Virtual workplace task automatic generation and distribution method and system based on knowledge graph

Abstract

The invention provides a virtual workplace task automatic generation and distribution method and system based on a knowledge graph, and relates to the technical field of virtual workplaces, and the method comprises the steps: carrying out the physical preprocessing of a battery material, processing the technological parameters of a leaching process through wavelet transformation and a long-short term memory network, determining an optimal leaching technological parameter combination, and carrying out the automatic generation and distribution of a virtual workplace task. According to the method, a self-adaptive weighted average algorithm and a time sequence feature extraction algorithm are adopted to analyze purification process data, optimal purification process parameters are determined in combination with a reinforcement learning algorithm, solid-liquid separation is performed to obtain the purified power battery material, efficient and intelligent purification of the power battery material is realized, and the resource recovery rate and the purification quality are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of virtual workplace, and in particular to a virtual workplace task automatic generation and distribution method and system based on a knowledge graph. BACKGROUND

[0002] With the deepening of digital transformation, virtual workplace as a new work mode is gradually being widely applied. Task management in virtual workplace involves key links such as task generation, decomposition and distribution, which is of great significance to improve work efficiency and collaboration quality. Traditional virtual workplace task management mainly relies on manual experience for planning and distribution, which is difficult to cope with complex and variable work scenarios.

[0003] The prior art lacks systematic modeling of complex dependency relationships between tasks, and cannot accurately express the preconditions, resource competition and timing constraints between tasks, resulting in frequent conflicts in the actual execution of the generated task decomposition scheme. Secondly, the traditional task allocation method usually adopts static rule matching or simple ability scoring mechanism, and does not fully consider the historical performance, professional skill development track and team collaboration characteristics of the executors, resulting in mismatch between task allocation and actual ability, and reducing the overall work efficiency. SUMMARY

[0004] The embodiment of the present application provides a virtual workplace task automatic generation and distribution method and system based on a knowledge graph, which can solve the problems in the prior art.

[0005] The first aspect of the embodiment of the present application provides a virtual workplace task automatic generation and distribution method based on a knowledge graph, comprising: Obtaining task information in a virtual workplace, constructing a multi-level workplace knowledge graph including a task layer and a role layer, wherein the task layer contains task type nodes and inter-task pre-dependence relationships, and the role layer contains execution role nodes and role collaboration relationships; Converting the task information into a word vector based on the multi-level workplace knowledge graph, constructing a k-neighbor graph structure based on the word vector, performing local information diffusion on the word vector through random walk, generating task semantic features; recursively decomposing the task semantic features, decomposing the overall task target into multiple sub-targets, combining the sub-targets based on the dependency relationships in the multi-level workplace knowledge graph, decomposing the time complexity and resource dependency relationships of each sub-task in the decomposition tree, and generating a task decomposition graph; The historical task data of the candidate performer is acquired from the virtual workplace task execution system, the capability profile of the candidate performer is constructed, the candidate performer is taken as an execution unit with autonomous decision-making capability according to the capability profile, a task allocation scheme is generated through task negotiation and bidding between execution units in combination with the dependency relationship constraint of the task decomposition graph, and the allocation scheme is dynamically adjusted according to the execution condition in the task execution process.

[0006] The multi-level workplace knowledge graph including a task layer and a role layer is constructed, including: The task layer includes a task type node set and a task dependency relationship edge set, the role layer includes an execution role node set and a role collaboration relationship edge set, each edge in the task dependency relationship edge set connects the associated task type node, and each edge in the role collaboration relationship edge set connects the execution role nodes with a collaboration relationship; The collaboration strength weight of each edge in the role collaboration relationship edge set is calculated based on role collaboration historical data, the collaboration strength weight is calculated by weighting the normalized value of the collaboration times and the normalized value of the collaboration effect score, and the collaboration effect score is evaluated based on the completion quality of the historical collaboration task; The task attribute features and the role attribute features are respectively vector transformed, the correlation degree between the transformed feature vectors is calculated to obtain a node pair correlation degree, and the node pair correlation degree represents the association degree between the task type node and the execution role node.

[0007] The task information is converted into a word vector based on the multi-level workplace knowledge graph, a k-neighbor graph structure is constructed based on the word vector, local information diffusion is performed on the word vector in a random walk manner, and the task semantic features are generated, including: The word sequence in the task description text of the task information is converted into a word vector; The semantic mutual information between the word vectors is calculated, the semantic mutual information is calculated by feature transformation and interaction matrix of the word vector, a k-neighbor graph structure is constructed based on the semantic mutual information, the connection relationship between each word vector and its k most similar nodes is reserved, and a decay weight is assigned to the reserved connection relationship, the decay weight is obtained by distance weighting of the semantic mutual information through a Gaussian kernel function; Multi-hop feature propagation is performed based on the decay weight of the connection relationship, local information diffusion is performed on the word vector in a random walk manner, and task semantic features are generated.

[0008] The sub-targets are combined based on the dependency relationship in the multi-level workplace knowledge graph, the time complexity and resource dependency relationship of each sub-task in the decomposition tree are decomposed, and a task decomposition graph is generated, including: According to the dependency relationship, correlation analysis is performed on the sub-targets, sub-targets with a correlation degree higher than a preset correlation degree threshold are combined to form a plurality of sub-target combination sets, an execution cost of each sub-target in the sub-target combination set is calculated according to a predicted execution time length and resource occupation amount of the sub-target, and a sub-target combination with the lowest execution cost and meeting a preset dependency strength requirement is selected as an optimal combination scheme; The input size and single operation time consumption of the sub-target in the optimal combination scheme are obtained, a fixed overhead time in the task execution process is determined, and a time complexity of each sub-target is calculated based on the input size, the single operation time consumption and the fixed overhead time; A dependency relationship matrix of tasks and resources is constructed based on the time complexity, the dependency relationship matrix records a dependency degree of each task on various resources, and a task decomposition graph is generated according to the dependency relationship matrix.

[0009] According to the capability profile, a candidate executor is taken as an execution unit with autonomous decision-making capability, a task allocation scheme is generated through task negotiation and bidding between execution units and in combination with a dependency relationship constraint of the task decomposition graph, including: A state vector of the execution unit is constructed, the state vector includes an image vector corresponding to the capability profile and a load level indicator, and an utility value of the execution unit to the task is calculated based on a matching degree of the image vector and a task demand and the load level indicator; A bidding function of the execution unit is constructed based on the utility value, the bidding function is calculated by a product of the utility value and an exponential decay term of an execution cost, and the execution cost includes a weighted combination of a time delay cost and a load transfer cost; The current load state and the remaining resource amount of the execution unit are normalized and then weighted and summed to obtain a resource consumption score, a task allocation probability is calculated by matching the resource consumption score with a bid value output by the bidding function, and a task allocation scheme is generated in combination with a dependency relationship constraint of the task decomposition graph.

[0010] The resource consumption score and the bid value output by the bidding function are matched to obtain a task allocation probability, and a task allocation scheme is generated in combination with a dependency relationship constraint of the task decomposition graph, including: When a plurality of tasks with a dependency relationship are allocated to different execution units, the task allocation probability is punished according to a communication overhead between the execution units, and the task allocation probability is reduced; when a plurality of tasks with a dependency relationship are allocated to the same execution unit, the task allocation probability is rewarded according to a parallelism degree of the tasks, and the task allocation probability is improved; A task allocation scheme is generated according to the adjusted task allocation probability, the task allocation scheme determines a target execution unit of each task and records an execution order and resource allocation information of the task.

[0011] In a second aspect of the embodiments of the present application, a virtual workplace task automatic generation and distribution system based on a knowledge graph is provided, comprising: A first unit is configured to obtain task information in a virtual workplace, and construct a multi-level workplace knowledge graph comprising a task layer and a role layer, wherein the task layer comprises task type nodes and inter-task precedence dependency relationships, and the role layer comprises execution role nodes and role collaboration relationships. A second unit is configured to convert the task information into word vectors based on the multi-level workplace knowledge graph, construct a k-nearest neighbor graph structure based on the word vectors, perform local information diffusion on the word vectors in a random walk manner to generate task semantic features, recursively decompose the task semantic features to decompose an overall task target into multiple sub-targets, combine the sub-targets based on the dependency relationships in the multi-level workplace knowledge graph, decompose the time complexity and resource dependency relationships of each sub-task in the decomposition tree, and generate a task decomposition graph. A third unit is configured to obtain historical task data of candidate executors from a virtual workplace task execution system, construct an ability portrait of the candidate executors, use the candidate executors as execution units with autonomous decision-making capabilities according to the ability portrait, perform task negotiation and bidding between the execution units, combine the dependency relationship constraints of the task decomposition graph, generate a task distribution scheme, and dynamically adjust the distribution scheme according to the execution during the task execution process.

[0012] In a third aspect of the embodiments of the present application, An electronic device is provided, comprising: A processor; A memory for storing processor-executable instructions; The processor is configured to invoke the instructions stored in the memory to execute the method described above.

[0013] In a fourth aspect of the embodiments of the present application, A computer-readable storage medium is provided, which stores computer program instructions, and the computer program instructions are executed by a processor to implement the method described above.

[0014] The beneficial effects of the present application are as follows: The present application realizes the structured representation of task information by constructing a multi-level workplace knowledge graph, converts abstract workplace tasks into executable units with clear semantic features, and improves the accuracy of task understanding and processing.

[0015] The present invention adopts a recursive decomposition method to refine tasks, and forms a task decomposition diagram based on the dependency relationship in the knowledge graph, so that complex tasks can be reasonably divided into manageable subtasks, while maintaining the logical connection between tasks and optimizing task planning and execution efficiency.

[0016] The present invention introduces an autonomous decision-making mechanism based on the executor's capability portrait. Through task negotiation and bidding between execution units, it realizes the intelligent allocation of tasks and can make dynamic adjustments according to the execution situation, thereby improving resource utilization and task completion quality, making virtual workplace management more flexible and efficient. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a flow chart of a method for automatically generating and allocating virtual workplace tasks based on a knowledge graph according to an embodiment of the present invention; Figure 2 A flowchart for generating a task allocation plan. DETAILED DESCRIPTION

[0018] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0019] The following specific embodiments are used to describe the technical solution of the present invention in detail. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0020] refer to Figure 1 and Figure 2 , a method for automatically generating and assigning virtual workplace tasks based on knowledge graph, including: Obtain task information in the virtual workplace and construct a multi-level workplace knowledge graph consisting of a task layer and a role layer. The task layer contains task type nodes and pre-dependencies between tasks, and the role layer contains execution role nodes and role collaboration relationships. Based on the multi-level workplace knowledge graph, the task information is converted into a word meta-vector, a k-nearest neighbor graph structure is constructed based on the word meta-vector, and local information diffusion of the word meta-vector is performed by random walk to generate task semantic features; the task semantic features are recursively decomposed to decompose the overall task goal into multiple sub-goals, and the sub-goals are combined based on the dependency relationships in the multi-level workplace knowledge graph. The time complexity and resource dependency relationship of each sub-task in the decomposition tree are calculated to generate a task decomposition graph; The historical task data of the candidate performer is acquired from a virtual workplace task execution system, an ability profile of the candidate performer is constructed, the candidate performer is taken as an execution unit with autonomous decision-making ability according to the ability profile, a task allocation scheme is generated through task negotiation and bidding between execution units in combination with the dependency relationship constraint of the task decomposition graph, and the allocation scheme is dynamically adjusted according to the execution condition in the task execution process.

[0021] In an optional implementation, constructing a multi-level workplace knowledge graph including a task layer and a role layer includes: The task layer includes a task type node set and a task dependency relationship edge set, the role layer includes an execution role node set and a role collaboration relationship edge set, each edge in the task dependency relationship edge set connects an associated task type node, and each edge in the role collaboration relationship edge set connects execution role nodes having a collaboration relationship; The collaboration strength weight of each edge in the role collaboration relationship edge set is calculated based on role collaboration historical data, the collaboration strength weight is calculated by weighted calculation of a normalized value of a collaboration frequency and a normalized value of a collaboration effect score, and the collaboration effect score is evaluated based on the completion quality of a historical collaboration task; The task attribute features and the role attribute features are respectively subjected to vector transformation, the correlation degree between the transformed feature vectors is calculated to obtain a node pair correlation degree, and the node pair correlation degree represents the degree of association between a task type node and an execution role node.

[0022] The construction process of the multi-level workplace knowledge graph includes three links of data acquisition, node generation, and relationship modeling. In the data acquisition stage, enterprise internal task execution records, role information, and historical collaboration data are acquired. The task execution records include fields such as task identification, task type, start time, end time, participating role, and completion status. The role information includes fields such as role identification, role name, skill list, and historical performance. The historical collaboration data includes fields such as collaboration role identification, collaboration task identification, collaboration duration, and collaboration score.

[0023] In the node generation link, the system creates a task type node set of the task layer and an execution role node set of the role layer based on the acquired data. The task type node is generated by analyzing historical task records to extract key features. For example, when a "data analysis" task appears in the historical data, the system creates a "data analysis" task type node and extracts attribute features such as average execution duration, resource consumption, and completion difficulty of this type of task. The execution role node is generated based on enterprise organizational structure and personnel information. For example, for a "data analyst" role, the system creates a corresponding node and records attribute features such as professional skills, historical task completion rate, and average task quality score.

[0024] During the relationship modeling phase, the system constructs a set of task dependency edges and a set of role collaboration edges. Task dependency edges are determined based on the order in which tasks are executed. For example, if a "data preprocessing" task is typically executed before a "model training" task, the system will create a directed edge between these two task type nodes and label the dependency strength. Role collaboration edges are generated based on historical collaboration records. For example, if a "data analyst" and a "product manager" frequently collaborate across multiple projects, the system will create an undirected edge between these two role nodes.

[0025] The calculation of the collaboration intensity weight for a role collaboration edge involves two key metrics: the number of collaborations and the collaboration effectiveness score. The number of collaborations is calculated by counting the number of tasks jointly participated in by two roles over a period of time (e.g., the past six months). For example, if roles A and B have jointly completed 15 tasks in the past six months, and the highest number of collaborations recorded by the system is 30, the normalized value of the number of collaborations is 0.5. The collaboration effectiveness score is calculated by analyzing the quality of jointly completed tasks, using metrics such as timeliness of task completion, quality of results, and communication efficiency. Assuming that the average score of tasks jointly completed by roles A and B is 85 points (out of a maximum of 100), the normalized value of the collaboration effectiveness score is 0.85.

[0026] The collaboration intensity weight is calculated using a weighted average method, assigning weight coefficients to the normalized value of the number of collaborations and the normalized value of the collaboration effectiveness score. In actual applications, the weight ratio can be adjusted according to business characteristics. For example, for scenarios with high quality requirements, the collaboration effectiveness score weight can be set to 0.7 and the number of collaborations weight to 0.3; for scenarios with high efficiency requirements, both weights can be set to 0.5. Assuming an equal weight of 0.5, the collaboration intensity weight between role A and role B in the above example is 0.5 × 0.5 + 0.5 × 0.85 = 0.675.

[0027] Calculating the degree of association between task type nodes and execution role nodes is key to cross-layer connectivity in the knowledge graph. This process begins by transforming task and role attribute features into vectors. Task attribute features include dimensions such as task complexity, required expertise, expected completion time, and resource requirements; role attribute features include professional skills, historical performance, task types handled, and resource control capabilities.

[0028] During vector transformation, the system converts the above features into a fixed-dimension numerical vector. For example, for the "data analysis" task, its complexity is medium (0.6), the required professional knowledge is data mining and statistical analysis (after vectorization, [0.8, 0.7]), the expected completion time is 3 days (after normalization, 0.5), and the resource requirement is computing resources and storage resources (after vectorization, [0.7, 0.4]). These values are combined and converted into a 16-dimensional vector through an embedding algorithm. Similarly, for the "data analyst" role, its professional skills are data mining and visualization (after vectorization, [0.9, 0.6]), the historical performance score is 90 points (after normalization, 0.9), the main processed task types are data cleaning and feature engineering (after vectorization, [0.8, 0.7]), and the resource control ability is computing resources and analysis tools (after vectorization, [0.8, 0.6]). Similarly, it is converted into a 16-dimensional vector through an embedding algorithm.

[0029] The node pair correlation degree calculation adopts a vector similarity measurement method. The system calculates the cosine similarity of the task feature vector and the role feature vector. The closer the cosine value is to 1, the higher the matching degree of the task and the role. For example, the feature vector similarity calculation result of the "data analysis" task and the "data analyst" role is 0.85, indicating a high degree of match; while the similarity with the "front-end development engineer" role is only 0.3, indicating a lower degree of match. The system sets a correlation threshold (such as 0.6), and when the node pair correlation degree exceeds the threshold, a cross-layer connection edge is established in the knowledge graph, and the correlation degree value is taken as the weight of the edge.

[0030] In actual application scenarios, the system regularly updates the node attributes and edge weights in the knowledge graph. When new task execution records are generated, the system analyzes the task completion and adjusts the attribute values of the task type nodes; when a role performs a new task and obtains an evaluation, the system updates the skill level and performance indicators of the role node; when two roles complete a new collaborative task, the system recalculates the collaboration strength weight. For example, if roles A and B have completed a task again and obtained a high evaluation of 95 points, the system will incorporate the new data into the calculation, causing the collaboration strength weight to increase to 0.7.

[0031] When the system assigns tasks using the multi-level job knowledge graph, it first finds similar task type nodes in the task layer according to the characteristics of the new task; then it finds a set of execution role nodes with high matching degree through the node pair correlation degree; finally, it considers the current load of the role and the collaboration strength between roles to select the optimal role combination to complete the task. For example, for a new "customer data analysis" task, the system identifies its similarity with the "data analysis" task type as 0.9, and then finds the "data analyst" role with high matching degree, and considers the high collaboration strength between this role and the "product manager", and assigns the task to these two roles to complete collaboratively.

[0032] In an alternative embodiment, the task information is converted into wordpiece vectors based on the multi-level career knowledge graph, a k-neighbor graph structure is constructed based on the wordpiece vectors, and local information diffusion is performed on the wordpiece vectors by random walk to generate task semantic features, including: converting a wordpiece sequence in a task description text of the task information into a wordpiece vector; calculating semantic mutual information between wordpiece vectors, the semantic mutual information being calculated through feature transformation and interaction matrix of the wordpiece vectors, constructing a k-neighbor graph structure based on the semantic mutual information, retaining connection relationships between each wordpiece vector and its k most similar nodes, and assigning a decay weight to the retained connection relationships, the decay weight being obtained by distance weighting of the semantic mutual information through a Gaussian kernel function; performing multi-hop feature propagation based on the decay weight of the connection relationship, performing local information diffusion on the wordpiece vectors by random walk to generate task semantic features.

[0033] In this embodiment, the task information is converted into wordpiece vectors, then a k-neighbor graph structure is constructed based on the wordpiece vectors, and finally local information diffusion is performed on the wordpiece vectors by random walk to generate task semantic features.

[0034] The multi-level career knowledge graph contains entity, relationship and attribute information in the career field, where the entity can be skills, positions, industries, etc., and the relationship can be "need", "belong to", "related" and the like. The knowledge graph is organized through a multi-level structure, so as to express career knowledge of different granularities. For example, for a software development task, the knowledge graph can provide relevant knowledge at multiple levels from programming languages, frameworks, development tools to project management methods.

[0035] The process of converting task information into wordpiece vectors includes two steps of text preprocessing and vectorization. In the text preprocessing stage, the system performs word segmentation, stop word removal and other operations on the task description text. For example, for the task description "develop a customer churn prediction system based on machine learning", the system decomposes it into wordpieces such as "develop", "machine learning", "customer churn", "predict", "system". In the vectorization stage, the system maps each wordpiece to a pre-trained word vector space to obtain the corresponding wordpiece vector. Assuming that a 300-dimensional word vector space is used, each wordpiece will be represented as a 300-dimensional floating-point number vector. For the word "machine learning" in the above task description, its vector representation is [0.23, 0.45, -0.12,..., 0.67].

[0036] Calculating the semantic mutual information between word vectors is the foundation for constructing the k-nearest neighbor graph structure. Semantic mutual information is obtained through feature transformation of word vectors and calculation of the interaction matrix. Specifically, a nonlinear transformation function is applied to each word vector to enhance its expressive power. For example, a word vector v can be transformed by applying the hyperbolic tangent function to obtain the transformed vector v'. Next, the interaction matrix between any two transformed word vectors v_i' and v_j' is calculated. This matrix reflects the relationship between the two words across different dimensions. The semantic mutual information value is calculated based on the interaction matrix; a higher value indicates a stronger semantic connection between the two words. For example, the semantic mutual information value between "machine learning" and "prediction" is 0.85, while the value between "machine learning" and "system" is 0.62.

[0037] When constructing a k-nearest neighbor graph based on semantic mutual information, the connections between each word vector and its k most similar nodes are retained. In practice, the value of k can be adjusted based on task complexity and the number of words, typically ranging from 5 to 15. For example, when k = 10, for the word "machine learning," its connections to the ten words with the highest semantic mutual information values ​​are retained. These retained connections are assigned a decaying weight, derived by weighting the semantic mutual information using a Gaussian kernel function. Specifically, the Gaussian kernel function is used to convert the semantic mutual information values ​​into weights between 0 and 1, with higher semantic mutual information values ​​corresponding to weights closer to 1. For example, the connection between "machine learning" and "prediction," with a semantic mutual information value of 0.85, has a weight of 0.92; while the connection between "machine learning" and "system," with a semantic mutual information value of 0.62, has a weight of 0.74.

[0038] Multi-hop feature propagation based on the decaying weight of the connection relationship is achieved through random walks. Random walks are a method of simulating the propagation of information in a graph structure. Starting from a node, it randomly selects the next node to be visited based on the weight of the edge, and repeats this process to form a path. In this embodiment, multiple random walks are performed starting from each word vector node, and the number of steps for each walk can be set to 3 to 5 steps. For example, starting from the "Machine Learning" node, a walk is performed along the path of "Machine Learning" → "Prediction" → "Customer Churn" → "System". During the walk, the transfer probability is adjusted according to the decaying weight of the edge. The higher the weight, the greater the probability of the edge being selected.

[0039] Through multiple random walks, the system acquires local contextual information for each token vector node. This information is aggregated to generate a context-enhanced representation for each token. For example, for the "machine learning" token, its context-enhanced representation would include information about related tokens such as "prediction" and "customer churn." The weighted average of the context-enhanced representations of all tokens yields a semantic feature representation for the entire task. This representation effectively captures the complex semantic relationships between tokens in task descriptions, providing strong support for subsequent task matching and recommendations.

[0040] Through the above method, the original task description text can be converted into a vector representation with rich semantic information. This representation not only takes into account the semantics of the word itself, but also integrates the association relationship between word elements, thereby more comprehensively depicting the semantic connotation of the task.

[0041] In an optional embodiment, sub-goals are combined based on dependencies in a multi-level workplace knowledge graph, and the time complexity and resource dependencies of each sub-task in the decomposition tree are decomposed to generate a task decomposition diagram, including: Performing a correlation analysis on the sub-goals based on the dependency relationships, combining sub-goals with correlations higher than a preset correlation threshold to form a plurality of sub-goal combination sets, calculating the execution cost of the sub-goals based on the estimated execution time and resource usage of each sub-goal in the sub-goal combination sets, and selecting the sub-goal combination with the lowest execution cost and meeting the preset dependency strength requirements as the optimal combination solution; Obtaining the task input size and single operation time for the sub-goals in the optimal combination solution, determining the fixed overhead time during task execution, and calculating the time complexity of each sub-goal based on the input size, the single operation time, and the fixed overhead time; A dependency matrix between tasks and resources is constructed based on the time complexity. The dependency matrix records the degree of dependence of each task on various resources. A task decomposition diagram is generated according to the dependency matrix.

[0042] In practical applications, a dependency network between sub-goals is constructed using the workplace knowledge graph. This network comprises three main types of dependencies: semantic, structural, and temporal. Semantic dependencies represent the degree of relevance between sub-goals in terms of business meaning and are derived by analyzing the functional mapping between skill nodes and task nodes. Structural dependencies represent the order in which sub-goals are executed and are derived by analyzing the execution paths between task nodes. Temporal dependencies represent the temporal relationship between sub-goals in terms of resource usage and are derived by analyzing the order in which resource nodes are called.

[0043] After the dependency relationship network is constructed, the correlation degree of the sub-goals is analyzed. The correlation degree calculation method is based on weighted summation of dependency strength, and the preset correlation degree threshold is set to 0.6. When the correlation degree between sub-goals exceeds the threshold, they are combined into the same set. For all sub-goals in the correlation network, the correlation degree of each pair of sub-goals is calculated by traversal, forming multiple sub-goal combination sets. Taking a project containing 10 sub-goals as an example, 3 sub-goal combination sets are formed through correlation degree analysis: set 1 contains sub-goals A, B, and C; set 2 contains sub-goals D, E, F, and G; and set 3 contains sub-goals H, I, and J.

[0044] For the formed sub-goal combination sets, the execution cost of each set is calculated, which consists of execution duration and resource occupation. The execution duration is obtained through historical data statistics or expert experience estimation, and the resource occupation includes multi-dimensional indicators such as computing resources, storage resources, and network resources. Weight coefficients are assigned to different resource types to calculate the comprehensive resource occupation. For example, for sub-goal A in set 1, the predicted execution duration is 8 hours, the computing resource occupation is CPU core number x usage duration = 4 cores x 8 hours = 32 core hours, the storage resource occupation is 500 GB, and the network resource occupation is bandwidth x transmission duration = 100 Mbps x 2 hours = 720 GB. The execution cost of sub-goal A can be represented as the product of execution duration and resource occupation. Similarly, the execution costs of other sub-goals in the set are calculated and summed to obtain the total execution cost of the set.

[0045] It is also necessary to consider whether the sub-goal combination meets the preset dependency strength requirement, which specifies the minimum correlation degree between sub-goals within the combination set and the maximum correlation degree between sub-goals in different combination sets. For example, the average correlation degree between sub-goals within the set should not be less than 0.65, and the maximum correlation degree between sub-goals in different sets should not exceed 0.4. The total execution cost of each combination scheme is calculated, and the schemes that meet the dependency strength requirement are selected, and the one with the lowest execution cost is chosen as the optimal combination scheme.

[0046] For each sub-goal in the determined optimal combination scheme, further calculate its time complexity. Time complexity calculation needs three key parameters: task input size, single operation time consumption and fixed overhead time. Task input size refers to the amount of data processed or the number of operation objects by the sub-goal, single operation time consumption refers to the time required to process a unit of data, and fixed overhead time refers to the preparation time independent of the input size. These parameters are obtained through historical execution data or prototype testing. For example, for a data processing sub-goal, the input size may be the number of records to be processed, such as 1 million; the single operation time consumption is the average time to process a record, such as 0.5 milliseconds; and the fixed overhead time includes data loading and initialization time, such as 30 seconds. Based on these parameters, the system can represent the time complexity of the sub-goal as the fixed overhead time plus the input size multiplied by the single operation time consumption. In this way, a time complexity model is established for each sub-goal in the optimal combination scheme.

[0047] According to the calculated time complexity, the system constructs a dependency relationship matrix between tasks and resources. This matrix is a two-dimensional table, with rows representing each sub-goal and columns representing each type of resource, and the matrix element value representing the dependency degree of the sub-goal on the resource. The dependency degree is calculated through the relationship between the time complexity of the sub-goal and the resource utilization efficiency. For example, if the time complexity of sub-goal A is highly dependent on CPU computing power, its dependency degree on computing resources is higher, which can be 0.8; if the access frequency to storage resources is low, the dependency degree on storage resources can be 0.3. The system determines the calculation method of the dependency degree based on business rules and historical data analysis to generate a complete dependency relationship matrix. Taking a project containing 5 sub-goals and 3 types of resources as an example, its dependency relationship matrix can be: the dependency degrees of sub-goal A on computing resources, storage resources and network resources are 0.8, 0.3 and 0.5 respectively; the dependency degrees of sub-goal B are 0.6, 0.7 and 0.2 respectively; the dependency degrees of sub-goal C are 0.4, 0.9 and 0.3 respectively; the dependency degrees of sub-goal D are 0.7, 0.4 and 0.8 respectively; and the dependency degrees of sub-goal E are 0.5, 0.6 and 0.7 respectively.

[0048] After the dependency matrix is constructed, a task decomposition graph is generated based on the matrix. The task decomposition graph is a directed graph structure, with nodes representing sub-goals and edges representing the dependency relationship between sub-goals. According to the dependency matrix, the competition situation of each sub-goal for resources is determined, and the resource conflict degree between sub-goals is calculated. Sub-goals with high resource conflict degree are usually not suitable for parallel execution. Combining the previously analyzed semantic dependency, structural dependency and time sequence dependency, as well as the resource conflict situation, the overall dependency relationship between sub-goals is determined. For example, if sub-goal A and sub-goal B are highly conflicted in resource use (such as both depending on the same computing resource), and there is a structural dependency, a directed edge from A to B is established in the task decomposition graph, indicating that B depends on the completion of A. Analyze all pairs of sub-goals to construct a complete directed graph structure.

[0049] After the task decomposition graph is generated, each sub-goal and its dependency relationship can be visualized and displayed, supporting project managers to plan tasks and allocate resources. When the actual execution process changes or the resource configuration is adjusted, the dependency matrix and the task decomposition graph can be updated in real time to ensure the rationality of the task execution plan and the efficiency of resource utilization. Through this task decomposition method based on the multi-level job knowledge graph, intelligent decomposition and optimized combination of complex projects are realized, and the project execution efficiency and resource utilization rate are improved.

[0050] In an optional implementation, according to the capability profile, the candidate executor is taken as an execution unit with autonomous decision-making capability, and a task allocation scheme is generated through task negotiation and bidding between execution units, combined with the dependency relationship constraints of the task decomposition graph, including: A state vector of the execution unit is constructed, which includes an image vector corresponding to the capability profile and a load level indicator. The utility value of the execution unit for the task is calculated based on the matching degree of the image vector and the task demand and the load level indicator; A bidding function of the execution unit is constructed based on the utility value. The bidding function is calculated by the product of the utility value and an exponential decay term of execution cost. The execution cost includes a weighted combination of time delay cost and load transfer cost; The current load state and the remaining resource amount of the execution unit are normalized and weighted summed to obtain a resource consumption score. The task allocation probability is calculated by matching the resource consumption score with the bid value output by the bidding function. Combined with the dependency relationship constraints of the task decomposition graph, a task allocation scheme is generated.

[0051] In this implementation, the candidate executor is taken as an execution unit with autonomous decision-making capability, and a task allocation scheme is generated through task negotiation and bidding between execution units, combined with the dependency relationship constraints of the task decomposition graph.

[0052] In a specific implementation method, a state vector of the execution unit is constructed. The state vector includes an image vector corresponding to the capability profile and a load level indicator. The image vector describes the skill characteristics of the execution unit, including multiple attributes such as technical field identification, professional level, experience years, response speed, and quality score. Taking a software development team as an example, the image vector of execution unit A can be represented as front-end development expertise 0.9, back-end development expertise 0.6, testing capability 0.5, and project management capability 0.7. The load level indicator reflects the current task carrying capacity of the execution unit, including the ratio of the number of allocated tasks to the maximum carrying capacity. For example, execution unit A currently undertakes 3 tasks, and the maximum carrying capacity is 5 tasks, so the load level indicator of execution unit A is 0.6.

[0053] After the state vector is constructed, the utility value of the execution unit for the task is calculated, which is calculated based on the matching degree of the image vector and the task demand and the load level indicator. The matching degree is calculated by using the cosine similarity method to compare the task demand vector and the execution unit image vector. For example, the task demand vector of a development task is front-end development demand 0.8, back-end development demand 0.4, and testing demand 0.3, and the matching degree calculated with the image vector of execution unit A is 0.85. The load level indicator is converted by an exponential function, so that the higher the load, the lower the utility value. For execution unit A with a load level indicator of 0.6, the load adjustment factor obtained after conversion is 0.75. The final utility value is the product of the matching degree and the load adjustment factor, that is, 0.85×0.75=0.6375.

[0054] Based on the utility value, a bidding function of the execution unit is constructed. The bidding function is calculated by multiplying the utility value and the exponential decay term of the execution cost. The execution cost includes the weighted combination of the time delay cost and the load transfer cost. The time delay cost reflects the difference between the expected time required for task execution and the expected completion time. For example, the task is expected to be completed in 3 days, and execution unit A is expected to take 4 days, so the time delay is 1 day, and the corresponding time delay cost can be set to 0.2. The load transfer cost measures the degree of influence of receiving a new task on the existing tasks of the execution unit. For example, after execution unit A receives a new task, the load level will rise from 0.6 to 0.8, and the load transfer cost can be set to 0.15. Assuming that the weights of the time delay cost and the load transfer cost are 0.6 and 0.4 respectively, the execution cost is 0.2×0.6+0.15×0.4=0.18.

[0055] Setting the exponential decay factor to 2.5, the bidding value of execution unit A for the task is calculated as: 0.6375×exponential decay term, where the exponential decay term is-0.18×2.5 power of 2.718, which is approximately equal to 0.64. Therefore, the bidding value is 0.6375×0.64=0.408.

[0056] The current load state and the remaining resource amount of the execution unit are normalized and then weighted and summed to obtain a resource consumption score. Assuming that the CPU utilization of execution unit A is 50%, the memory occupation is 40%, and the bandwidth usage is 30%, the normalized values are 0.5, 0.4, and 0.3 respectively. Assuming that the weights of the three resources are 0.5, 0.3, and 0.2 respectively, the resource consumption score is 0.5*0.5+0.4*0.3+0.3*0.2=0.43.

[0057] The resource consumption score is matched with the bid value output by the bid function to obtain a task allocation probability. The calculation method is to divide the bid value by the product of the resource consumption score and the adjustment factor. Assuming that the adjustment factor is 1.2, the task allocation probability is 0.408 / (0.43*1.2)=0.79.

[0058] The above calculation process is repeated for all execution units in the system to obtain the allocation probability of each execution unit for a specific task. For example, the allocation probabilities of execution units A, B, and C for task 1 are 0.79, 0.62, and 0.45 respectively.

[0059] Finally, a task allocation scheme is generated in combination with the dependency constraints of the task decomposition graph. The task decomposition graph describes the precedence relationship between tasks, for example, task 2 depends on the completion of task 1, and tasks 3 and 4 depend on the completion of task 2. The dependency constraint requires that tasks that depend on each other should be allocated to a combination of execution units with high collaboration efficiency as much as possible. By calculating the collaboration efficiency matrix between execution units, combining the task dependency graph, and using an iterative optimization method, the final allocation scheme is generated. For example, the collaboration efficiency of execution unit A and execution unit B is 0.85, which is higher than the collaboration efficiency of A and C, which is 0.62, so if task 1 is allocated to execution unit A, task 2 tends to be allocated to execution unit B.

[0060] The final generated task allocation scheme includes: task 1 is allocated to execution unit A (probability 0.79), task 2 is allocated to execution unit B (probability 0.82, considering the collaboration efficiency with A), task 3 is allocated to execution unit A (probability 0.71), and task 4 is allocated to execution unit C (probability 0.68).

[0061] Through the above technical implementation, the method can automatically generate an efficient task allocation scheme according to the capability profile, load state of the execution unit, and dependency relationship between tasks, thereby improving the overall execution efficiency of the system.

[0062] In an optional implementation, the resource consumption score is matched with the bid value output by the bid function to obtain a task allocation probability, and the task allocation scheme is generated in combination with the dependency constraints of the task decomposition graph, which includes: When multiple tasks with dependencies are assigned to different execution units, the task assignment probability is penalized according to the communication overhead between execution units, and the task assignment probability is reduced; when multiple tasks with dependencies are assigned to the same execution unit, the task assignment probability is rewarded according to the parallelism of the task, and the task assignment probability is increased. A task assignment scheme is generated according to the adjusted task assignment probability, which determines the target execution unit of each task and records the execution order and resource allocation information of the task.

[0063] The application provides a method for calculating task assignment probability by matching degree calculation according to resource consumption score and bidding value output by bidding function, and generating task assignment scheme combined with dependency constraints of task decomposition graph.

[0064] For each execution unit in the system, such as different processor cores or computing nodes, the resource consumption score of processing each type of task is calculated. The resource consumption score can be evaluated based on processor load, memory usage, energy consumption and other dimensions. For example, for execution unit A processing task 1, it may consume CPU resources 75%, memory resources 60%, and energy consumption 3.5W, and the resource consumption score is 0.68 by weighted calculation; while execution unit B processes the same task, it consumes CPU resources 60%, memory resources 80%, and energy consumption 2.8W, and the resource consumption score is 0.72.

[0065] The bidding function outputs the bidding value according to the current state of the execution unit and the characteristics of the task. The bidding value reflects the willingness of the execution unit to process the task. For example, execution unit A has a low current load, and the bidding value for task 1 is 0.85; execution unit B has a medium current load, and the bidding value for task 1 is 0.65.

[0066] The matching degree calculation is obtained by comparing the relationship between the resource consumption score and the bidding value. One implementation is to calculate the cosine similarity of the two, and another way is to calculate the weighted average value of the two. For example, for execution unit A processing task 1, the resource consumption score is 0.68 and the bidding value is 0.85, and the weights are 0.4 and 0.6 respectively, and the weighted average value is 0.782; for execution unit B processing task 1, the resource consumption score is 0.72 and the bidding value is 0.65, and the weighted average value is 0.678. These weighted average values are normalized to convert to preliminary task assignment probability, for example, the probability of execution unit A processing task 1 is 0.536, and the probability of execution unit B processing task 1 is 0.464.

[0067] When considering the dependency between tasks, the initial task allocation probability described above will be adjusted. If the tasks with dependencies are allocated to different execution units, the communication overhead between execution units needs to be considered. The communication overhead can be quantified by factors such as network bandwidth, communication delay, etc. For example, the communication overhead between execution unit A and execution unit B is 0.25 (normalized value), when task 1 is allocated to execution unit A and dependent task 2 is allocated to execution unit B, the task allocation probability will be penalized according to the communication overhead. Specifically, the probability of task 1 being allocated to execution unit A can be reduced from 0.536 to 0.536 x (1-0.25) = 0.402.

[0068] On the contrary, if the tasks with dependencies are allocated to the same execution unit, the system will reward according to the parallelism of the tasks, and increase the task allocation probability. The task parallelism can be obtained by analyzing the internal structure of the task or historical execution data. For example, if task 1 and task 2 can be partially parallel executed on the same execution unit, and the parallelism is 0.3 (normalized value), then the probability of allocating task 1 and task 2 to execution unit A will be increased from the original value to the original value x (1+0.3).

[0069] Through multiple rounds of iteration calculation, the target execution unit of each task is finally determined. For example, in a system containing 5 tasks and 3 execution units, the final task allocation scheme can be: task 1 and task 2 are allocated to execution unit A, task 3 is allocated to execution unit B, and task 4 and task 5 are allocated to execution unit C. At the same time, the execution order of the tasks and the resource allocation information are also recorded, such as task 1 is allocated 70% of the processor resources on execution unit A, task 2 is allocated 30% of the processor resources on execution unit A, and task 1 needs to be executed before task 2.

[0070] In practical applications, this method can adaptively adjust the task allocation scheme according to dynamic changes. For example, when the load of a certain execution unit suddenly increases, its competitive value will decrease accordingly, resulting in a decrease in task allocation probability, so that the task is allocated to other more suitable execution units. This dynamic adjustment mechanism ensures efficient use of system resources and timely processing of tasks.

[0071] Practical tests show that in a large system containing 50 tasks and 10 execution units, compared with the traditional static allocation method, the overall throughput of the system is improved by 23%, the average response time of the task is reduced by 18%, and the resource utilization is improved by 15%, verifying the effectiveness and practical value of the method.

[0072] The virtual job task automatic generation and distribution system based on the knowledge graph of the embodiment of the application comprises: The first unit is configured to acquire task information in a virtual workplace, and construct a multi-level workplace knowledge graph including a task layer and a role layer, wherein the task layer includes task type nodes and inter-task precedence dependency relationships, and the role layer includes execution role nodes and role collaboration relationships. The second unit is configured to convert the task information into wordpiece vectors based on the multi-level workplace knowledge graph, construct a k-nearest neighbor graph structure based on the wordpiece vectors, perform local information diffusion on the wordpiece vectors in a random walk manner, and generate task semantic features; recursively decompose the task semantic features, decompose an overall task target into multiple sub-targets, combine the sub-targets based on dependency relationships in the multi-level workplace knowledge graph, decompose time complexity and resource dependency relationships of each sub-task in a decomposition tree, and generate a task decomposition graph. The third unit is configured to acquire historical task data of candidate executors from a virtual workplace task execution system, construct a capability profile of the candidate executors, use the candidate executors as execution units with autonomous decision-making capabilities according to the capability profile, generate a task allocation scheme through task negotiation and bidding between the execution units, in combination with dependency relationship constraints of the task decomposition graph, and dynamically adjust the allocation scheme according to execution conditions in a task execution process.

[0073] In a third aspect, an electronic device is provided, including: a processor; a memory for storing processor-executable instructions; The processor is configured to invoke the instructions stored in the memory to execute the method described above.

[0074] In a fourth aspect, a computer-readable storage medium is provided, which stores computer program instructions, and the computer program instructions are executed by a processor to implement the method described above.

[0075] The present application can be a method, device, system and / or computer program product. The computer program product can include a computer readable storage medium having computer readable program instructions loaded thereon for implementing various aspects of the present application.

[0076] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for automatically generating and assigning virtual workplace tasks based on knowledge graphs, characterized in that: include: Obtain task information in the virtual workplace and construct a multi-level workplace knowledge graph consisting of a task layer and a role layer. The task layer contains task type nodes and pre-dependencies between tasks, and the role layer contains execution role nodes and role collaboration relationships. Converting the task information into a word meta-vector based on the multi-level workplace knowledge graph, constructing a k-nearest neighbor graph structure based on the word meta-vector, and performing local information diffusion on the word meta-vector by random walk to generate task semantic features; Recursively decompose the semantic features of the task, decompose the overall task goal into multiple sub-goals, combine the sub-goals based on the dependencies in the multi-level workplace knowledge graph, decompose the time complexity and resource dependencies of each sub-task in the tree, and generate a task decomposition diagram; The historical task data of candidate executors is obtained from the virtual workplace task execution system, and a capability profile of the candidate executor is constructed. Based on the capability profile, the candidate executor is regarded as an execution unit with autonomous decision-making ability. Through task negotiation and bidding between execution units, combined with the dependency constraints of the task decomposition diagram, a task allocation plan is generated, and the allocation plan is dynamically adjusted according to the execution status during the task execution process.

2. The method according to claim 1, characterized in that Building a multi-level workplace knowledge graph that includes task and role layers includes: The task layer includes a task type node set and a task dependency edge set, and the role layer includes an execution role node set and a role collaboration edge set, wherein each edge in the task dependency edge set connects the associated task type node, and each edge in the role collaboration edge set connects the execution role node with a collaboration relationship; Calculating the collaboration strength weight of each edge in the role collaboration relationship edge set based on the role collaboration history data. The collaboration strength weight is calculated by weighting the normalized value of the number of collaborations with the normalized value of the collaboration effect score. The collaboration effect score is based on the completion quality evaluation of historical collaboration tasks. The task attribute features and the role attribute features are respectively transformed into vectors, and the correlation between the transformed feature vectors is calculated to obtain the node pair correlation, which represents the degree of association between the task type node and the execution role node.

3. The method according to claim 1, characterized in that Based on the multi-level workplace knowledge graph, the task information is converted into a word meta-vector, a k-nearest neighbor graph structure is constructed based on the word meta-vector, and local information diffusion is performed on the word meta-vector by random walk to generate task semantic features, including: Converting a word sequence in the task description text of the task information into a word vector; Calculate the semantic mutual information between word meta-vectors, where the semantic mutual information is obtained by transforming the feature of the word meta-vectors and calculating the interaction matrix. Construct a k-nearest neighbor graph structure based on the semantic mutual information, retain the connection relationship between each word meta-vector and its k most similar nodes, and assign an attenuation weight to the retained connection relationship. The attenuation weight is obtained by distance-weighting the semantic mutual information using a Gaussian kernel function. Multi-hop feature propagation is performed based on the attenuation weight of the connection relationship, and local information diffusion of word element vectors is performed through random walks to generate task semantic features.

4. The method according to claim 1, wherein Based on the dependencies in the multi-level workplace knowledge graph, the sub-goals are combined, and the time complexity and resource dependencies of each sub-task in the decomposition tree are calculated to generate a task decomposition diagram including: Performing a correlation analysis on the sub-goals based on the dependency relationships, combining sub-goals with correlations higher than a preset correlation threshold to form a plurality of sub-goal combination sets, calculating the execution cost of the sub-goals based on the estimated execution time and resource usage of each sub-goal in the sub-goal combination sets, and selecting the sub-goal combination with the lowest execution cost and meeting the preset dependency strength requirements as the optimal combination solution; Obtaining the task input size and single operation time for the sub-goals in the optimal combination solution, determining the fixed overhead time during task execution, and calculating the time complexity of each sub-goal based on the input size, the single operation time, and the fixed overhead time; A dependency matrix between tasks and resources is constructed based on the time complexity. The dependency matrix records the degree of dependence of each task on various resources. A task decomposition diagram is generated according to the dependency matrix.

5. The method according to claim 1, wherein Based on the capability profile, candidate executors are considered as execution units with autonomous decision-making capabilities. Through task negotiation and bidding among the execution units, combined with the dependency constraints of the task decomposition diagram, a task allocation plan is generated, including: Constructing a state vector of the execution unit, the state vector including a portrait vector corresponding to the capability portrait and a load level indicator, and calculating the utility value of the execution unit for the task based on the matching degree between the portrait vector and the task requirements and the load level indicator; Constructing a bidding function for the execution unit based on the utility value, wherein the bidding function is calculated by multiplying the utility value by an exponential decay term of an execution cost, wherein the execution cost comprises a weighted combination of a time delay cost and a load transfer cost; The current load status and the remaining resource amount of the execution unit are normalized and then weighted summed to obtain a resource consumption score. The resource consumption score is matched with the bidding value output by the bidding function to calculate the task allocation probability. Combined with the dependency constraints of the task decomposition graph, a task allocation plan is generated.

6. The method according to claim 5, characterized in that The resource consumption score is matched with the bidding value output by the bidding function to obtain a task allocation probability. Combined with the dependency constraints of the task decomposition graph, a task allocation plan is generated, including: When multiple tasks with dependencies are assigned to different execution units, the task assignment probability is penalized according to the communication overhead between the execution units, thereby reducing the task assignment probability; when multiple tasks with dependencies are assigned to the same execution unit, the task assignment probability is rewarded according to the parallelism of the tasks, thereby increasing the task assignment probability; A task allocation plan is generated according to the adjusted task allocation probability. The task allocation plan determines the target execution unit of each task and records the execution order and resource allocation information of the tasks.

7. A virtual workplace task automatic generation and allocation system based on knowledge graph, used to implement the method according to any one of claims 1 to 6, characterized in that: include: The first unit is used to obtain task information in the virtual workplace and construct a multi-level workplace knowledge graph consisting of a task layer and a role layer. The task layer contains task type nodes and pre-dependencies between tasks, and the role layer contains execution role nodes and role collaboration relationships. The second unit is configured to convert the task information into a word meta-vector based on the multi-level workplace knowledge graph, construct a k-nearest neighbor graph structure based on the word meta-vector, perform local information diffusion on the word meta-vector by random walk, and generate task semantic features; Recursively decompose the semantic features of the task, decompose the overall task goal into multiple sub-goals, combine the sub-goals based on the dependencies in the multi-level workplace knowledge graph, decompose the time complexity and resource dependencies of each sub-task in the tree, and generate a task decomposition diagram; The third unit is used to obtain the historical task data of candidate executors from the virtual workplace task execution system, construct a capability profile of the candidate executor, and treat the candidate executor as an execution unit with autonomous decision-making capabilities based on the capability profile. Through task negotiation and bidding between execution units, combined with the dependency constraints of the task decomposition diagram, a task allocation plan is generated, and the allocation plan is dynamically adjusted according to the execution status during the task execution process.

8. An electronic device, characterized in that: include: processor; a memory for storing processor-executable instructions; The processor is configured to call the instructions stored in the memory to execute the method according to any one of claims 1 to 6.

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

Images