp.org. The network contains information about authors who have published at least two papers in different fields in multiple years, and the main fields in which the authors published their papers are used as the research direction attributes of the nodes. As can be seen from Table 1, the method of the present invention achieves the best performance at all seven moments. This is mainly because the method of the present invention considers the academic connections and research directions between authors at the same time, and also makes full use of historical information, thus obtaining real-time and accurate division.

[0074] Table 1 Experimental results of the algorithm on the DBLP collaborative author dataset Q

[0075]

[0076] 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 1At 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.

[0077] In one embodiment, Figure 3 As shown, a dynamic talent team discovery device based on complex network analysis is provided, comprising:

[0078] A parameter acquisition module 302 is used to construct a talent association matrix according to the academic association relationships between talents, and to construct a research direction matrix according to the research directions of talents;

[0079] The graph construction module 304 is used to sort the similarity of each talent with other talents according to the talent association matrix and the research direction matrix to obtain a sorting vector, perform multi-dimensional information fusion on the sorting vector to obtain a sorting matrix, perform singular value decomposition on the sorting matrix, extract the first column of the left singular vector, select the first K talents with the highest similarity as the neighbors of the current talent, and construct a unified talent graph according to each talent and the corresponding neighbor; the nodes of the unified talent graph represent talents, and the edges represent the association relationship between talents;

[0080] The model building module 306 is used to perform non-negative matrix decomposition on the talent unified graph to obtain the corresponding talent affiliation indicator matrix, calculate the similarity of talents in team affiliation according to the talent affiliation indicator matrix, minimize the difference between the talent unified graph and the similarity of talents in team affiliation, obtain the talent team division result, impose historical evolution constraints on the talent team division result, and construct the objective function; the talent affiliation indicator matrix represents the probability of each talent belonging to different teams;

[0081] The model solving module 308 is used to optimize and solve the objective function at each moment, obtain the corresponding optimal talent affiliation indicator matrix, and perform dynamic talent team division according to the optimal talent affiliation indicator matrix.

[0082] For the specific limitations of the dynamic talent team discovery device based on complex network analysis, please refer to the limitations of the dynamic talent team discovery method based on complex network analysis above, which will not be repeated here. Each module in the above-mentioned dynamic talent team discovery device based on complex network analysis can be implemented in whole or in part through 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 of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0083] 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 4 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 dynamic talent team discovery method based on complex network analysis 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 covering the display screen, or a button, trackball or touchpad set on the computer device housing, or an external keyboard, touchpad or mouse, etc.

[0084] Those skilled in the art will understand that Figure 4 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.

[0085] 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.

[0086] 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.

[0087] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed 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 can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an 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).

[0088] 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.

[0089] 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 dynamic talent team discovery method based on complex network analysis, characterized in that: The method comprises: Build a talent association matrix based on the academic associations between talents, and build a research direction matrix based on the research directions of talents; According to the talent association matrix and the research direction matrix, each talent is sorted by similarity with other talents to obtain a sorting vector, multi-dimensional information fusion is performed on the sorting vector to obtain a sorting matrix, singular value decomposition is performed on the sorting matrix, the first column of the left singular vector is extracted, and the first K talents with the highest similarity are selected as neighbors of the current talent, and a talent unified graph is constructed according to each talent and the corresponding neighbor; the nodes of the talent unified graph represent talents, and the edges represent the association relationship between talents; Performing non-negative matrix decomposition on the talent unified graph to obtain a corresponding talent affiliation indicator matrix, calculating the similarity of talents in team affiliation according to the talent affiliation indicator matrix, minimizing the difference between the talent unified graph and the similarity of talents in team affiliation, obtaining a talent team division result, applying historical evolution constraints to the talent team division result, and constructing an objective function; The objective function at each moment is optimized and solved to obtain the corresponding optimal talent affiliation indicator matrix, and dynamic talent team division is performed based on the optimal talent affiliation indicator matrix.

2. The method according to claim 1, characterized in that Minimize the difference between the talent unified graph and the similarity of talents in team affiliation, and obtain the talent team division results including: Minimize the difference between the talent unified graph and the similarity of talents in team affiliation, and obtain the talent team division result as follows: Among them, G t is the unified talent graph at time t, H t is the talent affiliation indicator matrix at time t, H ij,t is the probability that talent i belongs to team j at time t, is the square of the F-norm of the matrix, (·) T is the transpose of the matrix, n is the number of talents, and k is the number of talent teams.

3. The method according to claim 1, characterized in that The historical evolution constraints are: Among them, L2 is the historical evolution constraint, H is the talent affiliation indicator matrix, t is the time, (·) T is the transpose of the matrix, is the square of the F-norm of the matrix.

4. The method according to claim 1, characterized in that: The objective function is: Among them, G t is the unified talent graph at time t, H t is the talent affiliation indicator matrix at time t, H ij,t is the probability that talent i belongs to team j at time t, is the square of the F-norm of the matrix, (·) T is the transpose of the matrix, n is the number of talents, k is the number of talent teams, and α is the weight coefficient.

5. The method according to claim 1, characterized in that The objective function at each moment is optimized and solved to obtain the corresponding optimal talent affiliation indicator matrix, including: According to the preset updating rules of the talent affiliation indicator matrix, the objective function at each moment is optimized and solved to obtain the corresponding optimal talent affiliation indicator matrix.

6. The method according to claim 1, characterized in that Dynamic talent team division based on the optimal talent affiliation indicator matrix includes: According to the optimal talent affiliation indicator matrix, the probability of talents belonging to different teams is obtained, and the talents are divided into teams corresponding to the maximum probability values ​​to achieve dynamic talent team division.

7. A dynamic talent team discovery device based on complex network analysis, characterized in that: The device comprises: The parameter acquisition module is used to construct a talent association matrix based on the academic association relationships between talents, and to construct a research direction matrix based on the research directions of talents; A graph construction module is used to sort the similarity of each talent with other talents according to the talent association matrix and the research direction matrix to obtain a sorting vector, perform multi-dimensional information fusion on the sorting vector to obtain a sorting matrix, perform singular value decomposition on the sorting matrix, extract the first column of the left singular vector, select the first K talents with the highest similarity as neighbors of the current talent, and construct a unified talent graph according to each talent and the corresponding neighbor; the nodes of the unified talent graph represent talents, and the edges represent the association relationship between talents; A model building module is used to perform non-negative matrix decomposition on the talent unified graph to obtain a corresponding talent affiliation indicator matrix, calculate the similarity of talents in team affiliation according to the talent affiliation indicator matrix, minimize the difference between the talent unified graph and the similarity of talents in team affiliation, obtain the talent team division result, impose historical evolution constraints on the talent team division result, and construct an objective function; The model solving module is used to optimize and solve the objective function at each moment, obtain the corresponding optimal talent affiliation indicator matrix, and perform dynamic talent team division based on the optimal talent affiliation indicator matrix.

8. The device according to claim 7, characterized in that The objective function is: Among them, G t is the unified talent graph at time t, H t is the talent affiliation indicator matrix at time t, H ij,t is the probability that talent i belongs to team j at time t, is the square of the F-norm of the matrix, (·) T is the transpose of the matrix, n is the number of talents, k is the number of talent teams, and α is the weight coefficient.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images