GPT-based task allocation method and device

Abstract

The invention relates to a task allocation method and device based on GPT, and belongs to the technical field of computers.The method comprises the steps that target personnel and parameter information of the target personnel are obtained, the target personnel are classified based on the parameter information, and a resource pool is obtained; obtaining attribute data of a target person, and preprocessing the attribute data; taking the preprocessed attribute data as a training set to train a GPT model to obtain a target GPT model; and under the target GPT model, generating a target team matched with the task demand based on the relation graph of the target personnel and the task demand, and allocating the task to the target personnel in the target team based on the attribute information of the task demand. According to the method, accurate classification and efficient matching of the target personnel are realized, the accuracy and efficiency of resource allocation and task allocation are improved, team cooperation and individual occupational development are further promoted, and the overall decision-making ability and competitiveness of an organization are improved.

Description

Technical Field

[0001] The present invention belongs to the field of computer technology, and in particular, relates to a task allocation method and device based on GPT. Background Art

[0002] In the related art, in many organizations and enterprises, team building and task allocation are important aspects of daily work. Team building involves selecting appropriate members so that they can work together to achieve common goals. Task allocation involves distributing work tasks to team members to ensure efficient work execution and resource utilization. However, in large organizations or enterprises, team building and task allocation may face some challenges. Therefore, how to better achieve team building and task allocation has become an urgent problem to be solved. Summary of the invention

[0003] In view of the above shortcomings of the prior art, the purpose of the invention is to provide a task allocation method and device based on GPT. The method realizes accurate classification and efficient matching of target personnel, and improves the accuracy and efficiency of resource allocation and task allocation.

[0004] In a first aspect of the present invention, a task assignment method based on GPT is proposed, comprising: S1, obtaining a target person and parameter information of the target person, and classifying the target person based on the parameter information to obtain a resource pool; S2, obtaining attribute data of the target person, and preprocessing the attribute data; S3, training a GPT model using the preprocessed attribute data as a training set to obtain a target GPT model; S4, under the target GPT model, based on the relationship map of the target person and the task requirements, generating a target team matching the task requirements, and assigning tasks to the target person in the target team based on the attribute information of the task requirements.

[0005] Furthermore, the parameter information includes job responsibilities, wherein the target personnel and the parameter information of the target personnel are obtained, and the target personnel are classified based on the parameter information to obtain a resource pool, including: initializing the cluster center based on the target personnel; based on the Euclidean distance formula Calculate the distance from the target person to each cluster center, and assign the target person to the cluster center with the smallest distance from the target person, where d(x i , C k ) represents the data point x i With cluster center C k The Euclidean distance between them, j = 1 means the starting index of the sum is 1, m means the number of job responsibilities, x ij Represents data point x iThe score or attribute value of the jth job responsibility, C kj Represents the cluster center C k The j-th eigenvalue of ; based on the cluster center update formula The minimum cluster center is updated until the updated minimum cluster center is less than a preset threshold, and the resource pool is obtained, wherein S k represents the set of all data points in the kth cluster, For all S k The data point x of the set i The sum of .

[0006] Furthermore, the attribute data is preprocessed, including: formatting the attribute data, deduplicating and checking the integrity of the formatted attribute data; based on the attribute data after the integrity check, determining and storing the skill label and experience value of the target person.

[0007] Furthermore, it also includes determining a relationship map of the target personnel, including: taking the target personnel as nodes and target relationships as edges, and constructing a relationship map of the target personnel, wherein the weight of each edge represents the affinity and tacit understanding between the target personnel.

[0008] Furthermore, it also includes: obtaining the centrality score of the target person; based on Calculate the affinity and tacit understanding between the target persons, where Affinity(i, j) represents the affinity or tacit understanding between target person i and target person j. It represents half of the sum of the centrality scores of target person i and target person j, and InteractionWeight(i, j) represents the weight of the interaction relationship between target person i and target person j.

[0009] Furthermore, the attribute information of the task requirement includes task level and deadline, wherein allocating the task to the target person in the target team based on the attribute information of the task requirement includes: determining the workload, skill label and experience value of the target person in the target team; and allocating the task to the target person in the target team based on the task level and deadline.

[0010] Further, the task is assigned to the target person in the target team based on the task level and the deadline, including: when the first task level in the task requirement is level one and the deadline is the first deadline, the task is assigned to the first target person in the target team; when the second task level in the task requirement is level two and the deadline is the second deadline, the task is assigned to the second target person in the target team, wherein the workload level of the first target person is higher than the workload level of the second target person, and the experience value of the first target person is higher than the experience value of the second target person.

[0011] The second aspect of the present invention proposes a task assignment device based on GPT, including: a classification module, used to obtain a target person and parameter information of the target person, and classify the target person based on the parameter information to obtain a resource pool; a data processing module, used to obtain attribute data of the target person and preprocess the attribute data; a training module, used to train a GPT model using the preprocessed attribute data as a training set to obtain a target GPT model; an assignment module, used to generate a target team matching the task requirements based on the relationship map of the target person and the task requirements under the target GPT model, and assign tasks to the target person in the target team based on the attribute information of the task requirements.

[0012] According to a third aspect of the present invention, an electronic device is proposed, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute any of the methods described in the first aspect of the present invention.

[0013] According to a fourth aspect of the present invention, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable the computer to execute any one of the methods according to the first aspect of the present invention.

[0014] The beneficial effects of the present invention are as follows:

[0015] The GPT-based task allocation method and device of the present invention obtains target personnel and parameter information of target personnel, and classifies target personnel based on parameter information to obtain resource pool; obtains attribute data of target personnel, and preprocesses attribute data; uses preprocessed attribute data as training set to train GPT model to obtain target GPT model; under target GPT model, based on relationship map and task requirements of target personnel, generates target team matching task requirements, and assigns tasks to target personnel in target team based on attribute information of task requirements. This method realizes accurate classification and efficient matching of target personnel by introducing technologies such as GPT model and relationship map, improves accuracy and efficiency of resource allocation and task allocation, and thus promotes team collaboration and personal career development, and enhances overall decision-making ability and competitiveness of organization. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The accompanying drawings are only used to illustrate specific embodiments and are not considered to limit the present invention. In the entire drawings, the same reference symbols represent the same components. Obviously, the drawings described below are only some embodiments recorded in the embodiments of the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings.

[0017] Figure 1 is a flowchart of a GPT-based task allocation method according to an embodiment of the present invention;

[0018] Figure 2 is a flowchart of a GPT-based task allocation method according to a specific embodiment of the present invention;

[0019] Figure 3 is a structural block diagram of a GPT-based task allocation device according to an embodiment of the present invention;

[0020] Figure 4 It is a structural block diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0021] In order to enable those skilled in the art to better understand the technical solutions in the embodiments of the present invention, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. It should be understood that these descriptions are only exemplary and are not intended to limit the scope of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in the field without making creative work should fall within the scope of protection of the present invention.

[0022] Furthermore, in the following description, descriptions of well-known structures and technologies are omitted to avoid unnecessarily obscuring the concepts disclosed in the present invention.

[0023] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the orientation or position relationship indicated by the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inside", "outside", etc. is based on the orientation or position relationship shown in the drawings, which is only for the convenience of describing the present invention and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation of the present invention. In addition, the terms "first", "second", and "third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. The terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be a connection between the two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0024] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the 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 invention. Instead, they are merely examples of methods and systems consistent with some aspects of the present invention as detailed in the appended claims.

[0025] The present invention proposes a GPT-based task allocation method, apparatus, and related equipment. Specifically, the GPT-based task allocation method, apparatus, and related equipment of an embodiment of the present invention are described below with reference to the accompanying drawings.

[0026] Figure 1 1 is a flow chart of a GPT-based task allocation method according to an embodiment of the present invention. It should be noted that the GPT-based task allocation method of the embodiment of the present invention can be applied to the GPT-based task allocation device of the embodiment of the present invention. The GPT-based task allocation device can be configured on an electronic device or in a server. This embodiment of the present application does not limit this.

[0027] like Figure 1 As shown, the task allocation method based on GPT includes:

[0028] S110, acquiring target personnel and parameter information of the target personnel, and classifying the target personnel based on the parameter information to obtain a resource pool.

[0029] In an embodiment of the present invention, the parameter information includes job responsibilities. That is, when the job responsibilities of the target personnel are obtained, the target personnel can be classified based on the job responsibilities, that is, the target personnel can be divided into different categories based on the job responsibilities. The resource pool is used to store a collection of skill categories or job responsibilities.

[0030] For example, the target personnel responsible for front-end development are classified into the front-end resource pool, and the target personnel responsible for back-end development are classified into the back-end resource pool. Project managers, product managers, testers, etc. are also placed in the corresponding resource pools.

[0031] S120, acquiring attribute data of the target person and preprocessing the attribute data.

[0032] Among them, attribute data includes but is not limited to expertise, work experience, educational background, skills, interests, etc.

[0033] In the embodiment of the present invention, when the target personnel are divided into different resource pools based on job responsibilities, the attribute data of the target personnel can be provided by filling out questionnaires, entering skill information by themselves, and expert group evaluation, and then the attribute data is preprocessed. The specific implementation method can refer to the subsequent embodiments.

[0034] S130, using the preprocessed attribute data as a training set to train the GPT model to obtain a target GPT model.

[0035] In an embodiment of the present invention, when the attribute data is preprocessed, the preprocessed attribute data can be used as a training set, and the training set can be pre-trained in an unsupervised manner, and then fine-tuned using a supervised learning method to improve the accuracy and adaptability of the target GPT model. When training the model, matching labels between task requirements and team members can be introduced as supervisory signals to help the target GPT model learn the rules of task allocation.

[0036] S140, under the target GPT model, based on the relationship map of the target personnel and the task requirements, a target team matching the task requirements is generated, and the task is assigned to the target personnel in the target team based on the attribute information of the task requirements.

[0037] Among them, the attribute information of task requirements includes task level, deadline,

[0038] In the embodiment of the present invention, the workload, skill label and experience value of the target personnel in the target team are determined; and the task is assigned to the target personnel in the target team based on the task level and deadline. The specific implementation method can be referred to the subsequent embodiments.

[0039] According to the GPT-based task allocation method of an embodiment of the present invention, the target personnel and the parameter information of the target personnel are obtained, and the target personnel are classified based on the parameter information to obtain a resource pool; the attribute data of the target personnel are obtained, and the attribute data is preprocessed; the preprocessed attribute data is used as a training set to train the GPT model to obtain a target GPT model; under the target GPT model, based on the relationship map and task requirements of the target personnel, a target team matching the task requirements is generated, and the task is assigned to the target personnel in the target team based on the attribute information of the task requirements. This method achieves accurate classification and efficient matching of target personnel by introducing technologies such as the GPT model and relationship map, improves the accuracy and efficiency of resource allocation and task allocation, and thus promotes team collaboration and personal career development, and enhances the overall decision-making ability and competitiveness of the organization.

[0040] In order to make it easier for those skilled in the art to understand the present invention, Figure 2 is a task allocation method based on GPT according to a specific embodiment of the present invention, such as Figure 2 As shown, the GPT-based task allocation method includes:

[0041] S210, acquiring target personnel and parameter information of the target personnel, and classifying the target personnel based on the parameter information to obtain a resource pool.

[0042] Among them, the parameter information includes job responsibilities.

[0043] That is to say, the target personnel are divided into different resource pools according to their job responsibilities. Each resource pool represents a job responsibility, such as development, testing, design, etc. The Python programming language can be used to develop a program to classify job responsibilities through clustering algorithms (such as K-means), and the target personnel can be divided into different categories according to their job responsibilities.

[0044] Among them, taking the K-means clustering algorithm as an example, the target personnel are taken as data points and the job responsibilities are taken as clustering features. The K-means algorithm is used to divide the data points (target personnel) into K clusters (resource pools), and each cluster represents a job responsibility.

[0045] In an embodiment of the present invention, based on the target person, the cluster center is initialized; based on the Euclidean distance formula Calculate the distance from the target person to each cluster center, and assign the target person to the cluster center with the smallest distance from the target person, where d(x i , C k ) represents the data point x i With cluster center C k The Euclidean distance between them, j = 1 means the starting index of the sum is 1, m means the number of job responsibilities, xij Represents data point x i The score or attribute value of the jth job responsibility, C kj Represents the cluster center C k The jth eigenvalue of ; Calculate the distance from the target person to each cluster center, and assign the target person to the cluster center with the smallest distance from the target person, based on the cluster center update formula The minimum cluster center is updated until the updated minimum cluster center is less than a preset threshold, and the resource pool is obtained, wherein S k represents the set of all data points in the kth cluster, For all S k The data point x of the set i , until the updated minimum cluster center is less than a preset threshold, the resource pool is obtained.

[0046] For example, 1) there are N data points (target personnel) and K cluster centers (C1, C2, ..., CK). In the initialization stage, these K cluster centers are usually K points randomly selected from N data points; 2) for each data point (target personnel), calculate its Euclidean distance to each cluster center Ck, and then assign the data point to the cluster center Ck closest to it; 3) for each cluster (resource pool), recalculate the cluster center based on the cluster center update formula. The new cluster center is the mean of all data points in the cluster; 4) repeat steps 2) and 3) until the stop condition is met (such as the change in the cluster center is less than a certain threshold or reaches the preset maximum number of iterations; 5) eventually, each cluster (resource pool) will contain a group of personnel, and the personnel in each group are relatively close in job responsibilities. The center of each cluster can represent the average job responsibilities of the personnel in the cluster. Among them, the purpose of updating the cluster center is the mean of the data point, which can represent the overall characteristics or average level of employees in the resource pool. And it can optimize the clustering effect and achieve further subdivision.

[0047] The cluster center represents the core feature or level of a certain skill category or position, and the resource pool is used to store a collection of skill categories or job responsibilities.

[0048] S220, acquiring attribute data of the target person and preprocessing the attribute data.

[0049] Among them, attribute data includes but is not limited to expertise, work experience, educational background, skills, interests, etc.

[0050] In an embodiment of the present invention, when the target personnel are divided into different resource pools based on job responsibilities, the attribute data of the target personnel can be provided by filling out questionnaires, entering skill information by themselves, and expert group evaluation, and then the attribute data can be pre-processed.

[0051] For example, the target personnel can be employees. Questionnaires can be distributed to employees to fill in basic information mainly based on work experience and educational background. Professional skills can be scored and rated through examinations and assessments. An expert group can be set up to rate the positions based on the information and work performance of the employees. The above questionnaires, examinations and expert group assessments all need to be set up according to the resource pool. Among them, the relationship data between employees can be sorted out according to historical work content, and the healthy tacit understanding of employees can be analyzed in combination with achievements, performance and loopholes. Then the affinity between employees can be obtained through work status feedback.

[0052] In an embodiment of the present invention, when the attribute data of the target person is obtained, the attribute data can be formatted, and the formatted attribute data can be deduplicated and integrity checked; based on the attribute data after the integrity check, the skill label and experience value of the target person are determined and stored.

[0053] For example, formatting attribute data includes using full spelling or phonetic abbreviations for names, using dictionary values ​​for skills, etc. If attribute data contains multiple formats (such as text, pictures, tables, etc.), it needs to be converted into a unified format for easy processing.

[0054] For example, deduplication of formatted attribute data includes checking whether there are duplicate employee records in the formatted attribute data. If there are duplicate records, the duplicate records need to be deleted or merged.

[0055] For example, the completeness check includes verifying whether the record of each target person contains necessary information, such as name, skills, etc. For missing data, it is supplemented through other channels (such as internal databases, target person files, etc.).

[0056] In an embodiment of the present invention, based on the attribute data after the integrity check, the target personnel can be marked with corresponding skill tags based on the target personnel's skills and experience. For example, the skills can be divided into categories such as programming, design, project management, etc., and corresponding skill tags are assigned to each target person. Labeling helps to quickly find qualified employees in the subsequent team building and task assignment process.

[0057] Among them, the skill level can be divided into different levels such as beginner, intermediate, advanced, etc., and a numerical value is assigned to each level.

[0058] In an embodiment of the present invention, the skill labels and experience values ​​of the target personnel may be stored based on a database.

[0059] In the embodiments of the present invention, when processing the target personnel data, relevant laws and policies on data security and privacy protection must be strictly followed. It is ensured that only authorized personnel can access and use the data, and appropriate technical and management measures are taken to protect the security and privacy of the data.

[0060] For example, taking an employee of the front-end resource pool as an example, the employee data is as follows: Employee ID: 1001, Name: A, Gender: Male, Age: 30, Education: Bachelor's degree, Resource pool: Front-end resource pool, Position: Software Engineer, Skills: Skill name: Vue, Skill score: 5; Skill name: React, Skill score: 3; Skill name: NodeJS, Skill score: 2; Skill name: Python, Skill score: 1; Work experience: 5, Teamwork ability: 4, Problem-solving ability: 5, Communication ability: 4, Workload: Moderate.

[0061] S230, using the preprocessed attribute data as a training set to train the GPT model to obtain a target GPT model.

[0062] S240, determining the relationship map of the target personnel.

[0063] Among them, the relationship map can be understood as the cooperation record between target personnel, including the projects involved, the frequency of cooperation, the effect of cooperation, etc.

[0064] In an embodiment of the present invention, the target persons are taken as nodes and the target relationships are taken as edges to construct a relationship graph of the target persons, wherein the weight of each edge represents the affinity and tacit understanding between the target persons.

[0065] In an embodiment of the present invention, the centrality score of the target person is obtained; based on Calculate the affinity and tacit understanding between the target persons, where Affinity(i, j) represents the affinity or tacit understanding between target person i and target person j. It represents half of the sum of the centrality scores of target person i and target person j, and InteractionWeight(i, j) represents the weight of the interaction relationship between target person i and target person j.

[0066] For example, taking the target personnel as employees, 1) Initialization: Assume that there are N employees, and the initial "centrality" or "importance" score of each employee is 1 / N (that is, uniformly distributed). Create an N×N matrix M to represent the interaction relationship between employees. If employee i and employee j interact, the value of M[i][j] is the weight of the interaction (determined by factors such as the frequency and depth of the interaction); otherwise it is 0; 2) Iterative calculation: set a damping factor β (usually between 0.8 and 0.9), and a convergence threshold ε (used to determine when to stop iterating). In each iteration, for each employee i, update its centrality score. Among them, v[j] is the centrality score of employee j in the previous iteration, M[j][i] represents the matrix of the interaction relationship between employee j and employee i, that is, the interaction intensity or frequency between employee j and employee i, N represents the total number of employees, and the last item To avoid the score being too low or 0, similar to the random walk term in PageRank, when the difference between v' and v is less than the convergence threshold ε, the iteration stops; 3) After having the centrality score of each employee, the affinity and tacit understanding between employees can be calculated based on these scores. For example, the following formula can be used to calculate the affinity or tacit understanding between employee i and employee j: Based on Calculate the affinity and tacit understanding between the target persons, where Affinity(i, j) represents the affinity or tacit understanding between target person i and target person j. represents half of the sum of the centrality scores of target person i and target person j, and InteractionWeight(i, j) represents the weight of the interaction relationship between target person i and target person j. This formula combines the centrality scores of employees and their interaction weights to more comprehensively reflect the affinity and tacit understanding between them.

[0067] S250, under the target GPT model, generates a target team that matches the task requirements based on the relationship map of the target personnel and the task requirements.

[0068] In the embodiment of the present invention, when task requirements are obtained, key requirement information can be extracted based on analyzing the skill requirements and task content in the task requirements. A candidate set of employees that meet the task requirements is screened from the resource pool. Then, based on the relationship map, a target combination of personnel with high affinity and tacit understanding, i.e., a target team, is preferentially selected.

[0069] Among them, combinatorial optimization algorithms (such as genetic algorithms, simulated annealing, etc.) are used to optimize the team configuration to ensure the optimal team structure. At the same time, the workload and availability of target personnel are considered to avoid over-allocation of tasks.

[0070] Among them, the attribute information of the task requirement includes the task level and deadline.

[0071] S260, determine the workload, skill label, and experience value of the target personnel in the target team.

[0072] S270, allocating the task to the target person in the target team based on the task level and deadline.

[0073] In an embodiment of the present invention, when the first task level in the task requirement is level one and the deadline is the first deadline, the task is assigned to the first target person in the target team; when the second task level in the task requirement is level two and the deadline is the second deadline, the task is assigned to the second target person in the target team, wherein the workload level of the first target person is higher than the workload level of the second target person, and the experience value of the first target person is higher than the experience value of the second target person.

[0074] In other words, tasks can be reasonably allocated to target personnel in the target team according to the complexity of the tasks, deadlines and the load of target personnel in the target team, so as to generate the optimal task allocation plan.

[0075] Among them, if there are personnel changes or changes in task requirements, the system can respond quickly and regenerate a new task allocation plan to meet the changing needs of the project.

[0076] For example, there is a requirement for an activity display page in the task requirements. According to the model analysis, this requirement involves product, UI, front-end, back-end and testers, among which UI and front-end are the focus. The product, back-end and test tasks of this requirement can be assigned to personnel with corresponding capabilities and average experience, while UI and front-end tasks require personnel with corresponding skills, rich experience and low current workload to concentrate on them.

[0077] In one embodiment of the present invention, after the task is assigned, the target GPT model can be iterated and optimized by continuously collecting information, including task execution status and member feedback of the target team. The accuracy and effect of team member matching and task assignment can be improved by adjusting the target GPT model parameters, increasing the diversity of training data, and introducing more accurate task requirements and team member information.

[0078] According to the GPT-based task allocation method of the embodiment of the present invention, through the analysis and recommendation of the target GPT model, it is possible to more accurately understand the task requirements, achieve accurate matching of team members and optimize task allocation. It is possible to improve the execution efficiency and quality of tasks, reduce human resource costs, reduce the impact of human factors, reduce the error rate, thereby enhancing the overall competitiveness of the organization, and can allocate tasks according to the skills and personality characteristics of team members, which can better play the advantages of each target person and improve the job satisfaction and sense of belonging of the target person. In addition, the flexible adjustment and optimization mechanism of the present invention can also meet the changing needs of the project and improve the adaptability and response speed of the organization.

[0079] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods of each embodiment of the present application.

[0080] According to one aspect of an embodiment of the present invention, a task allocation device based on GPT is also proposed. Figure 3 is a structural block diagram of a task allocation device based on GPT according to an embodiment of the present invention; Figure 3 As shown, including:

[0081] The classification module 310 is used to obtain target personnel and parameter information of the target personnel, and classify the target personnel based on the parameter information to obtain a resource pool;

[0082] The data processing module 320 is used to obtain the attribute data of the target person and pre-process the attribute data;

[0083] A training module 330 is used to train a GPT model using the preprocessed attribute data as a training set to obtain a target GPT model;

[0084] The allocation module 340 is used to generate a target team matching the task requirements based on the relationship map of the target personnel and the task requirements under the target GPT model, and allocate the task to the target personnel in the target team based on the attribute information of the task requirements.

[0085] According to the GPT-based task allocation device of the embodiment of the present invention, the target personnel and the parameter information of the target personnel are obtained, and the target personnel are classified based on the parameter information to obtain a resource pool; the attribute data of the target personnel are obtained, and the attribute data are preprocessed; the preprocessed attribute data is used as a training set to train the GPT model to obtain a target GPT model; under the target GPT model, based on the relationship map and task requirements of the target personnel, a target team matching the task requirements is generated, and the task is assigned to the target personnel in the target team based on the attribute information of the task requirements. Thus, by introducing technologies such as the GPT model and the relationship map, accurate classification and efficient matching of target personnel are achieved, the accuracy and efficiency of resource allocation and task allocation are improved, and team collaboration and personal career development are promoted, and the overall decision-making ability and competitiveness of the organization are enhanced.

[0086] Optionally, the parameter information includes job responsibilities, wherein the classification module 310 is specifically used to initialize the cluster center based on the target personnel; based on the Euclidean distance formula Calculate the distance from the target person to each cluster center, and assign the target person to the cluster center with the smallest distance from the target person, where d(x i , C k ) represents the data point x i With cluster center C k The Euclidean distance between them, j = 1 means the starting index of the sum is 1, m means the number of job responsibilities, x ij Represents data point x i The score or attribute value of the jth job responsibility, C kj Represents the cluster center C k The j-th eigenvalue of ; based on the cluster center update formula The minimum cluster center is updated until the updated minimum cluster center is less than a preset threshold, and the resource pool is obtained, wherein S k represents the set of all data points in the kth cluster, For all S k The data point x of the set i The sum of .

[0087] Optionally, the data processing module 320 is specifically used to format the attribute data, deduplicate and perform integrity check on the formatted attribute data; based on the attribute data after the integrity check, determine the skill label and experience value of the target person and store them.

[0088] Optionally, a determination module is also included, which is used to construct a relationship graph of the target personnel by taking the target personnel as nodes and the target relationships as edges, wherein the weight of each edge represents the affinity and tacit understanding between the target personnel.

[0089] Optionally, a determination module is specifically used to obtain a centrality score of the target person; based on Calculate the affinity and tacit understanding between the target persons, where Affinity(i, j) represents the affinity or tacit understanding between target person i and target person j. It represents half of the sum of the centrality scores of target person i and target person j, and InteractionWeight(i, j) represents the weight of the interaction relationship between target person i and target person j.

[0090] Optionally, the attribute information of the task requirement includes task level and deadline, wherein the allocation module 340 is specifically used to determine the workload, skill label and experience value of the target personnel in the target team; and allocate the task to the target personnel in the target team based on the task level and deadline.

[0091] Optionally, the allocation module 340 is specifically used to allocate the task to a first target person in the target team when the first task level in the task requirement is level one and the deadline is the first deadline; and to allocate the task to a second target person in the target team when the second task level in the task requirement is level two and the deadline is the second deadline, wherein the workload level of the first target person is higher than the workload level of the second target person, and the experience value of the first target person is higher than the experience value of the second target person.

[0092] According to one aspect of an embodiment of the present invention, an electronic device is provided.

[0093] Figure 4 Schematic diagram of the structure of an electronic device according to an embodiment of the present invention. Figure 4 As shown, the electronic device may include one or more ( Figure 4Only one is shown in the figure) processor 102 (processor 102 may include but is not limited to a microprocessor (Microprocessor Unit, referred to as MPU) or a programmable logic device (Programmable logic device, referred to as PLD)) and a memory 104 for storing data. In an exemplary embodiment, the electronic device may also include a transmission device 106 and an input / output device 108 for communication functions. It can be understood by those skilled in the art that Figure 4 The structure shown is for illustration only and does not limit the structure of the above terminal device. Figure 4 More or fewer components as shown, or with Figure 4 Equivalent functions or comparisons shown Figure 4 A different configuration with more features is shown.

[0094] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the GPT-based task allocation method in the embodiment of the present invention. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, to implement the above method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the terminal device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0095] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the switching device. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0096] The present invention provides a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to enable the computer to execute a task allocation method based on GPT.

[0097] The applicant of the present invention has made a detailed explanation and description of the implementation examples of the present invention in conjunction with the drawings in the specification. However, those skilled in the art should understand that the above implementation examples are only preferred implementation schemes of the present invention, and the detailed description is only to help readers better understand the spirit of the present invention, but not to limit the scope of protection of the present invention. On the contrary, any improvements or modifications based on the inventive spirit of the present invention should fall within the scope of protection of the present invention.

[0098] In addition, each functional unit in each embodiment of the present invention may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0099] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present invention. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present invention.

[0100] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the embodiments of the present invention, rather than to limit them. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the above embodiments, or replace some of the technical features therein by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention. Any changes or replacements that can be easily thought of by those skilled in the art within the technical scope disclosed in the present invention should be covered within the protection scope of the present invention.

Claims

1. A task allocation method based on GPT, characterized in that: include: S1, obtaining target personnel and parameter information of the target personnel, and classifying the target personnel based on the parameter information to obtain a resource pool; S2, obtaining attribute data of the target person and preprocessing the attribute data; S3, training the GPT model using the preprocessed attribute data as a training set to obtain a target GPT model; S4, under the target GPT model, based on the relationship map of the target personnel and the task requirements, a target team matching the task requirements is generated, and the task is assigned to the target personnel in the target team based on the attribute information of the task requirements.

2. The GPT-based task allocation method according to claim 1, characterized in that: The parameter information includes job responsibilities, wherein a target person and parameter information of the target person are obtained, and the target person is classified based on the parameter information to obtain a resource pool, including: Initialize cluster centers based on the target personnel; Based on the Euclidean distance formula Calculate the distance between the target person and each cluster center, and assign the target person to the cluster center with the smallest distance to the target person, where d(x i , C k ) represents the data point x i With cluster center C k The Euclidean distance between them, j = 1 means the starting index of the sum is 1, m means the number of job responsibilities, x ij Represents data point x i The score or attribute value of the jth job responsibility, C kj Represents the cluster center C k The j-th eigenvalue of ; Based on the cluster center update formula The minimum cluster center is updated until the updated minimum cluster center is less than a preset threshold, and the resource pool is obtained, wherein S k represents the set of all data points in the kth cluster, For all S k The data point x of the set i The sum of .

3. The GPT-based task allocation method according to claim 1, characterized in that: Preprocessing the attribute data includes: Formatting the attribute data, and performing deduplication and integrity check on the formatted attribute data; Based on the attribute data after integrity check, the skill label and experience value of the target person are determined and stored.

4. The GPT-based task allocation method according to claim 1, characterized in that: Also included is determining a relationship map of the target person, including: The target persons are taken as nodes and the target relationships as edges to construct a relationship graph of the target persons, wherein the weight of each edge represents the affinity and tacit understanding between the target persons.

5. The GPT-based task allocation method according to claim 4, characterized in that: Also includes: Obtaining a centrality score of the target person; based on Calculate the affinity and tacit understanding between the target persons, where Affinity(i,j) represents the affinity or tacit understanding between target person i and target person j. It represents half of the sum of the centrality scores of target person i and target person j, and InteractionWeight(i, j) represents the weight of the interaction relationship between target person i and target person j.

6. The GPT-based task allocation method according to claim 1, characterized in that: The attribute information of the task requirement includes a task level and a deadline, wherein allocating the task to the target person in the target team based on the attribute information of the task requirement includes: Determine the workload, skill tag and experience value of the target personnel in the target team; The task is assigned to the target person in the target team based on the task level and the deadline.

7. The GPT-based task allocation method according to claim 6, characterized in that: Allocating the task to the target person in the target team based on the task level and the deadline includes: When the first task level in the task requirement is level one and the deadline is the first deadline, assigning the task to the first target person in the target team; When the second task level in the task requirement is level two and the deadline is the second deadline, the task is assigned to the second target person in the target team, wherein the workload level of the first target person is higher than the workload level of the second target person, and the experience value of the first target person is higher than the experience value of the second target person.

8. A task allocation device based on GPT, characterized in that: include: A classification module, used to obtain target personnel and parameter information of the target personnel, and classify the target personnel based on the parameter information to obtain a resource pool; A data processing module, used to obtain the attribute data of the target person and pre-process the attribute data; A training module, used to train the GPT model using the preprocessed attribute data as a training set to obtain a target GPT model; An allocation module is used to generate a target team matching the task requirements based on the relationship map of the target personnel and the task requirements under the target GPT model, and to allocate the task to the target personnel in the target team based on the attribute information of the task requirements.

9. An electronic device, characterized in that: include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium storing computer instructions, characterized in that: The computer instructions are used to cause the computer to execute the method according to any one of claims 1 to 7.

Images