Virtual team management method based on enterprise information sharing

Abstract

The invention relates to the technical field of enterprise team time management, in particular to a virtual team management method based on enterprise information sharing, and the method comprises the steps: setting timestamps for key task cooperation links on an information sharing platform, and collecting various types of data; a dynamic reminding threshold model is constructed by using a fusion transfer learning and generative adversarial network method, meanwhile, social network factors of tasks are considered, model parameters are optimized by adopting a quantum annealing algorithm, and in a task reminding stage, task processing time is monitored in real time according to the model, and reminding frequency is dynamically adjusted based on cognitive calculation. A block chain and edge calculation are utilized to obtain busy states of members, time sequence prediction and multi-agent negotiation are utilized to delay reminding time, emotion calculation and a virtual reality technology are utilized to trigger reminding, interference to work of the members is avoided, acceptance and attention of the members to the reminding are improved, the cooperation efficiency of key tasks of a virtual team is remarkably improved, and the user experience is improved. And the virtual team management effect is optimized.

Description

Technical Field

[0001] The present invention relates to the technical field of enterprise team time management. Specifically, it relates to a virtual team management method based on enterprise information sharing. Background Art

[0002] Enterprise team time management is an important technology. In the digital business environment, the application prospect of enterprise team time management is extensive. It can help enterprises reasonably plan resources, improve work efficiency, thereby gaining an advantage in market competition, promoting the efficient operation of business in various industries, and achieving sustainable development.

[0003] When a virtual team is carrying out a software development project, due to the wide geographical distribution of members and different time zones, it is difficult to establish a unified work rhythm and communication norms, resulting in a lack of effective time management and information sharing mechanisms. When it comes to software development projects, when front-end developers complete page design and transfer tasks to back-end personnel, due to these information communication barriers, it is very difficult for back-end personnel to clearly know the urgency and time requirements of tasks, which in turn leads to task backlogs. Existing reminder mechanisms often rely on simple preset rules and do not fully consider the actual work situation of members, unable to accurately grasp the status of members. When back-end personnel are busy handling high-priority emergency bug repair tasks, the existing reminder mechanism still sends new task reminders according to the established rules, and members are prone to have a resistant attitude towards reminders. When the time arrangement or urgency of the task itself changes, the existing mechanism cannot adjust the reminder in time, resulting in untimely task reminders and missing the best handling opportunity, causing the project progress to lag and the team collaboration efficiency to be low. To solve this technical problem, we thus provide a virtual team management method based on enterprise information sharing. Summary of the Invention

[0004] The purpose of the present invention is to provide a virtual team management method based on enterprise information sharing to solve the problems raised in the above background art.

[0005] To achieve the above purpose, one of the purposes of the present invention is to provide a virtual team management method based on enterprise information sharing, including the following steps:

[0006] S1. On the information sharing platform, set timestamps for each collaboration link of key tasks, record the time of task initiation, transfer, and waiting for processing, and collect historical task data, member busy period data, and task priority data;

[0007] S2. Use machine learning algorithms to analyze the collected historical task data, member busy period data, and task priority data, establish a dynamic reminder threshold model for different types of key task collaboration links, train the dynamic reminder threshold model using historical data, and regularly evaluate the model and adjust the model parameters based on actual application feedback;

[0008] S3. According to the dynamic reminder threshold model, monitor the time when the recipient processes the task in real time. When the recipient's processing time reaches the dynamic threshold from the time when the task is sent by the task initiator, trigger a reminder. If the recipient does not respond in time, as the time reaches the upper limit of the normal processing time, dynamically adjust the reminder frequency according to the remaining time ratio and the urgency of the task. Obtain the busy status of members on the platform in real time. Combining the urgency of the task, when the member is in a busy state and the urgency of the task permits, postpone the reminder time. After the member's busy state is lifted, immediately trigger a reminder;

[0009] S4. According to the reminder strategy, send reminder information including task basic information, remaining processing time, and urgency to members through the information sharing platform, collect the feedback of members on the reminder information, and provide the feedback information for model training and optimization.

[0010] As a further improvement of this technical solution, in S2, a method combining transfer learning and generative adversarial network is used to construct a dynamic reminder threshold model, specifically as follows:

[0011] For historical task data, member busy period data, and task priority data, perform preprocessing using feature mapping and normalization techniques to convert them into a unified feature space;

[0012] Select a pre-trained base model in the same task management scenario as the source model, and adjust it using the collected data. At the same time, use a metric-based transfer learning method to calculate the feature distance between the source domain data and the target domain data, and adjust the model parameters. The source domain data is the data that has been used for pre-training the base model in the same task management scenario, and the target domain data is the data corresponding to the specific task that the transfer learning model currently adapts to;

[0013] Then introduce a generative adversarial network model, where the generator is responsible for generating synthetic data with the same distribution as the real data, and the discriminator is used to distinguish between real data and synthetic data, and perform adversarial training on the real data and the synthetic data;

[0014] Fuse the above transfer learning model and generative adversarial network model to obtain a dynamic reminder threshold model.

[0015] As a further improvement of this technical solution, when establishing the dynamic reminder threshold model in S2, consider the social network factors of the task, specifically as follows:

[0016] Analyze the communication frequency, cooperation history, and information sharing patterns among members to construct the social network of members. Use graph theory methods to abstract members as nodes, the relationships between members as edges, and assign weights to the edges to represent the closeness of the relationships.

[0017] Combine the collaboration links of the task and the social network of members to analyze the propagation path and diffusion speed of the task in the network. Through social network analysis algorithms, determine the key members and key propagation paths.

[0018] Convert the centrality of members in the social network, the delay of task propagation, and information loss into features and incorporate them into the dynamic reminder threshold model. As the relationships between members change and the task progresses, update the social network structure in real time, and dynamically adjust the parameters of the dynamic reminder threshold model according to the changes in the network structure.

[0019] As a further improvement of this technical solution, when regularly evaluating the model and adjusting the model parameters according to the actual application feedback in S2, an optimization strategy based on the quantum annealing algorithm is adopted, specifically as follows:

[0020] Define the task delay risk assessment indicators and construct an energy function that includes the performance evaluation indicators and parameter constraint conditions of the model. Convert the problem of adjusting the model parameters into a quantum annealing problem, simulate the optimization process of the model parameters through the interaction of qubits, and use the quantum annealing algorithm to find the global minimum of the energy function, that is, the optimal parameter combination of the model.

[0021] During the quantum annealing process, by controlling the temperature and interaction strength of the qubits, guide the algorithm to converge to the optimal solution, and update the parameters of the energy function in real time according to the feedback of members on reminder information and the task processing results.

[0022] As a further improvement of this technical solution, when dynamically adjusting the reminder frequency according to the remaining time ratio and task urgency in S3, an adaptive reminder strategy based on cognitive computing is adopted, specifically as follows:

[0023] Construct a cognitive model of members, which includes the attention allocation mechanism and information processing speed of members. Combine the complexity, urgency, and remaining time ratio of the task to evaluate the cognitive load of the task on members and the cognitive resources required for computing the task. The cognitive load represents the amount of load borne by members when performing the task, and the cognitive resources represent the processing ability of members when processing this task.

[0024] Dynamically adjust the reminder frequency according to the cognitive model of members and the cognitive load of the task, collect the feedback information of members on the reminder to update the cognitive model of members, and then dynamically adjust the reminder strategy according to the update of the cognitive model.

[0025] As a further improvement of this technical solution, when obtaining the busy status of members on the platform in S3, a distributed monitoring method based on blockchain and edge computing is adopted, which is specifically as follows:

[0026] Store the operation data of members on the platform on the blockchain. At the same time, assign a unique digital identity to each member, and deploy edge computing nodes on the edge devices of the information sharing platform to collect and process the operation data of members in real time. The edge computing nodes are used to analyze and process the data and extract key features;

[0027] Through the blockchain network, data sharing and collaborative computing are carried out among edge computing nodes, and a busy status monitoring model is jointly trained using distributed machine learning algorithms. The busy status monitoring model is used to monitor the busy status of members in real time and feedback the monitoring results to the reminder system;

[0028] During the distributed monitoring process, an anomaly detection algorithm is used to detect abnormal behaviors in the operation data of members in real time. When abnormal behaviors are found, early warnings are immediately issued.

[0029] As a further improvement of this technical solution, when a member is in a busy state and the task urgency permits in S3, a method based on time series prediction and multi-agent negotiation is adopted to postpone the reminder time, which is specifically as follows:

[0030] Use a deep learning model to model and predict the historical working time series data of members, predict the busy status and idle time of members in the future period, and build agents for each task and member. The task agent is responsible for managing the progress of the task and the reminder strategy, and the member agent is responsible for maintaining the status information of the member;

[0031] When a member is in a busy state and the task urgency permits, negotiations are carried out between the task agent and the member agent. During the negotiation process, the task agent proposes a time suggestion for postponing the reminder according to the remaining time ratio and urgency of the task, and the member agent evaluates and feedbacks the suggestion according to its predicted busy status and work arrangement. Through multiple negotiations, a postponed reminder time acceptable to both parties is reached;

[0032] During the execution of the task, the working time series data of the member and the status information of the task are updated in real time. When the busy status of the member or the urgency of the task changes, the agents re-negotiate and dynamically adjust the postponed reminder time.

[0033] As a further improvement of this technical solution, when triggering a reminder in S3, a reminder strategy based on affective computing and virtual reality technology is adopted, which is specifically as follows:

[0034] By analyzing the text interaction information, voice information, and facial expression information of members on the platform, using emotion computing technology to identify the emotional states of members, and creating virtual reality reminder scenarios according to the emotional states of members and the urgency of tasks;

[0035] And through virtual reality devices, bring members into the created virtual reality reminder scenarios, display the basic information, remaining processing time, and urgency of tasks in the scenarios, collect the emotional feedback information of members in the virtual reality reminder scenarios, and optimize the virtual reality reminder scenarios according to the feedback information.

[0036] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0037] In the virtual team management method based on enterprise information sharing, a dynamic reminder threshold model is constructed by integrating transfer learning and generative adversarial networks, using the knowledge of pre-trained models and expanding data, improving the model performance and generalization ability, considering task social network factors and adjusting in real time to make the reminder threshold more in line with actual needs, using quantum annealing algorithm to optimize model parameters, enhancing model performance and stability, an adaptive reminder strategy based on cognitive computing, adjusting the reminder frequency according to members' cognition and task load, avoiding interference and improving effectiveness, using blockchain and edge computing for distributed monitoring of busy states, ensuring data security in real time and being able to detect anomalies, using time series prediction and multi-agent negotiation to postpone the reminder time, making the reminder more practical, triggering reminders by combining emotion computing and virtual reality technology, improving members' acceptance and attention, and enhancing the collaboration efficiency of key tasks in virtual teams. Brief Description of the Drawings

[0038] Figure 1 It is the overall work flow chart of the present invention. Detailed Embodiments

[0039] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0040] Please refer to Figure 1 As shown, this embodiment provides a virtual team management method based on enterprise information sharing, including the following steps:

[0041] S1. On the information sharing platform, set time stamps for each collaboration link of the key task, record the time of task initiation, transfer, and waiting for processing, and collect historical task data, member busy period data, and task priority data.

[0042] S2. Analyze the collected historical task data, member busy period data, and task priority data using machine learning algorithms, establish a dynamic reminder threshold model for different types of critical task collaboration links, train the dynamic reminder threshold model using historical data, and regularly evaluate the model and adjust the model parameters based on actual application feedback.

[0043] In S2, a method that combines transfer learning and generative adversarial networks is used to construct the dynamic reminder threshold model, which is specifically as follows:

[0044] The historical task data, member busy period data, and task priority data have different feature representations and data distributions, which will affect the subsequent training effect of the model. For the historical task data, member busy period data, and task priority data, feature mapping and normalization techniques are used for preprocessing to convert them into a unified feature space. The preprocessed data has a unified feature space and similar data distributions, which can accelerate the convergence speed of the model, improve the stability and generalization ability of the model.

[0045] The base model pre-trained in the same task management scenario has learned some general features and patterns. Select a base model pre-trained in the same task management scenario as the source model and adjust it using the collected data. At the same time, use the metric-based transfer learning method to adjust the model parameters by calculating the feature distances between the source domain data and the target domain data. The source domain data is the data that has been used for pre-training the base model in the same task management scenario, and the target domain data is the data corresponding to the specific task that the transfer learning model is currently adapted to.

[0046] In practical applications, the data in the target domain may have problems such as insufficient samples or unbalanced distributions, which will affect the generalization ability of the model. Then introduce the generative adversarial network model, where the generator is responsible for generating synthetic data with the same distribution as the real data, and the discriminator is used to distinguish between real data and synthetic data and conduct adversarial training on the real data and synthetic data. The generative adversarial network model can generate high-quality synthetic data, expand the scale and diversity of the training data, alleviate the problems of data insufficiency and unbalanced distributions, and improve the generalization ability of the model.

[0047] Fuse the above transfer learning model and generative adversarial network model to obtain the dynamic reminder threshold model. The model fusion combines the advantages of transfer learning and generative adversarial networks, utilizes the knowledge of the pre-trained model, expands the training data, and improves the performance and generalization ability of the model.

[0048] When establishing the dynamic reminder threshold model in S2, consider the social network factors of the tasks, which are specifically as follows:

[0049] In a virtual team, social relationships such as the communication frequency, cooperation history, and information sharing patterns among members will have an important impact on the execution and progress of tasks. By analyzing the communication frequency, cooperation history, and information sharing patterns among members, constructing the social network of members, using graph theory methods to abstract members as nodes, the relationships among members as edges, and assigning weights to the edges to represent the degree of closeness of the relationships, through constructing the social network, we can better understand the interaction patterns among members and provide an intuitive basis for analyzing task propagation and determining key members.

[0050] The propagation path and diffusion speed of tasks in the team will affect the execution efficiency and completion time of tasks. Combining the collaboration links of tasks and the social network of members, analyzing the propagation path and diffusion speed of tasks in the network, through social network analysis algorithms, determining key members and key propagation paths, using the shortest path algorithm to find the shortest propagation path of tasks from the initiating node to the receiving node, and setting the weight ω of the edge ij represents the time for the task to propagate between members v i to v j . For a given starting node v s and target node v t , the shortest path algorithm finds the shortest path by continuously updating the shortest distances of nodes.

[0051] Calculate the diffusion speed according to the propagation time of tasks between different nodes. Let the time for the task to propagate from node v i to node v j be t ij , then the diffusion speed on the edge (v i , v j ) Use the centrality analysis algorithm to determine key members. Degree centrality where a ij is an element of the adjacency matrix, indicating whether there is an edge connection between nodes v i and v j . Betweenness centrality where σ st is the number of shortest paths from node s to t, and σ st (v i ) is the number of shortest paths passing through node v i . The key propagation path can be determined by analyzing the edges whose betweenness centrality values exceed the preset threshold. Through social network analysis algorithms, the key paths and key members of task propagation can be accurately found, providing targeted strategies for task management.

[0052] The centrality of members in the social network, the latency of task dissemination, and information loss affect the processing time and efficiency of tasks. These factors are transformed into features and incorporated into the dynamic reminder threshold model. As the relationships between members change and tasks progress, the social network structure is updated in real time. Based on the changes in the network structure, the parameters of the dynamic reminder threshold model are adjusted dynamically. Over time, the relationships between members may change, and the edge weights of the social network need to be updated in real time. Based on the updated social network, the centrality, dissemination latency, and information loss are recalculated, and the gradient descent optimization algorithm is used to adjust the parameters of the dynamic reminder threshold model to minimize the loss function where T k is the reminder threshold predicted by the model, is the true reminder threshold, N is the number of samples, θ is the parameter of the model. The reminder threshold is more in line with the actual requirements of the task, can timely and effectively remind members to process tasks, and improve the completion efficiency and quality of tasks.

[0053] When regularly evaluating the model and adjusting the model parameters according to the actual application feedback in S2, an optimization strategy based on the quantum annealing algorithm is adopted, as follows:

[0054] In the optimization process of the model, clear evaluation indicators are needed to measure the performance of the model, and at the same time, the constraint conditions of the parameters are considered to ensure that the obtained parameter combination is reasonable and effective. Define the task delay risk evaluation indicator R and construct an energy function E(θ), which includes the performance evaluation indicators of the model and the parameter constraint conditions. Transform the parameter adjustment problem of the model into a quantum annealing problem. In the quantum annealing process, by controlling the temperature and interaction strength of the qubits, the algorithm is guided to converge to the optimal solution.

[0055] The core of the quantum annealing algorithm is to simulate the evolution of the quantum system. Let the Hamiltonian of the quantum system be H = H0 + s(t)H p , where H0 is the initial Hamiltonian, H p is the problem Hamiltonian, s(t) is the annealing parameter, and s(0) = 0, s(T) = 1, T is the annealing time. Initially, the quantum system is in the ground state of H0. As time t increases, it gradually evolves to the ground state of H p , and this ground state corresponds to the minimum value of the energy function E(θ).

[0056] The temperature T q of the qubit and the interaction strength J ij will affect the evolution of the system. During the annealing process, the temperature T q gradually decreases, and the interaction strength J ijGradually adjust so that the system gradually converges from a disordered state to an ordered ground state, which can find the global minimum of the energy function, that is, the optimal parameter combination of the model, more quickly, thereby improving the performance and stability of the model.

[0057] According to the feedback of members on reminder information and the task processing results, update the parameters of the energy function in real time to improve the effect of task management and reduce the risk of task delay.

[0058] S3. According to the dynamic reminder threshold model, monitor the time for the recipient to process tasks in real time. When the time for the recipient to process tasks reaches the dynamic threshold from the time when the task originator transmits the tasks, trigger a reminder. If the recipient does not respond in time, as the time reaches the upper limit of the normal processing time, dynamically adjust the reminder frequency according to the remaining time ratio and the urgency of the task, and obtain the busy status of members on the platform in real time. Combining the urgency of the task, when the member is in a busy state and the urgency of the task permits, postpone the reminder time, and immediately trigger a reminder after the member's busy state is lifted.

[0059] When dynamically adjusting the reminder frequency according to the remaining time ratio and the urgency of the task in S3, adopt an adaptive reminder strategy based on cognitive computing, which is specifically as follows:

[0060] Different members have different attention allocation mechanisms and information processing speeds. These individual differences will affect their task processing capabilities and responses to reminders. Construct a cognitive model of members, and combine the complexity, urgency, and remaining time ratio of tasks to evaluate the cognitive load of tasks on members and the cognitive resources required for computing tasks.

[0061] Let the attention allocation weight of member i on different types of tasks j be ω ij , and define the information processing speed v i of member i as the number of information units processed per unit time. Let the complexity of task j be C j , the urgency be U j , the remaining time ratio be R j , and the cognitive load L ij of task j on member i can be calculated by the following formula:

[0062] L ij = aC j + bU j + (1 - a - b)(1 - R j ), where a and b are weight coefficients, and the cognitive resources R ij required by member i to process task j = L ij / v i . Accurately evaluate the cognitive load of tasks on members and the required cognitive resources, providing a scientific basis for subsequent adjustment of reminder strategies.

[0063] Dynamically adjusting the reminder frequency according to the cognitive model of the member and the cognitive load of the task can ensure that the reminder neither causes excessive interference to the member nor fails to timely remind the member to handle the task when necessary. Let the initial reminder frequency be f0, and according to the cognitive load L ij Adjust the reminder frequency f ij , and use a piecewise function for adjustment, that is where L low and L high are the thresholds of the cognitive load, k1 > 1, k2 < 1. Dynamically adjusting the reminder frequency can flexibly provide reminder services according to the cognitive state of the member and the task requirements, improving the effectiveness and pertinence of the reminder, and improving the efficiency of task processing.

[0064] The feedback information of the member on the reminder reflects the effectiveness of the current reminder strategy and the change of the member's cognitive state. Collect the feedback satisfaction of the member on the reminder, update the attention allocation weight according to the feedback satisfaction, and then update the information processing speed according to the new time for the member to complete the task, so as to update the cognitive model of the member. According to the updated cognitive model, readjust L low 、L high 、k1 and k2. Over time, the reminder strategy will increasingly conform to the cognitive characteristics of the member and the task requirements, further improving the efficiency and quality of task management.

[0065] When obtaining the busy status of the member on the platform in S3, a distributed monitoring method based on blockchain and edge computing is adopted, specifically as follows:

[0066] Store the operation data of the member on the platform on the blockchain. At the same time, assign a unique digital identity to each member, deploy edge computing nodes on the edge devices of the information sharing platform, and collect and process the operation data of the member in real time. The edge computing nodes are used to analyze and process the data, and extract the time interval of the operation as the key feature. The blockchain ensures the security and traceability of the data, the digital identity facilitates member management and data attribution confirmation, and edge computing reduces data transmission latency and cloud computing pressure, improving the real-time performance and efficiency of data processing.

[0067] The data collected by each edge computing node is local. Through the blockchain network for data sharing and collaborative computing, the data of each node can be integrated to obtain more comprehensive member operation information. Through the blockchain network, data sharing and collaborative computing are carried out between each edge computing node, and a busy status monitoring model is jointly trained using a distributed machine learning algorithm. The busy status monitoring model is used to monitor the busy status of the member in real time and feedback the monitoring result to the reminder system.

[0068] The edge computing node k will extract the feature data f kPackaged as a transaction T k , and broadcast to the blockchain network. Other nodes obtain data by verifying the legitimacy of the transaction. Let the local model parameters on the edge computing node k be θ k , and the global model parameters be Θ. Each node updates its local model according to the local data, where α is the learning rate, is the gradient of the local loss function L(θ k , f k ). Then, each node uploads the updated local model parameters to the blockchain network and updates the global model parameters through weighted averaging, where m is the number of nodes, ω k is the weight of node k. The trained busy state monitoring model MΘ predicts the real-time input feature data f, outputs the busy state S = M(Θ, f) of the member, and feeds the result back to the reminder system, providing a reliable basis for the reminder system and improving the pertinence and effectiveness of the reminder.

[0069] During the distributed monitoring process, an anomaly detection algorithm is used to detect abnormal behaviors in the member operation data in real time. When an abnormal behavior is found, an early warning is immediately issued, reducing task delays and losses caused by abnormal behaviors and improving the reliability of virtual team management.

[0070] In S3, when the member is in a busy state and the task urgency permits, the reminder time is postponed using a method based on time series prediction and multi-agent negotiation, as follows:

[0071] The work of members has certain regularity and periodicity. A deep learning model is used to model and predict the historical work time series data of members, predicting the busy state and idle time of members in the future for a period of time, providing a basis for subsequent reminder time adjustment. Let the historical work time series data of members be X = {x1, x2,..., x n}, and a long short-term memory network is used for modeling. By training the long short-term memory network model, the work state Y in the next T time steps is predicted, and the expression is Y = {y n+1 , y n+2 ,..., y n+T}. The task agent A ta maintains the progress P and reminder strategy S of the task re , and the member agent A me maintains the status information S of the member me , including the predicted busy state and work arrangement, providing a reliable basis for the adjustment of the reminder time. At the same time, the modular design of the agent improves the scalability and flexibility of the system.

[0072] When the member is busy and the urgency of the task permits, negotiation takes place between the task agent and the member agent. During the negotiation process, the task agent proposes a time suggestion for postponing the reminder based on the remaining time ratio and urgency of the task. The member agent evaluates and gives feedback on the suggestion according to its predicted busy state and work arrangement. Through multiple rounds of negotiation, a mutually acceptable postponed reminder time is reached.

[0073] Let the remaining time ratio of the task be r, the urgency be e, and the time suggestion t for postponing the reminder proposed by the task agent su can be calculated through the function t su = f(r, e), where t su = k1(1 - r) + k2e, and k1 and k2 are weight coefficients. The member agent calculates the probability p = g(S bu and work arrangement S sc , S bu , t sc ) of accepting the suggestion. If p is greater than a certain threshold θ, the suggestion is accepted; otherwise, a new postponed reminder time t su is given as feedback. Through multiple iterative negotiations, until a mutually agreed postponed reminder time t new is reached, it avoids member interference and task delays caused by unreasonable reminder times, makes the reminder more in line with the actual work situation, and improves the collaboration efficiency of the virtual team. final

[0074] During the task execution process, the working time series data of the member and the status information of the task are updated in real time. When the busy state of the member or the urgency of the task changes, the agents negotiate again and dynamically adjust the postponed reminder time.

[0075] New work data of the member is obtained in real time and added to the historical time series data. The updated time series is X = {x1, x2,..., x n new , x new}, and the long short-term memory network model is retrained for prediction. At the same time, the progress and urgency of the task are updated. When the change in the busy state of the member or the urgency of the task exceeds a certain threshold, the task agent and the member agent negotiate again. According to the new remaining time ratio, urgency, predicted busy state, and work arrangement, the above negotiation process is repeated to obtain a new postponed reminder time, ensuring that during the task execution process, the reminder time always conforms to the actual situation, reducing task management mistakes caused by state changes, and improving the work efficiency and task completion quality of the virtual team.

[0076] When a member is in a busy state and the task urgency permits in S3, the reminder time is postponed using a method based on time series prediction and multi-agent negotiation, as follows:

[0077] The emotional state of a member affects their acceptance of reminders and the efficiency of task processing. By analyzing the text interaction information, voice information, and facial expression information of the member on the platform, emotion computing technology is used to identify the emotional state of the member. Based on the emotional state of the member and the urgency of the task, a virtual reality reminder scenario is created.

[0078] For text interaction information, a pre-trained sentiment analysis model is used to output the probability distribution of sentiment categories, representing the probabilities of positive, negative, and neutral sentiments respectively. For voice information, linguistic features are extracted and a support vector machine is used for sentiment classification to obtain the sentiment category. For facial expression information, a convolutional neural network is used for emotion recognition, and the output of the convolutional neural network is the probability distribution of the emotion category.

[0079] Taking into account the analysis results of text, voice, and facial expressions, the final emotional state is obtained through weighted averaging. Based on the emotional state and urgency, a virtual reality reminder scenario is selected from a preset scenario library, which can more accurately grasp the emotional state of the member, provide a more suitable reminder scenario for the member, improve the member's acceptance and attention of the reminder, and further improve the effectiveness of the reminder and the member's work efficiency.

[0080] In S4, according to the reminder strategy, reminder information including the basic information of the task, the remaining processing time, and the urgency is sent to the member through the information sharing platform, the feedback of the member on the reminder information is collected, and the feedback information is provided for model training and optimization.

[0081] The above virtual team management method based on enterprise information sharing sets timestamps for key task collaboration links on the information sharing platform and collects various types of data, constructs a dynamic reminder threshold model using a method that combines transfer learning and generative adversarial networks, takes into account the social network factors of tasks, optimizes model parameters using the quantum annealing algorithm. In the task reminder stage, the model is used to monitor the task processing time in real time, dynamically adjust the reminder frequency based on cognitive computing, obtain the busy state of the member using blockchain and edge computing, postpone the reminder time using time series prediction and multi-agent negotiation, and also trigger reminders using emotion computing and virtual reality technology to avoid interfering with the member's work, improve the member's acceptance and attention of the reminder, thereby significantly improving the collaboration efficiency of key tasks in the virtual team and optimizing the virtual team management effect.

[0082] The foregoing has shown and described the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments, and the above embodiments and the descriptions in the specification are only preferred examples of the present invention, which are not used to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of the present invention claimed is defined by the appended claims and their equivalents.

Claims

1. A virtual team management method based on enterprise information sharing, characterized in that: It includes the following steps: S1. On the information sharing platform, set timestamps for each collaboration link of the critical task, record the time of task initiation, transmission, and waiting for processing, and collect historical task data, member busy period data, and task priority data; S2. Use machine learning algorithms to analyze the collected historical task data, member busy period data, and task priority data, establish a dynamic reminder threshold model for different types of critical task collaboration links, train the dynamic reminder threshold model with historical data, and regularly evaluate the model and adjust model parameters according to actual application feedback; S3. According to the dynamic reminder threshold model, monitor the time for the recipient to process the task in real time. When the recipient's processing time reaches the dynamic threshold from the time the task initiator transmits the task, trigger a reminder. If the recipient does not respond in time, as the time reaches the upper limit of the normal processing time, dynamically adjust the reminder frequency according to the remaining time ratio and the urgency of the task, and obtain the busy status of the member on the platform in real time. Combining the urgency of the task, when the member is in a busy state and the urgency of the task permits, postpone the reminder time, and immediately trigger a reminder after the member's busy state is lifted; S4. According to the reminder strategy, send reminder messages containing task basic information, remaining processing time, and urgency to members through the information sharing platform, collect the feedback of members on the reminder messages, and provide the feedback information for model training and optimization.

2. The virtual team management method based on enterprise information sharing according to claim 1, wherein: In S2, the method of fusing transfer learning and generative adversarial network is used to construct the dynamic reminder threshold model, which is specifically as follows: For historical task data, member busy period data, and task priority data, use feature mapping and normalization techniques for preprocessing to convert them into a unified feature space; Select a pre-trained base model in the same task management scenario as the source model, and adjust it using the collected data. At the same time, use a metric-based transfer learning method to calculate the feature distance between the source domain data and the target domain data, and adjust the model parameters. The source domain data is the data that has been used for pre-training the base model in the same task management scenario, and the target domain data is the data corresponding to the specific task that the transfer learning model currently adapts to; Then introduce a generative adversarial network model, where the generator is responsible for generating synthetic data with the same distribution as the real data, and the discriminator is used to distinguish between real data and synthetic data, and perform adversarial training on the real data and the synthetic data; Fuse the above transfer learning model and generative adversarial network model to obtain the dynamic reminder threshold model.

3. The virtual team management method based on enterprise information sharing according to claim 2, characterized in that, When establishing the dynamic reminder threshold model in S2, consider the social network factors of the task, which are specifically as follows: Analyze the communication frequency, cooperation history, and information sharing mode among members, construct the social network of members, abstract members as nodes using graph theory methods, abstract the relationship between members as edges, and assign weights to the edges to represent the closeness of the relationship; Combining the collaboration links of the task and the member social network, analyze the propagation path and diffusion speed of the task in the network, and determine the key members and key propagation paths through social network analysis algorithms; Convert the centrality of members in the social network, the latency of task propagation, and information loss into features, incorporate them into the dynamic reminder threshold model, update the social network structure in real time as the relationships between members change and the task progresses, and dynamically adjust the parameters of the dynamic reminder threshold model according to the changes in the network structure.

4. The virtual team management method based on enterprise information sharing according to claim 3, characterized in that: When regularly evaluating the model and adjusting the model parameters according to the actual application feedback in S2, an optimization strategy based on the quantum annealing algorithm is adopted, which is specifically as follows: Define the task delay risk assessment index and construct an energy function, which includes the performance evaluation index of the model and parameter constraints. Transform the problem of adjusting the model parameters into a quantum annealing problem, simulate the optimization process of the model parameters through the interaction of qubits, and use the quantum annealing algorithm to find the global minimum of the energy function, that is, the optimal parameter combination of the model; During the quantum annealing process, by controlling the temperature and interaction strength of the qubits, guide the algorithm to converge to the optimal solution, and update the parameters of the energy function in real time according to the feedback of members on reminder information and the task processing results.

5. The virtual team management method based on enterprise information sharing according to claim 4, wherein: When dynamically adjusting the reminder frequency according to the remaining time ratio and task urgency in S3, an adaptive reminder strategy based on cognitive computing is adopted, which is specifically as follows: Construct a cognitive model of members, which includes the attention allocation mechanism and information processing speed of members. Combine the complexity, urgency, and remaining time ratio of the task to evaluate the cognitive load of the task on members and the cognitive resources required for computing the task. The cognitive load represents the amount of load borne by members when performing the task, and the cognitive resources represent the processing ability of members when processing this task; Dynamically adjust the reminder frequency according to the cognitive model of members and the cognitive load of the task, collect the feedback information of members on the reminder to update the cognitive model of members, and then dynamically adjust the reminder strategy according to the update of the cognitive model.

6. The virtual team management method based on enterprise information sharing according to claim 5, characterized in that: When obtaining the busy status of members on the platform in real time in S3, a distributed monitoring method based on blockchain and edge computing is adopted, which is specifically as follows: Store the operation data of members on the platform on the blockchain. At the same time, assign a unique digital identity to each member, and deploy edge computing nodes on the edge devices of the information sharing platform to collect and process the operation data of members in real time. The edge computing nodes are used to analyze and process the data and extract key features; Through the blockchain network, data sharing and collaborative computing are carried out between edge computing nodes, and a busy status monitoring model is jointly trained using distributed machine learning algorithms. The busy status monitoring model is used to monitor the busy status of members in real time and feedback the monitoring results to the reminder system; During the distributed monitoring process, use anomaly detection algorithms to detect abnormal behaviors in the operation data of members in real time, and issue early warnings immediately when abnormal behaviors are found.

7. The virtual team management method based on enterprise information sharing according to claim 6, characterized in that: When members are in a busy state and the task urgency permits, postpone the reminder time using a method based on time series prediction and multi-agent negotiation, which is specifically as follows: Use a deep learning model to model and predict the historical working time series data of members, predict the busy status and idle time of members in the future for a period of time, and build agents for each task and member. The task agent is responsible for managing the progress and reminder strategy of the task, and the member agent is responsible for maintaining the status information of the member; When the member is in a busy state and the urgency of the task permits, negotiation is carried out between the task agent and the member agent. During the negotiation process, the task agent proposes a time suggestion for postponing the reminder based on the remaining time ratio and urgency of the task, and the member agent evaluates and gives feedback on the suggestion according to its predicted busy status and work arrangement. Through multiple negotiations, a postponed reminder time acceptable to both parties is reached; During the execution of the task, the working time series data of the member and the status information of the task are updated in real time. When the busy status of the member or the urgency of the task changes, the agents negotiate again to dynamically adjust the postponed reminder time.

8. The virtual team management method based on enterprise information sharing according to claim 7, characterized in that: In S3, when triggering a reminder, a reminder strategy based on emotion computing and virtual reality technology is adopted, specifically as follows: By analyzing the text interaction information, voice information and facial expression information of the member on the platform, emotion computing technology is used to identify the emotional state of the member, and a virtual reality reminder scenario is created according to the emotional state of the member and the urgency of the task; And through virtual reality devices, the member is brought into the created virtual reality reminder scenario, where the basic information, remaining processing time and urgency of the task are displayed. The emotional feedback information of the member in the virtual reality reminder scenario is collected, and the virtual reality reminder scenario is optimized according to the feedback information.

Images