Strategy data matching analysis method and system applied to task management

Abstract

The invention discloses a strategy data matching analysis method and system applied to task management, relates to the technical field of task data matching, and solves the problems that when tasks are sequentially processed in the prior art, if the strategy data of multiple tasks are repeated, the strategy data need to be repeatedly extracted and matched for multiple times, and the task processing efficiency is high. The data processing amount is increased, and the task processing efficiency is influenced. According to the method, to-be-processed tasks are sequenced through priorities or task relevance, and a first task list is generated; the task sequence is adjusted according to the resource similarity of all the to-be-processed tasks in the first task list, and a second task list is obtained; finally, strategy resources are matched for the to-be-processed tasks in the second task list, and the to-be-processed tasks are executed based on the strategy resources. According to the method, the task sequence is adjusted through the resource similarity, the matching efficiency of the data resources can be improved, and meanwhile, human resources are reasonably allocated to ensure that all the to-be-processed tasks are processed in time.

Description

Technical Field

[0001] The present application relates to the field of task data matching, and specifically to a strategy data matching analysis method and system applied to task management. Background Art

[0002] With the increasing knowledge density and explosive growth of data, the number and types of tasks on many platforms are also increasing. As the number of platform tasks increases, different tasks have different priorities and resource requirements. If reasonable strategy data cannot be matched for each task, task execution will be chaotic.

[0003] In order to ensure the processing efficiency and quality of tasks, it is necessary to match appropriate strategic resources for each task in the platform. These strategic resources include manpower, time, and data. When performing multi-task processing, concurrent tasks are generally analyzed and processed according to priority or processing order, and then the tasks are processed according to the analysis results. During the task processing process, the policy data is analyzed and matched. In this solution, the policy data of each task is matched separately, and when computing resources are insufficient, each task needs to be processed in sequence. If there is duplication between the policy data of multiple tasks, the policy data needs to be extracted and matched repeatedly, which will increase the amount of data processing and affect the efficiency of task processing.

[0004] The present application provides a strategy data matching analysis method and system applied to task management to solve the above technical problems. Summary of the invention

[0005] The present application aims to solve at least one of the technical problems existing in the prior art; to this end, the present application proposes a strategy data matching analysis method and system applied to task management, which is used to solve the technical problem that when the prior art processes each task in sequence, if there is duplication between the strategy data of multiple tasks, the strategy data needs to be repeatedly extracted and matched, which increases the data processing volume and affects the task processing efficiency.

[0006] To achieve the above-mentioned purpose, the first aspect of the present application provides a strategy data matching analysis method applied to task management, comprising: Receiving a number of pending tasks; sorting the number of pending tasks according to a sorting rule to generate a task list 1; wherein the sorting rule is set based on task priority or task relevance; Analyze the strategic resources required by each task to be processed in task list one; adjust task list one according to the similarity of each resource in the strategic resources to obtain task list two; wherein the strategic resources include data resources and human resources; Based on the demand policy resources, the policy resources are matched to each pending task in the task list 2 in turn, and the corresponding pending task is executed according to the matched policy resources.

[0007] Preferably, the plurality of tasks to be processed are sorted according to a sorting rule, including: Identify the priorities and task relevance of several pending tasks; wherein the task relevance is used to determine the order in which the multiple pending tasks are processed; According to the task relevance, a number of pending tasks are divided into a number of task groups, and the number of task groups are sorted according to the priority to obtain a task list 1.

[0008] Preferably, several task groups are sorted according to priority, including: Extract the task attributes of each task to be processed; quantify the task attributes to obtain attribute tags; the task attributes include time requirements, resource quantity, complexity and risk; The weight of each task to be processed is calculated based on the entropy weight method and attribute labeling, and the priority of each task group is obtained by weighted average of the weight and the priority. The order of several task groups is achieved through the priority of each task group.

[0009] Preferably, analyzing the strategic resources required by each task to be processed in the task list 1 includes: Collect historical task data; perform data cleaning and feature extraction on historical task data to obtain basic data; the historical task data includes task description, required resource type and task results; Perform correlation analysis on task features and resource types in basic data, and build model training data based on the analysis results; train the machine learning model built with model training data, and mark the machine learning model as a resource matching model after verification and testing; the machine learning model includes decision tree, random forest or neural network model; The task characteristics of each task to be processed in the task list 1 are input into the resource matching model to obtain the corresponding strategic resources.

[0010] Preferably, adjusting the task list 1 according to the similarity of each resource in the strategy resource includes: Extract the policy resources corresponding to each pending task in the task list 1, and calculate the data similarity and manpower similarity of any two pending tasks in the task list 1 according to the policy resources; The order of pending tasks in task list one is adjusted according to data similarity and manpower similarity to obtain task list two; the higher the data similarity, the closer the corresponding pending tasks are in task list two; the higher the manpower similarity, the farther the corresponding pending tasks are in task list two.

[0011] Preferably, adjusting the task list 1 according to the similarity of each resource in the strategy resource includes: Extract the policy resources corresponding to each task group in the task list 1, and calculate the data similarity and manpower similarity of any two task groups in the task list 1 according to the policy resources; The order of the task groups in task list one is adjusted according to the data similarity and the manpower similarity to obtain task list two; the higher the data similarity, the closer the corresponding task groups are in task list two; the higher the manpower similarity, the farther the corresponding task groups are in task list two.

[0012] Preferably, a task list 2 is obtained, including: Add the target object in the task list and target audience The data similarity is marked as , human similarity is marked as ; Among them, when the task list 1 is obtained by sorting the task group, the target object is the task group; when the task list 1 is not obtained by sorting the task group, the target object is the task to be processed; Define the similarity influence function ,and ; Similarity influence function and sorting algorithms to obtain sorting results; wherein the sorting algorithms include greedy algorithms, dynamic programming algorithms or graph theory algorithms, and is the weight coefficient; The order of the target objects is adjusted according to the sorting results to obtain task list 2.

[0013] A second aspect of the present application provides a policy data matching and analysis system for task management, comprising: a resource matching module, and a task processing module connected thereto; Task processing module: used for receiving a number of tasks to be processed; sorting the number of tasks to be processed to obtain a task list 1, and adjusting the task list 1 to obtain a task list 2; Resource matching module: used to analyze the policy resources required by each pending task in task list one, and match the policy resources to the pending tasks in task list two based on the required policy resources; and, Execute the corresponding pending tasks according to the matching policy resources.

[0014] Compared with the prior art, the beneficial effects of this application are: 1. This application first sorts the pending tasks by priority or task relevance to generate a task list one; then adjusts the task order according to the resource similarity of each pending task in task list one to obtain a task list two; finally, matches policy resources for the pending tasks in task list two, and executes each pending task based on the policy resources; this application adjusts the task order by resource similarity, which can improve the matching efficiency of data resources, and at the same time reasonably allocates human resources to ensure the timely processing of each pending task.

[0015] 2. This application first divides a number of pending tasks into a number of task groups according to the task relevance, calculates the priority of each task group according to the task attributes of the pending tasks in each task group and the entropy weight method, and sorts the task groups according to the priority to obtain task list one; adjusts task list one according to the resource similarity between task groups to obtain task list two; this application matches strategic resources in the form of task groups, and does not only rely on priority to sort task groups, but also comprehensively considers the correlation between each pending task, which can improve the matching efficiency of strategic resources and task processing efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 This is a method flow diagram of a strategy data matching and analysis method applied to task management in Embodiment 1 of the present application; Figure 2 This is a flow chart of generating a task list 1 according to a task group in Embodiment 2 of the present application; Figure 3 This is a schematic diagram of the system principle of a strategy data matching and analysis system applied to task management in this application. DETAILED DESCRIPTION

[0018] The technical solution of the present application will be described clearly and completely in conjunction with the embodiments below. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application.

[0019] When processing tasks, the platform needs to allocate strategic resources for each task. The strategic resources mainly include data resources and human resources, that is, the data and specific executors required to complete the task. When the computing resources of the platform are sufficient, multiple tasks can be processed simultaneously through parallel processing; when the computing resources are insufficient, tasks need to be processed in sequence according to priority. Regardless of whether the computing resources are sufficient or not, the matching of human resources and data resources is involved. If efficient matching cannot be achieved, it will not only increase the workload of policy resource matching, but also affect the efficiency of task processing. In order to solve the efficiency of policy resource matching during task processing, the present application provides a policy data matching analysis method and system for task management.

[0020] Example 1: Please refer to Figure 1 The first aspect of the present application provides a strategy data matching analysis method for task management, including: Receive a number of pending tasks; sort the pending tasks according to the sorting rules to generate a task list one; analyze the policy resources required by each pending task in the task list one; adjust the task list one according to the similarity of each resource in the policy resources to obtain a task list two; match the policy resources for each pending task in the task list two in turn based on the required policy resources, and execute the corresponding pending tasks according to the matched policy resources.

[0021] When the platform receives several pending tasks, it generally determines the order of task processing based on the time they were received or their priority. When processing tasks according to the order, it determines the processing strategy and matches the corresponding strategy resources, such as the basic data needed to process the tasks and staff with the required skills, that is, data resources and human resources.

[0022] In the above general task processing process, there are mainly two problems: 1) Matching and calling of data resources. Data resources are matched when tasks are executed sequentially, that is, data resources are matched in real time. However, the identification, matching and calling of data resources will also take up more computing resources and time, which will obviously affect the efficiency of task execution. After the task is executed, the data resources are released. If the data resources required by adjacent tasks are the same or partially overlap, the same or overlapping data resources need to be called repeatedly, which not only wastes computing resources, but also affects the matching efficiency of data resources.

[0023] 2) Matching and calling of human resources. Human resources are matched when executing tasks in sequence. If the current task cannot be matched with suitable human resources, the current task cannot be processed in time, such as waiting for other tasks to release human resources, which will cause subsequent tasks to be postponed. If the current task is suspended, the priority of the task will be ignored when executing the next task.

[0024] In this embodiment, the sorting rules are mainly set according to the task priority or task relevance. Specifically, if the task to be processed has been set with a task priority, it can be sorted according to the task priority; or, if the tasks to be processed are related to each other, such as task A is a predecessor task of task B, then according to the relevance, the execution order of task A is placed before task B. Of course, in other preferred embodiments, the task priority and task relevance can also be combined to set the sorting rules, first sorting according to the task priority, and then adjusting according to the task relevance.

[0025] After sorting the tasks to be processed according to the sorting rules, a task list 1 can be obtained. In order to solve the efficiency and rationality of the policy resource configuration, this embodiment adjusts the task list 1 according to the resource similarity to obtain a task list 2, which specifically includes the following steps: First, analyze the strategic resources required by each pending task in Task List 1, including: Collect historical task data; perform data cleaning and feature extraction on historical task data to obtain basic data; perform correlation analysis on task features and resource types in the basic data, and build model training data based on the analysis results; train the constructed machine learning model with the model training data, and mark the machine learning model as a resource matching model after verification and testing; input the task features of each pending task in Task List 1 into the resource matching model to obtain the corresponding policy resources.

[0026] Before intelligently matching the strategic resources of each task to be processed, the association relationship between tasks and resources should be established. In this embodiment, the historical task data is collected to build the relationship between each task and its required resources. Therefore, the historical task data at least includes task description, required resource type, task results, etc. Next, the historical task data is cleaned and feature extracted to remove invalid data to ensure data quality; extract task-related features, such as task type, required data type, and participants.

[0027] After obtaining the basic data, analyze the relationship between different tasks and the required strategic resources, use statistical methods (such as correlation analysis) to determine which strategic resources have a strong correlation with task characteristics, and integrate the analysis results to generate model training data.

[0028] Build a machine learning model, train it according to the conventional model training method, and mark the obtained model as a resource matching model after cross-validation and test set evaluation. The machine learning model here can be selected according to the data type of the policy resources and task characteristics, such as decision trees, random forests, and various neural network models.

[0029] It should be noted that the resource matching model in this embodiment mainly learns the mapping relationship between task characteristics and required resources. In the process of using the resource matching model, the corresponding policy resources can be obtained by inputting task characteristics.

[0030] Extract the pending tasks from task list one in turn, input the task features of the pending tasks into the resource matching model to obtain the required strategic resources, so that the data resources and human resources required by each pending task during execution can be obtained.

[0031] When the policy resources of each pending task in task list 1 are matched according to the above scheme, the policy resources can be retrieved and cached in advance. When a pending task is processed, the corresponding policy resources can be extracted from the cache space, such as retrieving data resources and allocating human resources, and the execution of the pending task can be completed based on the corresponding policy resources. However, when multiple pending tasks coexist, if the policy resources are cached through the cache space, a larger cache space needs to be built to avoid a large number of policy resources for a single pending task, and the cache efficiency of the cache space also needs to be guaranteed.

[0032] In order to improve the cache efficiency of policy resources, this embodiment adjusts the order of tasks in the task list 1 according to the resource similarity between the policy resources of each task to be processed, which specifically includes the following steps: Extract the policy resources corresponding to each pending task in task list one, calculate the data similarity and manpower similarity of any two pending tasks in task list one according to the policy resources; adjust the order of the pending tasks in task list one according to the data similarity and manpower similarity to obtain task list two.

[0033] It is worth noting that the higher the data similarity of two pending tasks, the closer the corresponding pending tasks are in Task List 2; the higher the manpower similarity, the farther the corresponding pending tasks are in Task List 2. It can also be understood that the higher the data similarity of two pending tasks and the lower the manpower similarity, the closer the two are in the task list.

[0034] Data similarity is used to evaluate the similarity between data resources. The higher the similarity of the data resources corresponding to the two pending tasks, the more overlapped the data required when executing the two pending tasks. If the first pending task caches the overlapping data resources to the cache space when executing, the corresponding overlapping data resources do not need to be released after the pending task is completed, and will be used when the next pending task is executed, which can reduce the workload of the cache space and improve cache efficiency. Data similarity can be calculated using methods such as cosine similarity, Jaccard similarity, and edit distance.

[0035] Human resource similarity is used to evaluate the similarity between human resources. The higher the human resource similarity between two pending tasks, the more overlapped the staff members required to execute the two pending tasks. If the first pending task is being executed, it may not be possible to free up enough qualified staff members to execute the second task. At this time, using human resource similarity as an evaluation indicator to adjust the execution order of pending tasks can improve the timeliness of processing pending tasks. Human resource similarity can be calculated using methods such as Pearson correlation coefficient and cosine similarity.

[0036] The order of the tasks to be processed in task list 1 is adjusted according to the data similarity and the manpower similarity, which specifically includes the following steps: Add pending tasks to the task list and pending tasks The data similarity is marked as , human similarity is marked as ; Among them, when the task list 1 is obtained by sorting the task group, the pending task is the task group; when the task list 1 is not sorted by the task group, the pending task is the pending task; define the similarity influence function ,and ; Similarity influence function The sorting algorithm is used to solve the problem and obtain the sorting result; wherein the sorting algorithm includes a greedy algorithm, a dynamic programming algorithm or a graph theory algorithm; and the order of the tasks to be processed is adjusted according to the sorting result to obtain a task list 2. and Used to distinguish different pending tasks.

[0037] Based on the above solution, you can refer to the following example to adjust the task list: Suppose a software development company has a series of development tasks that need to be assigned to different development teams. We hope to optimize the order of task assignment based on the technology stack (data similarity) and required skill sets (human similarity) required for the tasks to improve team efficiency and project success rate.

[0038] The following task list has been determined: Task Technology stack (data similarity) Skill set (human similarity) T1 Java,MySQL Full stack engineer T2 Python, MongoDB Data Scientist T3 Java, React Front-end Engineer

[0039] Data similarity ( ): Define a simple similarity scoring system based on the matching degree of the technology stacks. For example, if the technology stacks of two tasks match perfectly, the similarity is 1, a partial match is 0.5, and a mismatch is 0.

[0040] Human Similarity ( ): Scoring is done based on how well the required skill sets match. For example, if two tasks require exactly the same skills, the similarity is 1, partial overlap in skills is 0.5, and irrelevant skills are 0.

[0041] Similarity influence function ,in and are weight coefficients, which are used to balance the influence of data similarity and human similarity. and ; According to the above table and the similarity evaluation method, the value of the similarity influence function is calculated as: ; ; ; Choose a greedy algorithm to influence the function based on similarity Sorting tasks. The greedy algorithm will give priority to task pairs with the highest similarity impact function values, that is, tasks with high data similarity and low human similarity should be put together. Based on the sorting results of the greedy algorithm, we get the following task list 2: Task list 2: T1->T3->T2; this order means that we execute T1 and T3 first because their data similarity is the highest and human similarity is the lowest, and then execute T2.

[0042] After determining the task list 2, cache data resources (which may also include human resources) for each pending task in the task list 2. Execute the pending tasks in sequence based on the cached policy resources. If the current pending task is completed, update the cache space according to the policy resources corresponding to the next pending task. Since the data resources of adjacent pending tasks in the task list 2 are highly similar, the cache space only needs to replace part of the data resources, and does not need to completely replace the data resources in the cache space. This does not require a particularly large cache space and can also reduce the data processing volume of the cache space.

[0043] In some other preferred embodiments, when executing the pending tasks, the data resources of the first pending task in the second task list are pre-cached, and when the first processing task is about to be completed, the data resources corresponding to the second pending task and different from the first pending task are cached into the cache space. When processing the second pending task, the data resources can be directly retrieved, and the processing can be carried out in sequence to achieve efficient execution of the pending tasks.

[0044] It should be noted that the policy resources in this embodiment mainly include data resources and human resources. Data resources refer to the data required to complete the tasks to be processed, such as data type, data content, etc. When necessary, the data can also be filtered by built-in filtering rules. Human resources mainly refer to the staff who perform the tasks; if there are not many staff, the human resources can be the unique ID corresponding to the staff, and the unique ID is associated with the skills of the staff; if there are many staff, the human resources can be the ID of the staff at each skill level. When adjusting Task List 2, it is also necessary to comprehensively consider the number of staff at each skill level. For example, if the human resources corresponding to two tasks to be processed are highly similar, but the human resources are very abundant, the human resource similarity can be ignored when calculating the first four-degree influence function. For example, Set to 0.

[0045] In this embodiment, several pending tasks in Task List 1 are verified. For example, after receiving a task, the built-in rules are used to verify whether the task belongs to the task of this example platform. If it meets the requirements, it is marked as a pending task. Of course, other verifications of tasks are also possible. It is worth noting that in other preferred embodiments, the policy resources required by each pending task can also be directly analyzed without generating Task List 1.

[0046] This embodiment first sorts the pending tasks by priority or task relevance to generate a task list 1; then adjusts the task order according to the resource similarity of each pending task in task list 1 to obtain a task list 2; finally, matches policy resources to the pending tasks in task list 2, and executes each pending task based on the policy resources. This embodiment adjusts the task order by resource similarity, which can improve the matching efficiency of data resources, and at the same time reasonably allocates human resources to ensure the timely processing of each pending task.

[0047] Embodiment 2: The difference from Embodiment 1 is that the tasks to be processed are divided into several task groups based on task relevance, and the several task groups are sorted according to their priorities to generate Task List 1; the task order is adjusted according to the resource similarity between the task groups to obtain Task List 2.

[0048] See also Figure 2 , sort several pending tasks according to the sorting rules, including: Identify the priorities and task relevance of a number of pending tasks; divide the number of pending tasks into a number of task groups according to the task relevance, sort the number of task groups according to the priorities, and obtain a task list 1.

[0049] This embodiment refers to both the priority and task relevance sorting principles when sorting a number of pending tasks. Task relevance is used to determine the order of processing between multiple pending tasks. For example, although task A has a lower priority than task B, the execution result of task A is the basis for executing task B, so task A should be placed before task B. Based on task relevance, the pending tasks can be divided into several task groups, and the pending tasks in each task group have been sorted.

[0050] After obtaining several task groups, sort each task group according to their priority. Please refer to the following steps: Extract the task attributes of each task to be processed; quantify the task attributes to obtain attribute tags; calculate the weight of each task to be processed based on the entropy weight method and attribute tags, and perform weighted average of the weight and priority to obtain the priority of each task group; and sort several task groups according to the priority of each task group.

[0051] The task attributes of this embodiment include time requirements, resource quantity, complexity and risk, etc., which are mainly used to measure the importance of the corresponding pending tasks. When calculating the weight of each pending task, each indicator in the task attribute is first standardized to eliminate the influence of dimension and order of magnitude. The standardization method is such as minimum-maximum standardization and Z-score standardization. The standardization formula is: ;in, For the The standardized value of the indicator, For the The original value of each indicator (obtained by quantifying each indicator in the task attributes, such as expert scoring, experience scoring, etc.), and Respectively The maximum and minimum values ​​of the indicators.

[0052] Then, the entropy weight method is used to calculate the information entropy of each indicator, and then the weight of each task is calculated. According to the formula ,in, It should be noted that the entropy weight method is an existing mature method, and its specific calculation process is not described here. In addition to the entropy weight method, the priority of the task group can also be calculated by methods such as hierarchical analysis method and weighted average model.

[0053] After obtaining the priority of each task group, each task group is sorted according to the priority to obtain task list 1. In other preferred embodiments, the average priority of the tasks to be processed in each task group can also be used as the priority of the task group to complete the sorting of the task groups.

[0054] When adjusting the task list 1 of this example, refer to the following steps: Extract the policy resources corresponding to each task group in task list one, and calculate the data similarity and manpower similarity of any two task groups in task list one according to the policy resources; adjust the order of the task groups in task list one according to the data similarity and manpower similarity to obtain task list two; among which, the higher the data similarity, the closer the corresponding task groups are in task list two; the higher the manpower similarity, the farther the corresponding task groups are in task list two.

[0055] Group the tasks in the task list and Task Force The data similarity is marked as , human similarity is marked as ; Define similarity influence function ,and ; Similarity influence function and sorting algorithms to obtain sorting results; wherein the sorting algorithms include greedy algorithms, dynamic programming algorithms or graph theory algorithms, and is the weight coefficient; the order of the task groups is adjusted according to the sorting result to obtain task list 2. The calculation process is similar to the adjustment of task list 1 according to the data similarity and manpower similarity of the tasks to be processed in embodiment 1, and the calculation method in embodiment 1 can be referred to, which will not be repeated here.

[0056] This embodiment first divides a number of pending tasks into a number of task groups according to task relevance, calculates the priority of each task group according to the task attributes of the pending tasks in each task group and the entropy weight method, sorts the task groups according to the priority to obtain task list 1; and adjusts task list 1 according to the resource similarity between task groups to obtain task list 2. This embodiment matches policy resources in the form of task groups, not only relying on priority to sort task groups, but also comprehensively considering the correlation between each pending task, which can improve the matching efficiency of policy resources and task processing efficiency.

[0057] See also Figure 3 , the second aspect of the present application provides a policy data matching analysis system for task management, including: a resource matching module, and a task processing module connected thereto; Task processing module: used for receiving a number of tasks to be processed; sorting the number of tasks to be processed to obtain a task list 1, and adjusting the task list 1 to obtain a task list 2; Resource matching module: used to analyze the strategic resources required by each pending task in task list one, and match the strategic resources to the pending tasks in task list two based on the strategic resources required; and, Execute the corresponding pending tasks according to the matching policy resources.

[0058] A cache space is provided in the resource matching module, which is equivalent to a temporary storage area for data and is used to cache data resources required for executing tasks to be processed.

[0059] The above embodiments are only used to illustrate the technical method of the present application and are not intended to limit it. Although the present application has been described in detail with reference to the preferred embodiments, a person of ordinary skill in the art should understand that the technical method of the present application may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present application.

Claims

1. A strategy data matching analysis method applied to task management, characterized in that: include: Receiving a number of pending tasks; sorting the pending tasks according to a sorting rule to generate a task list 1; wherein the sorting rule is set based on task priority or task relevance; Analyze the strategic resources required by each of the pending tasks in the task list 1; adjust the task list 1 according to the similarity of each resource in the strategic resources to obtain a task list 2; wherein the strategic resources include data resources and human resources; Based on the demand, the policy resources are matched with the policy resources for each of the pending tasks in the second task list in turn, and the corresponding pending tasks are executed according to the matched policy resources.

2. A strategy data matching analysis method for task management according to claim 1, characterized in that: Analyze the strategic resources required by each of the pending tasks in the task list 1, including: Collecting historical task data; performing data cleaning and feature extraction on the historical task data to obtain basic data; wherein the historical task data includes task description, required resource type and task result; Performing correlation analysis on the task features and resource types in the basic data, and constructing model training data based on the analysis results; training the machine learning model constructed by the model training data, and marking the machine learning model as a resource matching model after verification and testing; wherein the machine learning model includes a decision tree, a random forest or a neural network model; The task characteristics of each task to be processed in the task list 1 are input into the resource matching model to obtain the corresponding strategic resources.

3. A strategy data matching analysis method for task management according to claim 2, characterized in that: Sorting the plurality of pending tasks according to the sorting rules, including: Identify the priorities and task relevance of the plurality of tasks to be processed; wherein the task relevance is used to determine the order in which the plurality of tasks to be processed are processed; The plurality of tasks to be processed are divided into a plurality of task groups according to the task relevance, and the plurality of task groups are sorted according to the priority to obtain a task list 1.

4. A strategy data matching analysis method for task management according to claim 3, characterized in that: Sorting the task groups according to the priority level includes: Extracting the task attributes of each of the tasks to be processed; quantifying the task attributes to obtain attribute tags; wherein the task attributes include time requirements, resource quantities, complexity and risks; The weight of each of the tasks to be processed is calculated based on the entropy weight method and the attribute label, and the priority of each of the task groups is obtained by weighted average of the weight and the priority; The order of the task groups is achieved by the priority of each task group.

5. The strategy data matching and analysis method for task management according to claim 2, characterized in that: The task list 1 is adjusted according to the similarity of each resource in the strategy resource, including: Extracting the policy resources corresponding to each pending task in the task list 1, and calculating the data similarity and manpower similarity of any two pending tasks in the task list 1 according to the policy resources; The order of the tasks to be processed in the task list one is adjusted according to the data similarity and the manpower similarity to obtain the task list two; wherein, the higher the data similarity, the closer the corresponding tasks to be processed are in the task list two; and the higher the manpower similarity, the farther the corresponding tasks to be processed are in the task list two.

6. A strategy data matching analysis method for task management according to claim 4, characterized in that: The task list 1 is adjusted according to the similarity of each resource in the strategy resource, including: Extracting the policy resources corresponding to each task group in the task list one, and calculating the data similarity and manpower similarity of any two task groups in the task list one according to the policy resources; The order of the task groups in the task list one is adjusted according to the data similarity and the manpower similarity to obtain the task list two; wherein, the higher the data similarity, the closer the corresponding task groups are in the task list two; and the higher the manpower similarity, the farther the corresponding task groups are in the task list two.

7. A strategy data matching analysis method for task management according to claim 5 or 6, characterized in that: The task list 2 is obtained, including: The target object in the task list and target audience The data similarity is marked as , human similarity is marked as ; Among them, when the task list 1 is obtained by sorting the task group, the target object is the task group; when the task list 1 is not obtained by sorting the task group, the target object is the task to be processed; Define the similarity influence function ,and ; Similarity influence function and sorting algorithms to obtain sorting results; wherein the sorting algorithms include greedy algorithms, dynamic programming algorithms or graph theory algorithms, and is the weight coefficient; The order of the target objects is adjusted according to the sorting result to obtain a second task list.

8. A strategy data matching and analysis system for task management, used to execute a strategy data matching and analysis method for task management as claimed in any one of claims 1 to 6, characterized in that: include: Resource matching module, and the task processing module connected to it; Task processing module: used for receiving a number of tasks to be processed; sorting the tasks to be processed to obtain a task list 1, and adjusting the task list 1 to obtain a task list 2; Resource matching module: used for analyzing the policy resources required by each task to be processed in the task list 1, and matching the policy resources with the tasks to be processed in the task list 2 based on the policy resources required; and, The corresponding task to be processed is executed according to the matched policy resources.

Images