A big data analysis management system and method based on intelligent cloud platform

Abstract

The present invention discloses a big data analysis management system and method based on an intelligent cloud platform, and belongs to the field of data analysis technology. The system of the present invention comprises: a data collection module, a task difficulty assessment module, an execution object efficiency assessment module, a correlation analysis module and a real-time task allocation module; the data collection module collects historical task data and execution object data from the intelligent cloud platform; the task difficulty assessment module performs task difficulty assessment on the historical task data and generates the task difficulty level of the historical task data; the execution object efficiency assessment module analyzes the execution object data and generates the efficiency level of the execution object; the correlation analysis module compares and analyzes the task difficulty level of the historical task data and the efficiency level of the execution object to obtain the correlation relationship model between the task event and the execution object; the real-time task allocation module receives the real-time task data and allocates the corresponding execution object to the real-time task data.

Description

Technical Field

[0001] The present invention relates to the field of data analysis technology, and specifically to a big data analysis management system and method based on an intelligent cloud platform. Background Art

[0002] In today's digital age, big data analysis plays a vital role in all walks of life. As the core infrastructure for data storage and computing, cloud platforms are responsible for more and more data processing tasks. However, there are some challenges and limitations in the existing cloud platform big data analysis management system.

[0003] First, with regard to the problem of task allocation for background execution objects, the existing technology often only determines the task allocation results based on the level of the execution object or randomly, but ignores the relationship between different tasks and execution objects of different levels. This judgment method may cause the execution objects to be assigned unskilled tasks, thereby making the system's task execution efficiency low; secondly, the existing technology lacks intelligent task scheduling, which may lead to unbalanced task allocation. Some execution objects may be overloaded or the task execution time may be too long, affecting the overall data processing efficiency. Summary of the invention

[0004] The purpose of the present invention is to provide a big data analysis management system and method based on an intelligent cloud platform to solve the problems raised in the above-mentioned background technology.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] A big data analysis and management method based on an intelligent cloud platform, the method comprising the following steps:

[0007] S100. Collect historical task data from the intelligent cloud platform, divide the historical task data into several task events according to the selected time period, arrange and number them in chronological order to form a historical task data set; perform task difficulty evaluation on each element in the historical task data set to obtain a task difficulty level;

[0008] S200. Obtain the execution object data of the intelligent cloud platform, divide the execution object data according to the division method of the historical task data, and form an execution object data set; perform efficiency evaluation on each element in the execution object data set to obtain the execution object efficiency level;

[0009] S300. Compare and analyze the task difficulty level and execution object efficiency level of each element in the historical task data set and the execution object data set in order to obtain the correlation between the two, thereby obtaining the corresponding relationship between the task event and the execution object;

[0010] S400. Receive real-time task data, calculate similarity between the real-time task data and historical task data, combine the similarity calculation result with the correspondence between task events and execution objects, and assign corresponding execution objects to each task corresponding to the real-time task data.

[0011] Furthermore, step S100 includes:

[0012] S101. The historical task data refers to the record of tasks completed on the intelligent cloud platform in the past, including task description, execution time, execution steps and execution interface; the task description is a text description of the task content and requirements, including the task objectives, specific requirements and required skills; the historical task data is divided according to the selected time period to obtain a number of task events, and each task event is independent and completed by an execution object; all task events are numbered in chronological order to form a historical task data set A, and A={a1,a2,...,an}, where a1 represents the first historical task event, a2 represents the second historical task event, and so on, an represents the nth historical task event, and n represents the historical task event number, which is a positive integer; when there are historical task events with the same execution time in the historical task event set, the sorting number of the corresponding historical task event is randomly determined;

[0013] S102. Obtain the task description of each historical task event in the historical task data set A, extract keywords, and compare them with the preset task difficulty keywords, and record the number of times the preset task difficulty keywords appear m; according to the execution time of the historical task event, obtain the completion time length T of each historical task event in the historical task data set A; obtain the execution steps and execution interface of each historical task event, and record the number of execution steps M and the number of execution interfaces N;

[0014] S103. Normalize the number of times m, the completion time length T, the number of execution steps M, and the number of execution interfaces N corresponding to each element in the historical task data set A, respectively, to obtain the feature vector V1 corresponding to each element in the historical task data set A, and V1=[v1,v2,v3,v4], where v1 represents the number of times m, the preset task difficulty keyword corresponding to each element in the historical task data set A, v2 represents the completion time length T corresponding to each element in the historical task data set A, v3 represents the number M of execution steps corresponding to each element in the historical task data set A, and v4 represents the number N of execution interfaces corresponding to each element in the historical task data set A;

[0015] S104. Mark the feature vector V1 corresponding to each element in the historical task data set A as data points on the radar chart in order, and the radar chart includes four axes with the same starting point, namely k1, k2, k3 and k4, and k1 corresponds to the value of v1, k2 corresponds to the value of v2, k3 corresponds to the value of v3, and k4 corresponds to the value of v4; connect the data points of two adjacent axes on the radar chart to form a quadrilateral, calculate the radar chart area S corresponding to each element in the historical task data set A, and thus obtain the task difficulty score set B corresponding to the historical task data set A, and B={S1,S2,...,Sn}, wherein S1 represents the radar chart area S of the first historical task event, S2 represents the radar chart area S of the second historical task event, and so on, Sn represents the radar chart area S of the nth historical task event; the elements with the same number in the historical task data set A and the task difficulty score set B correspond one to one, and the task difficulty of the historical task event is determined by comparing the sizes of the elements in the task difficulty score set B of the historical task event, and one radar chart area S represents a task difficulty level.

[0016] By analyzing multiple aspects such as keyword extraction, completion time, number of execution steps and number of execution interfaces of historical task data, a comprehensive task difficulty assessment index is formed. Compared with traditional methods that only rely on subjective judgment or evaluation of a few indicators, the present invention can evaluate task difficulty more objectively; by converting historical task data into feature vectors and displaying them on a radar chart, the differences between different historical task events can be intuitively compared. Based on the task difficulty score set, tasks and execution objects can be accurately matched to ensure that each task is handled by the most suitable execution object.

[0017] Further, step S200 includes:

[0018] S201. The execution object data refers to the relevant information of the execution object of the intelligent cloud platform, including the execution object number, task proficiency index and the completion status of historical task events; the task proficiency index refers to the ratio of the target completion time of the historical task event to the actual completion time; the execution object data is divided according to the execution object number to obtain the execution object data set C, and C={c1,c2,...,ct}, where c1 represents the first execution object data, c2 represents the second execution object data, and so on, ct represents the tth execution object data, and t represents the execution object data number, which is a positive integer;

[0019] S202. For each element in the execution object data set, extract the data information of task satisfaction to obtain the task proficiency index Q; and extract the completion status of historical task events, calculate the deviation between the completion status data E1 of the historical task events and the preset target data E2 to obtain the completion index W. The specific calculation formula is:

[0020] ,

[0021] Wherein, S(E1, E2) represents the similarity function between the completion status data E1 and the preset target data E2;

[0022] S203. The task proficiency index Q and the completion index W of each element in the execution object data set are combined into a feature vector V2, and V2=[Q,W]; each data point of the feature vector V2 is marked in the order of the execution object data set in a plane rectangular coordinate system, and the horizontal axis of the plane rectangular coordinate system represents the task proficiency index Q, and the vertical axis represents the completion index W, and a bar chart is obtained in the plane rectangular coordinate system, so as to evaluate the efficiency of each element in the execution object data set and obtain an efficiency score set X, and each element in the efficiency score set X has: X=Q*W, and one efficiency score corresponds to one execution object efficiency level.

[0023] By extracting the task proficiency index and historical task completion data and converting them into feature vectors, a bar chart is drawn in a plane rectangular coordinate system for efficiency evaluation, which is conducive to objectively evaluating the work performance and efficiency level of each execution object; by analyzing the efficiency score set X of each element in the execution object data set, personalized management suggestions and guidance can be provided for different execution objects to help them improve their work efficiency and task completion quality; based on each element in the efficiency score set X (X=Q*W), management strategies and decisions can be formulated more scientifically, resource allocation can be optimized, and overall work efficiency and management level can be improved.

[0024] Furthermore, step S300 includes:

[0025] S301. Match each element in the task difficulty score set B with each element in the efficiency score set X according to the historical task events, thereby obtaining the data point coordinates (Si, Xi), where i is the historical task event number, which is a positive integer;

[0026] S302. Mark the data point coordinates (Si, Xi) of the selected time period in a plane rectangular coordinate system in chronological order, and the horizontal axis of the plane rectangular coordinate system is the task difficulty score, and the vertical axis is the efficiency score, so as to obtain a scatter plot of the task difficulty score and the efficiency score; draw a scatter plot of the task difficulty score and the efficiency score corresponding to all historical task events of the same time length, and analyze the distribution of data points in each scatter plot to determine the type of correlation between the task difficulty score and the efficiency score; if the data points are linearly distributed, draw the best fitting regression line in the scatter plot; if they are curvedly distributed, then count the corresponding proportions between the task difficulty score and the efficiency score in the scatter plot; through the above analysis, obtain the correlation model G between the task difficulty score and the efficiency score, and input the historical task data into the correlation model G to obtain the corresponding relationship between the task event and the execution object.

[0027] By matching the task difficulty score set and the efficiency score set according to historical task events and showing their relationship in a scatter plot, we can conduct a more in-depth analysis and exploration of the correlation between the task difficulty score and the efficiency score; by statistically analyzing the scatter plot of the task difficulty score and the efficiency score, we can establish a correlation model G between the task difficulty score and the efficiency score, which helps to understand the impact of task difficulty on the efficiency of the execution object and provide a more scientific basis for management decisions.

[0028] Furthermore, step S400 includes:

[0029] S401 receives real-time task data, divides the real-time task data according to the division method of historical task data, obtains several real-time task events, extracts the corresponding feature vector V1' for each real-time task event, and represents each data point of the feature vector V1' on the radar chart;

[0030] S402. Calculate the similarity between the radar chart of the real-time task event and the radar chart of all historical task events, select the historical task event with the greatest similarity as the matching result, and input the matched historical task event into the association relationship model, and output the corresponding execution object level; obtain the corresponding execution object number in real time, determine the idle status of the execution object, issue a real-time task event to the idle execution object, and let the execution object process the real-time task event.

[0031] A big data analysis and management system based on an intelligent cloud platform, the system includes: a data collection module, a task difficulty assessment module, an execution object efficiency assessment module, a correlation relationship analysis module and a real-time task allocation module;

[0032] The data collection module collects historical task data and execution object data from the intelligent cloud platform and performs preprocessing; the task difficulty assessment module assesses the task difficulty of the historical task data, analyzes it according to the task description, execution time, execution steps and execution interface, and generates the task difficulty level of the historical task data; the execution object efficiency assessment module analyzes the task satisfaction of the execution object and the completion status of historical task events, and generates the efficiency level of the execution object; the association analysis module compares and analyzes the task difficulty level of the historical task data and the efficiency level of the execution object, and obtains the association relationship model between the task event and the execution object; the real-time task allocation module receives the real-time task data, calculates the similarity between the real-time task data and the historical data, and allocates the corresponding execution object to the real-time task data based on the results of the association analysis module.

[0033] Further, the data collection module includes a historical task data collection unit and an execution object data collection unit;

[0034] The historical task data collection unit is responsible for collecting historical task data from the intelligent cloud platform and dividing it into several task events according to the selected time period to form a historical task data set; the execution object data collection unit obtains execution object data from the intelligent cloud platform, divides the execution object data according to the division method of historical task data, and forms an execution object data set.

[0035] Furthermore, the task difficulty assessment module includes a task description extraction and keyword matching unit, a feature vector generation and normalization unit, and a difficulty score calculation unit;

[0036] The task description extraction and keyword matching unit extracts the task description in the historical task data, extracts keywords and compares them with the preset task difficulty keywords, and records the number of times the preset task difficulty keywords appear; the feature vector generation and normalization unit normalizes the task difficulty related features of each element in the historical task data set, generates a feature vector and labels it; the difficulty score calculation unit generates a radar chart based on the feature vector, calculates the radar chart area corresponding to each element in the historical task data set, and obtains a task difficulty score set.

[0037] Furthermore, the execution object efficiency evaluation module includes a task proficiency index extraction unit, a completion index calculation unit, and an efficiency score calculation unit;

[0038] The task proficiency index extraction unit extracts the task proficiency index from the execution object data set; the completion index calculation unit calculates the deviation between the completion status data of the historical task event and the preset target data to obtain the completion index; the efficiency score calculation unit forms a feature vector based on the task proficiency index and the completion index, performs efficiency evaluation and obtains an efficiency score set.

[0039] Furthermore, the association relationship analysis module includes a data point coordinate generation unit and an association relationship type analysis unit;

[0040] The data point coordinate generation unit matches the task difficulty score and the efficiency score according to the historical task events to obtain the data point coordinates; the association relationship type analysis unit marks the data points in the plane rectangular coordinate system to obtain a scatter plot of the task difficulty score and the efficiency score, and determines the association relationship type by analyzing the distribution of the data points;

[0041] The real-time task allocation module includes a real-time task feature extraction unit and a similarity calculation and matching unit;

[0042] The real-time task feature extraction unit receives real-time task data and extracts the corresponding feature vector; the similarity calculation and matching unit calculates the similarity between the feature vector of the real-time task event and the historical task event, selects the historical task event with the greatest similarity as the matching result, outputs the corresponding execution object level, determines the idle status of the execution object and performs real-time task allocation.

[0043] Compared with the prior art, the beneficial effects achieved by the present invention are: the system can intelligently perform task allocation and execution object matching based on the correlation between historical task data, execution object data and real-time task data, thereby improving task execution efficiency and resource utilization; the method establishes an association model between task difficulty scores and execution object efficiency scores by analyzing historical task data and execution object data, which can more accurately match tasks and execution objects; at the same time, when real-time task data arrives, the system can intelligently select the most appropriate execution object for task processing based on the similarity calculation results of historical data, thereby improving the efficiency and quality of task execution. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0045] Figure 1 This is a schematic diagram of a big data analysis and management system module based on an intelligent cloud platform of the present invention;

[0046] Figure 2 It is a historical task data radar chart of a big data analysis management system and method based on an intelligent cloud platform of the present invention. DETAILED DESCRIPTION

[0047] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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 creative work are within the scope of protection of the present invention.

[0048] See also Figure 1 , the present invention provides a technical solution:

[0049] A big data analysis and management system based on an intelligent cloud platform, the system includes: a data collection module, a task difficulty assessment module, an execution object efficiency assessment module, a correlation relationship analysis module and a real-time task allocation module;

[0050] The data collection module collects historical task data and execution object data from the intelligent cloud platform and performs preprocessing; the task difficulty assessment module assesses the task difficulty of the historical task data, analyzes it according to the task description, execution time, execution steps and execution interface, and generates the task difficulty level of the historical task data; the execution object efficiency assessment module analyzes the task satisfaction of the execution object and the completion status of historical task events, and generates the efficiency level of the execution object; the association analysis module compares and analyzes the task difficulty level of the historical task data and the efficiency level of the execution object, and obtains the association relationship model between the task event and the execution object; the real-time task allocation module receives the real-time task data, calculates the similarity between the real-time task data and the historical data, and allocates the corresponding execution object to the real-time task data based on the results of the association analysis module.

[0051] The data collection module includes a historical task data collection unit and an execution object data collection unit;

[0052] The historical task data collection unit is responsible for collecting historical task data from the intelligent cloud platform and dividing it into several task events according to the selected time period to form a historical task data set; the execution object data collection unit obtains execution object data from the intelligent cloud platform, divides the execution object data according to the division method of historical task data, and forms an execution object data set.

[0053] The task difficulty assessment module includes a task description extraction and keyword matching unit, a feature vector generation and normalization unit, and a difficulty score calculation unit;

[0054] The task description extraction and keyword matching unit extracts the task description in the historical task data, extracts keywords and compares them with the preset task difficulty keywords, and records the number of times the preset task difficulty keywords appear; the feature vector generation and normalization unit normalizes the task difficulty related features of each element in the historical task data set, generates a feature vector and labels it; the difficulty score calculation unit generates a radar chart based on the feature vector, calculates the radar chart area corresponding to each element in the historical task data set, and obtains a task difficulty score set.

[0055] The execution object efficiency evaluation module includes a task proficiency index extraction unit, a completion index calculation unit and an efficiency score calculation unit;

[0056] The task proficiency index extraction unit extracts the task proficiency index from the execution object data set; the completion index calculation unit calculates the deviation between the completion status data of the historical task event and the preset target data to obtain the completion index; the efficiency score calculation unit forms a feature vector based on the task proficiency index and the completion index, performs efficiency evaluation and obtains an efficiency score set.

[0057] The association relationship analysis module includes a data point coordinate generation unit and an association relationship type analysis unit;

[0058] The data point coordinate generation unit matches the task difficulty score and the efficiency score according to the historical task events to obtain the data point coordinates; the association relationship type analysis unit marks the data points in the plane rectangular coordinate system to obtain a scatter plot of the task difficulty score and the efficiency score, and determines the association relationship type by analyzing the distribution of the data points;

[0059] The real-time task allocation module includes a real-time task feature extraction unit and a similarity calculation and matching unit;

[0060] The real-time task feature extraction unit receives real-time task data and extracts the corresponding feature vector; the similarity calculation and matching unit calculates the similarity between the feature vector of the real-time task event and the historical task event, selects the historical task event with the greatest similarity as the matching result, outputs the corresponding execution object level, determines the idle status of the execution object and performs real-time task allocation.

[0061] A big data analysis and management method based on an intelligent cloud platform, the method comprising the following steps:

[0062] S100. Collect historical task data from the intelligent cloud platform, divide the historical task data into several task events according to the selected time period, arrange and number them in chronological order to form a historical task data set; perform task difficulty evaluation on each element in the historical task data set to obtain a task difficulty level;

[0063] S200. Obtain the execution object data of the intelligent cloud platform, divide the execution object data according to the division method of the historical task data, and form an execution object data set; perform efficiency evaluation on each element in the execution object data set to obtain the execution object efficiency level;

[0064] S300. Compare and analyze the task difficulty level and execution object efficiency level of each element in the historical task data set and the execution object data set in order to obtain the correlation between the two, thereby obtaining the corresponding relationship between the task event and the execution object;

[0065] S400. Receive real-time task data, calculate similarity between the real-time task data and historical task data, combine the similarity calculation result with the correspondence between task events and execution objects, and assign corresponding execution objects to each task corresponding to the real-time task data.

[0066] Step S100 includes:

[0067] S101. The historical task data refers to the record of tasks completed on the intelligent cloud platform in the past, including task description, execution time, execution steps and execution interface; the task description is a text description of the task content and requirements, including the task objectives, specific requirements and required skills; the historical task data is divided according to the selected time period to obtain a number of task events, and each task event is independent and completed by an execution object; all task events are numbered in chronological order to form a historical task data set A, and A={a1,a2,...,an}, where a1 represents the first historical task event, a2 represents the second historical task event, and so on, an represents the nth historical task event, and n represents the historical task event number, which is a positive integer; when there are historical task events with the same execution time in the historical task event set, the sorting number of the corresponding historical task event is randomly determined;

[0068] S102. Obtain the task description of each historical task event in the historical task data set A, extract keywords, and compare them with the preset task difficulty keywords, and record the number of times the preset task difficulty keywords appear m; according to the execution time of the historical task event, obtain the completion time length T of each historical task event in the historical task data set A; obtain the execution steps and execution interface of each historical task event, and record the number of execution steps M and the number of execution interfaces N;

[0069] S103. Normalize the number of times m, the completion time length T, the number of execution steps M, and the number of execution interfaces N corresponding to each element in the historical task data set A, respectively, to obtain the feature vector V1 corresponding to each element in the historical task data set A, and V1=[v1,v2,v3,v4], where v1 represents the number of times m, the preset task difficulty keyword corresponding to each element in the historical task data set A, v2 represents the completion time length T corresponding to each element in the historical task data set A, v3 represents the number M of execution steps corresponding to each element in the historical task data set A, and v4 represents the number N of execution interfaces corresponding to each element in the historical task data set A;

[0070] S104. Mark the feature vector V1 corresponding to each element in the historical task data set A as data points on the radar chart in order, and the radar chart includes four axes with the same starting point, namely k1, k2, k3 and k4, and k1 corresponds to the value of v1, k2 corresponds to the value of v2, k3 corresponds to the value of v3, and k4 corresponds to the value of v4; connect the data points of two adjacent axes on the radar chart to form a quadrilateral, calculate the radar chart area S corresponding to each element in the historical task data set A, and thus obtain the task difficulty score set B corresponding to the historical task data set A, and B={S1,S2,...,Sn}, wherein S1 represents the radar chart area S of the first historical task event, S2 represents the radar chart area S of the second historical task event, and so on, Sn represents the radar chart area S of the nth historical task event; the elements with the same number in the historical task data set A and the task difficulty score set B correspond one to one, and the task difficulty of the historical task event is determined by comparing the sizes of the elements in the task difficulty score set B of the historical task event, and one radar chart area S represents a task difficulty level.

[0071] By analyzing multiple aspects such as keyword extraction, completion time, number of execution steps and number of execution interfaces of historical task data, a comprehensive task difficulty evaluation index is formed. Compared with the traditional method that only relies on subjective judgment or a few indicators for evaluation, the present invention can evaluate the task difficulty more objectively; by converting historical task data into feature vectors and displaying them on radar charts, such as Figure 2 As shown, the differences between different historical task events can be intuitively compared. Based on the task difficulty score set, tasks and execution objects can be accurately matched to ensure that each task is handled by the most suitable execution object.

[0072] Step S200 includes:

[0073] S201. The execution object data refers to the relevant information of the execution object of the intelligent cloud platform, including the execution object number, task proficiency index and the completion status of historical task events; the task proficiency index refers to the ratio of the target completion time of the historical task event to the actual completion time; the execution object data is divided according to the execution object number to obtain the execution object data set C, and C={c1,c2,...,ct}, where c1 represents the first execution object data, c2 represents the second execution object data, and so on, ct represents the tth execution object data, and t represents the execution object data number, which is a positive integer;

[0074] S202. For each element in the execution object data set, extract the data information of task satisfaction to obtain the task proficiency index Q; and extract the completion status of historical task events, calculate the deviation between the completion status data E1 of the historical task events and the preset target data E2 to obtain the completion index W. The specific calculation formula is:

[0075] ,

[0076] Wherein, S(E1, E2) represents the similarity function between the completion status data E1 and the preset target data E2;

[0077] S203. The task proficiency index Q and the completion index W of each element in the execution object data set are combined into a feature vector V2, and V2=[Q,W]; each data point of the feature vector V2 is marked in the order of the execution object data set in a plane rectangular coordinate system, and the horizontal axis of the plane rectangular coordinate system represents the task proficiency index Q, and the vertical axis represents the completion index W, and a bar chart is obtained in the plane rectangular coordinate system, so as to evaluate the efficiency of each element in the execution object data set and obtain an efficiency score set X, and each element in the efficiency score set X has: X=Q*W, and one efficiency score corresponds to one execution object efficiency level.

[0078] By extracting the task proficiency index and historical task completion data and converting them into feature vectors, a bar chart is drawn in a plane rectangular coordinate system for efficiency evaluation, which is conducive to objectively evaluating the work performance and efficiency level of each execution object; by analyzing the efficiency score set X of each element in the execution object data set, personalized management suggestions and guidance can be provided for different execution objects to help them improve their work efficiency and task completion quality; based on each element in the efficiency score set X (X=Q*W), management strategies and decisions can be formulated more scientifically, resource allocation can be optimized, and overall work efficiency and management level can be improved.

[0079] Step S300 includes:

[0080] S301. Match each element in the task difficulty score set B with each element in the efficiency score set X according to the historical task events, thereby obtaining the data point coordinates (Si, Xi), where i is the historical task event number, which is a positive integer;

[0081] S302. Mark the data point coordinates (Si, Xi) of the selected time period in a plane rectangular coordinate system in chronological order, and the horizontal axis of the plane rectangular coordinate system is the task difficulty score, and the vertical axis is the efficiency score, so as to obtain a scatter plot of the task difficulty score and the efficiency score; draw a scatter plot of the task difficulty score and the efficiency score corresponding to all historical task events of the same time length, and analyze the distribution of data points in each scatter plot to determine the type of correlation between the task difficulty score and the efficiency score; if the data points are linearly distributed, draw the best fitting regression line in the scatter plot; if they are curvedly distributed, then count the corresponding proportions between the task difficulty score and the efficiency score in the scatter plot; through the above analysis, obtain the correlation model G between the task difficulty score and the efficiency score, and input the historical task data into the correlation model G to obtain the corresponding relationship between the task event and the execution object.

[0082] By matching the task difficulty score set and the efficiency score set according to historical task events and showing their relationship in a scatter plot, we can conduct a more in-depth analysis and exploration of the correlation between the task difficulty score and the efficiency score; by statistically analyzing the scatter plot of the task difficulty score and the efficiency score, we can establish a correlation model G between the task difficulty score and the efficiency score, which helps to understand the impact of task difficulty on the efficiency of the execution object and provide a more scientific basis for management decisions.

[0083] In this embodiment, it is assumed that there is a set of historical task event data, which includes a task difficulty score set B and an efficiency score set X. The specific data is shown in Table 1:

[0084]

[0085] Table 1

[0086] First, the task difficulty score and efficiency score are matched according to historical task events to obtain the data point coordinates (Si, Xi), and are marked in the plane rectangular coordinate system; in the scatter plot, the horizontal axis represents the task difficulty score and the vertical axis represents the efficiency score; by observing the distribution of data points in the scatter plot, the type of correlation relationship between the task difficulty score and the efficiency score can be analyzed.

[0087] Assuming that in this example, the data points are linearly distributed, the best fit regression line can be drawn in the scatter plot to represent the linear correlation between the task difficulty score and the efficiency score, thereby obtaining the association relationship model G. Assume that G(Si)=g(Xi), which represents the execution object number corresponding to the input task difficulty score and the output efficiency score.

[0088] Step S400 includes:

[0089] S401 receives real-time task data, divides the real-time task data according to the division method of historical task data, obtains several real-time task events, extracts the corresponding feature vector V1' for each real-time task event, and represents each data point of the feature vector V1' on the radar chart;

[0090] S402. Calculate the similarity between the radar chart of the real-time task event and the radar chart of all historical task events, select the historical task event with the greatest similarity as the matching result, and input the matched historical task event into the association relationship model, and output the corresponding execution object level; obtain the corresponding execution object number in real time, determine the idle status of the execution object, issue a real-time task event to the idle execution object, and let the execution object process the real-time task event.

[0091] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0092] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A big data analysis and management method based on an intelligent cloud platform, characterized in that: The method comprises the following steps: S100. Collect historical task data from the intelligent cloud platform, divide the historical task data into several task events according to the selected time period, arrange and number them in chronological order to form a historical task data set; perform task difficulty evaluation on each element in the historical task data set to obtain a task difficulty level; S200. Obtain the execution object data of the intelligent cloud platform, divide the execution object data according to the division method of the historical task data, and form an execution object data set; perform efficiency evaluation on each element in the execution object data set to obtain the execution object efficiency level; The step S200 includes: S201. The execution object data refers to the relevant information of the execution object of the intelligent cloud platform, including the execution object number, task proficiency index and the completion status of historical task events; the task proficiency index refers to the ratio of the target completion time of the historical task event to the actual completion time; the execution object data is divided according to the execution object number to obtain the execution object data set C, and C={c1,c2,...,ct}, where c1 represents the first execution object data, c2 represents the second execution object data, and so on, ct represents the tth execution object data, and t represents the execution object data number, which is a positive integer; S202. For each element in the execution object data set, extract the task proficiency index Q; and extract the completion status of historical task events, calculate the deviation between the completion status data E1 of the historical task events and the preset target data E2, and obtain the completion index W. The specific calculation formula is: ， Wherein, S(E1, E2) represents the similarity function between the completion status data E1 and the preset target data E2; S203. The task proficiency index Q and the completion index W of each element in the execution object data set are combined into a feature vector V2, and V2=[Q,W]; each data point of the feature vector V2 is marked in a plane rectangular coordinate system according to the order of the execution object data set, and the horizontal axis of the plane rectangular coordinate system represents the task proficiency index Q, and the vertical axis represents the completion index W, and a bar chart is obtained in the plane rectangular coordinate system, so as to evaluate the efficiency of each element in the execution object data set and obtain an efficiency score set X, and each element in the efficiency score set X has: X=Q*W, and one efficiency score corresponds to one execution object efficiency level; S300. Compare and analyze the task difficulty level and execution object efficiency level of each element in the historical task data set and the execution object data set in order to obtain the correlation between the two, thereby obtaining the corresponding relationship between the task event and the execution object; The step S300 includes: S301. Match each element in the task difficulty score set B with each element in the efficiency score set X according to the historical task events, thereby obtaining the data point coordinates (Si, Xi), where i is the historical task event number, which is a positive integer; S302. The coordinates of the data points (Si, Xi) of the selected time period are marked in a plane rectangular coordinate system in chronological order, and the horizontal axis of the plane rectangular coordinate system is the task difficulty score, and the vertical axis is the efficiency score, so as to obtain a scatter plot of the task difficulty score and the efficiency score; draw a scatter plot of the task difficulty score and the efficiency score corresponding to all historical task events of the same time length, and analyze the distribution of data points in each scatter plot to determine the type of correlation relationship between the task difficulty score and the efficiency score; if the data points are linearly distributed, draw the best fitting regression line in the scatter plot; if they are curvedly distributed, count the corresponding proportions between the task difficulty score and the efficiency score in the scatter plot; obtain the correlation relationship model G between the task difficulty score and the efficiency score through the above analysis, and input the historical task data into the correlation relationship model G to obtain the corresponding relationship between the task event and the execution object; S400 receives real-time task data, calculates similarity between the real-time task data and the historical task data, combines the similarity calculation result with the correspondence between the task event and the execution object, and assigns a corresponding execution object to each task corresponding to the real-time task data; The step S400 includes: S401. Receive real-time task data, divide the real-time task data according to the division method of historical task data, obtain several real-time task events, extract the corresponding feature vector V1' for each real-time task event, and represent each data point of the feature vector V1' on the radar chart; S402. Calculate the similarity between the radar chart of the real-time task event and the radar chart of all historical task events, select the historical task event with the greatest similarity as the matching result, and input the matched historical task event into the association relationship model, and output the corresponding execution object level; obtain the corresponding execution object number in real time, determine the idle status of the execution object, issue a real-time task event to the idle execution object, and let the execution object process the real-time task event.

2. The big data analysis and management method based on the intelligent cloud platform according to claim 1 is characterized by: The step S100 includes: S101. The historical task data refers to the record of tasks completed on the intelligent cloud platform in the past, including task description, execution time, execution steps and execution interface; the task description is a text description of the task content and requirements, including the task objectives, specific requirements and required skills; the historical task data is divided according to the selected time period to obtain a number of task events, and each task event is independent and completed by an execution object; all task events are numbered in chronological order to form a historical task data set A, and A={a1,a2,...,an}, where a1 represents the first historical task event, a2 represents the second historical task event, and so on, an represents the nth historical task event, and n represents the historical task event number, which is a positive integer; when there are historical task events with the same execution time in the historical task event set, the sorting number of the corresponding historical task event is randomly determined; S102. Obtain the task description of each historical task event in the historical task data set A, extract keywords, and compare them with the preset task difficulty keywords, and record the number of times the preset task difficulty keywords appear m; according to the execution time of the historical task event, obtain the completion time length T of each historical task event in the historical task data set A; obtain the execution steps and execution interface of each historical task event, and record the number of execution steps M and the number of execution interfaces N; S103. Normalize the number of times m, the completion time length T, the number of execution steps M, and the number of execution interfaces N corresponding to each element in the historical task data set A, respectively, to obtain the feature vector V1 corresponding to each element in the historical task data set A, and V1=[v1,v2,v3,v4], where v1 represents the number of times m, the preset task difficulty keyword corresponding to each element in the historical task data set A, v2 represents the completion time length T corresponding to each element in the historical task data set A, v3 represents the number M of execution steps corresponding to each element in the historical task data set A, and v4 represents the number N of execution interfaces corresponding to each element in the historical task data set A; S104. Mark the feature vector V1 corresponding to each element in the historical task data set A as data points on the radar chart in order, and the radar chart includes four axes with the same starting point, namely k1, k2, k3 and k4, and k1 corresponds to the value of v1, k2 corresponds to the value of v2, k3 corresponds to the value of v3, and k4 corresponds to the value of v4; connect the data points of two adjacent axes on the radar chart to form a quadrilateral, calculate the radar chart area S corresponding to each element in the historical task data set A, and thus obtain the task difficulty score set B corresponding to the historical task data set A, and B={S1,S2,...,Sn}, wherein S1 represents the radar chart area S of the first historical task event, S2 represents the radar chart area S of the second historical task event, and so on, Sn represents the radar chart area S of the nth historical task event; the elements with the same number in the historical task data set A and the task difficulty score set B correspond one to one, and the task difficulty of the historical task event is determined by comparing the sizes of the elements in the task difficulty score set B of the historical task event, and one radar chart area S represents a task difficulty level.

3. A big data analysis and management system based on an intelligent cloud platform, using a big data analysis and management method based on an intelligent cloud platform as described in any one of claims 1-2, characterized in that: The system includes: a data collection module, a task difficulty assessment module, an execution object efficiency assessment module, a correlation relationship analysis module and a real-time task allocation module; The data collection module collects historical task data and execution object data from the intelligent cloud platform and performs preprocessing; the task difficulty assessment module performs task difficulty assessment on the historical task data, analyzes the task description, execution time, execution steps and execution interface, and generates the task difficulty level of the historical task data; the execution object efficiency assessment module analyzes the task satisfaction of the execution object and the completion status of historical task events, and generates the efficiency level of the execution object; the association analysis module compares and analyzes the task difficulty level of the historical task data and the efficiency level of the execution object to obtain an association relationship model between the task event and the execution object; the real-time task allocation module receives the real-time task data, calculates the similarity between the real-time task data and the historical data, and allocates corresponding execution objects to the real-time task data in combination with the results of the association analysis module.

4. The big data analysis and management system based on the intelligent cloud platform according to claim 3 is characterized by: The data collection module includes a historical task data collection unit and an execution object data collection unit; The historical task data collection unit is responsible for collecting historical task data from the intelligent cloud platform and dividing it into several task events according to the selected time period to form a historical task data set; the execution object data collection unit obtains execution object data from the intelligent cloud platform, divides the execution object data according to the division method of historical task data, and forms an execution object data set.

5. The big data analysis and management system based on the intelligent cloud platform according to claim 3 is characterized in that: The task difficulty assessment module includes a task description extraction and keyword matching unit, a feature vector generation and normalization unit, and a difficulty score calculation unit; The task description extraction and keyword matching unit extracts the task description in the historical task data, extracts the keywords and compares them with the preset task difficulty keywords, and records the number of occurrences of the preset task difficulty keywords; The feature vector generation and normalization unit normalizes the task difficulty related features of each element in the historical task data set, generates a feature vector and labels it; the difficulty score calculation unit generates a radar chart based on the feature vector, calculates the radar chart area corresponding to each element in the historical task data set, and obtains a task difficulty score set.

6. The big data analysis and management system based on the intelligent cloud platform according to claim 3 is characterized by: The execution object efficiency evaluation module includes a task proficiency index extraction unit, a completion index calculation unit and an efficiency score calculation unit; The task proficiency index extraction unit extracts the task proficiency index in the execution object data set; the completion index calculation unit calculates the deviation between the completion status data of the historical task event and the preset target data to obtain the completion index; the efficiency score calculation unit forms a feature vector based on the task proficiency index and the completion index, performs efficiency evaluation and obtains an efficiency score set.

7. The big data analysis and management system based on the intelligent cloud platform according to claim 3 is characterized by: The association relationship analysis module includes a data point coordinate generation unit and an association relationship type analysis unit; The data point coordinate generation unit matches the task difficulty score and the efficiency score according to the historical task events to obtain the data point coordinates; the association relationship type analysis unit marks the data points in the plane rectangular coordinate system to obtain a scatter plot of the task difficulty score and the efficiency score, and determines the association relationship type by analyzing the distribution of the data points; The real-time task allocation module includes a real-time task feature extraction unit and a similarity calculation and matching unit; The real-time task feature extraction unit receives real-time task data and extracts corresponding feature vectors; the similarity calculation and matching unit performs similarity calculation on the feature vectors of real-time task events and historical task events, selects the historical task event with the greatest similarity as the matching result, outputs the corresponding execution object level, determines the idle status of the execution object and performs real-time task allocation.

Images