Intelligent software development system based on Internet big data

Abstract

The invention discloses an intelligent software development system based on internet big data, relates to the technical field of software development, is used for solving the problems that the complexity evaluation dimension of a development task is single and an optimization strategy lacks data driving support, and constructs a modularized intelligent software development system through the internet big data. Comprising a data acquisition unit, a task screening unit, a complexity evaluation unit and an optimization generation unit. The data acquisition unit monitors code quantity of each module and transmits the code quantity to the task screening unit, and the screening unit marks high-complexity modules and sets a screening period. The acquisition unit periodically acquires running time, calling frequency and the number of nested layers, transmits the running time, the calling frequency and the number of nested layers to the complexity evaluation unit, calculates a complexity coefficient by utilizing linear regression, sets an experimental period, then acquires structural node characteristics, and calculates a task complexity coefficient based on nonlinear regression. And the optimization generation unit generates an optimization strategy to assist a user in adjusting system configuration.

Description

Technical Field

[0001] The present invention relates to the field of software development technology, and more specifically, to an intelligent software development system based on Internet big data. Background Art

[0002] As software projects continue to expand in scale and their functional complexity rapidly increases, traditional software development management methods are increasingly exposed to challenges such as difficulty accurately assessing development task complexity, inefficient resource allocation, and reliance on manual experience for optimization strategies. Existing software development management systems typically allocate and optimize tasks based on static code size or subjective assessments, lacking systematic analysis of actual operational data, module structural characteristics, and the dynamic behavior of code. This results in difficulties improving development efficiency, uneven resource utilization, and increased module maintenance costs.

[0003] The existing technology has the following deficiencies: Currently, leveraging the massive data collection and analysis capabilities of the internet, some systems attempt to incorporate big data analysis and machine learning methods into the software development process for intelligent processing. However, these systems still face limitations, such as a single dimension for assessing development task complexity, a lack of data-driven support for optimization strategies, and unclear feedback cycles. These limitations make it difficult to achieve continuous and refined intelligent optimization and control in multi-module concurrent development scenarios. Therefore, an intelligent software development system based on internet big data is proposed.

[0004] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides an intelligent software development system based on Internet big data, which solves the problems raised in the above-mentioned background technology by using a complexity modeling algorithm combining linear regression and nonlinear regression, and a dynamic optimization recommendation mechanism based on task weights.

[0006] To achieve the above-mentioned object, the present invention provides the following technical solution: an intelligent software development system based on Internet big data, comprising a data acquisition unit, a task screening unit, a complexity evaluation unit, and an optimization generation unit, wherein the units are signal-connected; The data collection unit monitors the amount of code in each development module and collects the running time, call frequency, and nesting level of the marked modules during the screening cycle. After receiving the experimental cycle data returned by the complexity evaluation unit, it further collects the number of branch nodes, loop nodes, and node activity. The task screening unit classifies the modules by complexity based on the amount of code collected, marks high-complexity modules, sets a screening cycle, and sends the relevant information to the data collection and complexity assessment unit; The complexity evaluation unit calculates the first and second complexity coefficients based on the running time, call frequency, and number of nested layers, builds a linear regression model to determine the experimental period, and calculates the task complexity coefficient based on the structural node data and activity through a nonlinear regression algorithm, which is then passed to the optimization generation unit. The optimization generation unit combines the preset optimization weights and complexity coefficients to generate recommended optimization strategies for users to adjust system configuration parameters.

[0007] In a preferred embodiment, the number of lines of code of each development task module is monitored as the code amount, and the code amount of each module is transmitted to the task screening unit. The task screening unit receives the code amount of each development task module transmitted by the data acquisition unit, and classifies each module into a low-complexity module and a high-complexity module according to the code amount, and marks the high-complexity module as a marked module; The task screening unit sets a screening cycle and transmits the marking module and the screening cycle to the data acquisition unit and the complexity evaluation unit.

[0008] In a preferred embodiment, the data collection unit receives the screening cycle input by the task screening unit and performs periodic data collection on the marking module; The data acquisition unit periodically monitors the process start and end events of the marking module, records and accumulates the actual running time of the module during the screening cycle; The data collection unit monitors the call events of the tag module and counts the call frequency within the screening cycle; The data collection unit analyzes the code structure in the tag module, counts the number of nested layers of the code, and records the maximum number of nested layers;

[0009] The data acquisition unit synchronously transmits the actual running time, module call frequency, and code nesting level to the complexity evaluation unit.

[0010] In a preferred embodiment, the complexity evaluation unit calculates the product of the actual running time and the module calling frequency based on the screening cycle, the actual running time, the module calling frequency, and the number of code nesting layers, and calculates the ratio of the product to the screening cycle to obtain the first complexity coefficient; The second complexity coefficient is obtained by multiplying the number of code nesting layers and the logarithm of the code amount.

[0011] In a preferred embodiment, the complexity evaluation unit integrates the first complexity coefficient and the second complexity coefficient, performs normalization processing, and constructs a comprehensive complexity coefficient based on a linear regression model; Based on the comprehensive complexity coefficient, the experimental period is set to adjust the data collection frequency. When the comprehensive complexity coefficient is larger, the experimental period is shorter to increase the data collection frequency; when the comprehensive complexity coefficient is smaller, the experimental period is longer to reduce system resource overhead; The complexity evaluation unit transmits the calculated experimental period to the data acquisition unit. The data acquisition unit performs a new data acquisition task according to the experimental period and feeds back the new data set to the complexity evaluation unit.

[0012] In a preferred embodiment, after receiving the experimental cycle transmitted by the complexity evaluation unit, the data collection unit collects the number of branch nodes, the number of loop nodes and the node activity of the marking module according to the time window within the experimental cycle; Detect and mark the branch control structure inside the module and count its number. If the module contains multiple nested branch structures, count and accumulate them layer by layer to get the number of branch nodes. Detect the loop control structure inside the module and count its number. If the module contains multiple nested loop structures, count and accumulate them layer by layer to get the number of loop nodes. The node activity is obtained by statistical calculation based on the number of calls and cumulative running time of each node in the module; The number of branch nodes, the number of loop nodes, and the node activity of the marking module are normalized, packaged into a data structure, and passed to the complexity evaluation unit.

[0013] In a preferred embodiment, the number of branch nodes, the number of loop nodes, and the node activity of the marking module are substituted into a nonlinear regression algorithm to calculate the task complexity coefficient; The complexity evaluation unit calculates the task complexity coefficient and passes it to the optimization generation unit.

[0014] In a preferred embodiment, the optimization generation unit is used to receive the task complexity coefficient input by the complexity evaluation unit and perform weighted calculation with the preset optimization weight coefficient; The generation of optimization weight coefficients is based on a neural network model. By training historical optimization data, the weighted impact of the task complexity coefficient on the optimization of system configuration parameters is determined; The neural network model adopts a multi-layer perceptron structure and is trained and optimized through the back-propagation algorithm.

[0015] In a preferred embodiment, the optimization generation unit obtains the task complexity coefficient and the optimization weight coefficient, and generates a recommended optimization strategy based on the weighted operation; For each configuration parameter, the optimization generation unit calculates its recommended optimization strategy value respectively; After the recommended optimization strategy is generated, the optimization generation unit sends the recommended optimization strategy to the user terminal to guide the user to adjust the system configuration parameters.

[0016] Technical effects and advantages of the present invention: 1. The present invention uses Internet big data to construct a modular intelligent software development system including a data acquisition unit, a task screening unit, a complexity evaluation unit and an optimization generation unit. The data acquisition unit first monitors the amount of code of each development task module and passes it to the task screening unit. The task screening unit divides the modules into low-complexity modules and high-complexity modules based on the amount of code, marks the high-complexity modules, and sets a screening cycle at the same time. The data acquisition unit obtains the actual running time, module call frequency and code nesting layer number of the marked modules according to the screening cycle, and passes them to the complexity evaluation unit. The complexity evaluation unit calculates the first complexity coefficient and the second complexity coefficient through a linear regression model, and sets the experimental cycle accordingly, and then collects the number of branch nodes, the number of loop nodes and the node activity again, and calculates the task complexity coefficient based on the nonlinear regression algorithm. The optimization generation unit generates a recommended optimization strategy based on the task complexity coefficient and the preset optimization weight, guides the user to optimize the system configuration, and improves the level of intelligent software development. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a flowchart of the implementation of the intelligent software development system based on Internet big data of the present invention.

[0018] Figure 2 This is a module diagram of the intelligent software development system based on Internet big data of the present invention DETAILED DESCRIPTION

[0019] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0020] Example 1, an intelligent software development system based on Internet big data, such as Figure 1 As shown, it includes a data acquisition unit, a task screening unit, a complexity evaluation unit and an optimization generation unit, and the signals between the units are connected; The data acquisition unit is used to monitor the code volume of each development task module and pass it into the task screening unit. According to the screening cycle, the actual running time, module call frequency and code nesting layer number of the marked module are detected and passed into the complexity evaluation unit. After detecting the experimental cycle passed into the complexity evaluation unit, the number of branch nodes, the number of loop nodes and node activity in the marked module are obtained and passed into the complexity evaluation unit. The task screening unit is used to receive the code volume of each development task module transmitted by the data acquisition unit, classify each development task module into low-complexity modules and high-complexity modules according to the code volume, screen out the high-complexity modules for marking, set the screening cycle, and transmit the marked module and the screening cycle to the data acquisition unit and the complexity evaluation unit; The complexity evaluation unit receives the screening cycle, actual running time, module call frequency, and code nesting layer number of the marked module, calculates the first complexity coefficient and the second complexity coefficient, constructs a linear regression model based on the first complexity coefficient and the second complexity coefficient, determines the experimental cycle in the screening cycle, sends the experimental cycle to the data acquisition unit, calculates the task complexity coefficient using a nonlinear regression algorithm based on the number of branch nodes, the number of loop nodes, and the node activity in each marked module, and transmits the task complexity coefficient to the optimization generation unit; The optimization generation unit receives the task complexity coefficient transmitted by the complexity evaluation unit and performs weighted operation with the preset optimization weight coefficient to generate a recommended optimization strategy for the intelligent software development system, and sends the recommended optimization strategy to the user end for the user to adjust the configuration parameters of the intelligent software development system.

[0021] The specific implementation is as follows: The data acquisition unit monitors the number of lines of code in each development task module as the code volume ,in is the module identifier. The task screening unit receives the code amount of each development task module passed in by the data acquisition unit. , and classify each module according to the amount of code. The module classification is based on the following: If the module meets It is divided into low-complexity modules; If the module meets It is divided into high-complexity modules; in, is the average amount of code for all modules, is the standard deviation of the code amount, and k is the classification adjustment coefficient. ,The task screening unit marks it as a marking module.

[0022] Task screening unit sets the screening cycle , that is, the time interval for the data acquisition unit to collect data from the high-complexity module, marking the module and the screening cycle Incoming data acquisition unit and complexity evaluation unit.

[0023] The data acquisition unit receives the screening cycle passed in by the task screening unit , perform periodic data collection on the marking module.

[0024] The actual running time refers to the cumulative running time of the marked module during the screening cycle, which is used to reflect the module activity. The data acquisition unit periodically monitors the process start and termination events of the marked module, records and accumulates the actual running time of the module during the screening cycle, and records the actual running time as ; The module call frequency is the total number of times the marked module is called during the screening cycle. The data acquisition unit monitors the call events of the marked module, counts the call frequency of the module during the screening cycle, and records the module call frequency as ; The code nesting level is used to describe the complexity of the code structure in the marking module. The data acquisition unit analyzes the code structure in the marking module, counts the code nesting levels, and records the maximum nesting level. The code nesting level is recorded as ; The data acquisition unit will actually run time , module call frequency , the number of code nesting levels Synchronously transmitted to the complexity evaluation unit for subsequent calculation.

[0025] The complexity evaluation unit receives the screening cycle from the data acquisition unit , actual running time , module call frequency and the number of code nesting levels , the following complexity coefficient calculation is performed based on the data: First complexity coefficient It is used to reflect the comprehensive workload of the marking module. The specific calculation formula is: ,in, is the first complexity coefficient, To mark the actual running time of the module during the screening cycle, The module call frequency of the marked module in the screening cycle, For the screening cycle.

[0026] If the marking module is frequently called during the screening cycle and the actual running time is long, then A larger value indicates that the module has a heavier workload; if the marked module is called less frequently or has a shorter running time, The value is smaller.

[0027] Second complexity coefficient It is used to reflect the nested complexity of the code structure within the module. The specific calculation formula is: ,in, is the second complexity coefficient, To mark the nesting level of module code; The amount of code for the marked module; Take the logarithm of the code quantity to avoid the influence of extreme values.

[0028] If the number of nested levels in the tag module is high and the amount of code is large, then A larger value indicates that the internal logical structure of the module is complex; if the number of nested levels of the marked module is small or the amount of code is small, then The value is smaller.

[0029] Complexity evaluation unit comprehensive first complexity coefficient and the second complexity coefficient , construct the comprehensive complexity coefficient based on the linear regression model. First, to unify the data dimensions of the above two complexity coefficients, and Normalization is performed, and the specific calculation formula is: ; ,in, and They are and The normalized results are all in the interval [0,1]. and is the minimum and maximum value of the first complexity coefficient, and are the minimum and maximum values ​​of the second complexity coefficient.

[0030] The first complexity coefficient and the second complexity coefficient The comprehensive complexity coefficient is calculated as follows: ,in, is the comprehensive complexity coefficient, and They are and The normalized result is and is the linear regression coefficient, and b is the bias term, which reflects the system baseline complexity level.

[0031] Based on the comprehensive complexity coefficient , the complexity assessment unit sets the experimental cycle , to adjust the data collection frequency and experimental period Set the formula to ,in, For the experimental period, For the screening cycle, is the comprehensive complexity coefficient, is the adjustment coefficient, which controls the influence of complexity on the experimental period. Its value range is .

[0032] when The larger the value, the shorter the experimental period. shortened to increase the frequency of data collection; when The smaller the value, the shorter the experimental period. Extended to reduce system resource consumption.

[0033] The complexity evaluation unit calculates the experimental period The data is transmitted to the data acquisition unit, which performs new data acquisition tasks according to the experimental cycle and feeds back the new data set to the complexity evaluation unit for the next round of complexity evaluation and experimental cycle optimization.

[0034] The data acquisition unit receives the experimental cycle from the complexity evaluation unit. After that, the number of branch nodes, the number of loop nodes and the node activity of the marking module are collected according to the time window within the experimental cycle. The data collection unit detects the branch control structure inside the marking module and counts its number. If the module contains multiple nested branch structures, the number of branch nodes is obtained by layer-by-layer counting and accumulation, which is recorded as At the same time, the data acquisition unit detects the loop control structure inside the module and counts its number. If the module contains multiple nested loop structures, the number of loop nodes is counted layer by layer and accumulated, which is recorded as Node activity is used to reflect the calling frequency of nodes in the marked module during the experimental period. The data acquisition unit performs statistical calculations based on the number of calls and cumulative running time of each node in the module. The calculation formula is: ,The data acquisition unit calculates the activity of both branch nodes and loop nodes, and normalizes the node activity to avoid extreme values ​​affecting subsequent operations.

[0035] The number of data branch nodes collected , the number of loop nodes and node activity are packaged into a data structure and passed to the complexity evaluation unit.

[0036] After receiving the above data, the complexity evaluation unit calculates the task complexity coefficient based on the nonlinear regression algorithm , the task complexity coefficient Defined as: ,in, , , is the weight coefficient of the nonlinear regression model, is the activity index adjustment coefficient, which is used to adjust the influence of activity on the complexity coefficient. is a bias term used to adjust the model output range.

[0037] In order to avoid the interference of the differences in the dimensions of each data on the output results of the regression model, the data are first normalized. The calculation formula is as follows: ; ; ;

[0038] Based on the above normalized data, the calculation formula for the optimized task complexity coefficient is: ,in, is the normalized number of branch nodes, is the normalized number of loop nodes, is the normalized node activity. The complexity evaluation unit calculates the task complexity coefficient Then it is passed to the optimization generation unit.

[0039] The optimization generation unit is used to receive the task complexity coefficient passed in by the complexity evaluation unit And with the preset optimization weight coefficient Perform weighted operations to generate recommended optimization strategies for each configuration parameter .

[0040] Optimize weight coefficient The generation of is based on the neural network model. The neural network is trained on historical optimization data to determine the weight influence of the task complexity coefficient on the optimization of system configuration parameters. Defined as ,in, is the complexity coefficient of the current task; Configure the parameter vector for the system in the previous period, is the historical optimization error, which is used to calibrate the neural network output. NN is the neural network model, which is trained and optimized through the back propagation algorithm.

[0041] The neural network model adopts a multi-layer perceptron structure, and its output layer is defined as , where W is the weight matrix and b is the bias term. is the activation function, and the output range is (0,1).

[0042] Optimize the generation unit to obtain the task complexity coefficient and optimized weight coefficient Finally, a recommended optimization strategy is generated based on the weighted operation of the two , the calculation formula is ,in, is the recommended optimization strategy vector, and r is the adjustment coefficient, which is used to control the output amplitude of the optimization strategy to prevent abnormal fluctuations in the optimization results.

[0043] For each configuration parameter , the optimization generation unit calculates its recommended optimization strategy value , the calculation formula is in, is the optimized weight coefficient for the kth configuration parameter, is the kth indicator in the task complexity coefficient, It is an adjustment coefficient used to prevent the optimization range from being too large.

[0044] Recommended optimization strategies After generation, the optimization generation unit sends the recommended optimization strategy to the user end to guide the user to adjust the system configuration parameters. Assume that the current system configuration parameter vector is , then the optimization generation unit is based on the recommended optimization strategy Update the system configuration parameters. The specific calculation formula is: ,in, is the kth configuration parameter value after optimization, It is an adjustment coefficient used to control the adjustment range of configuration parameters to ensure the stability of system operation.

[0045] The recommended optimization strategy generated by the optimization generation unit according to the above operation process will be transmitted to the user end in the form of a vector to guide the user to dynamically adjust the configuration parameters of each system, thereby achieving continuous optimization of system performance and reducing the complexity of the development task module.

[0046] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0047] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0048] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0049] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0050] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0051] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0052] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0053] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0054] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0055] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. An intelligent software development system based on Internet big data, characterized by: It includes data acquisition unit, task screening unit, complexity evaluation unit and optimization generation unit, and the signal connections between the units; The data collection unit monitors the amount of code in each development module and collects the running time, call frequency, and nesting level of the marked modules during the screening cycle. After receiving the experimental cycle data returned by the complexity evaluation unit, it further collects the number of branch nodes, loop nodes, and node activity. The task screening unit classifies the modules by complexity based on the amount of code collected, marks high-complexity modules, sets a screening cycle, and sends the relevant information to the data collection and complexity assessment unit; The complexity evaluation unit calculates the first and second complexity coefficients based on the running time, call frequency, and number of nested layers, builds a linear regression model to determine the experimental period, and calculates the task complexity coefficient based on the structural node data and activity through a nonlinear regression algorithm, which is then passed to the optimization generation unit. The optimization generation unit combines the preset optimization weights and complexity coefficients to generate recommended optimization strategies for users to adjust system configuration parameters.

2. The intelligent software development system based on Internet big data according to claim 1, characterized in that: Monitor the number of code lines of each development task module as the code volume, and transfer the code volume of each module to the task screening unit. The task screening unit receives the code volume of each development task module transferred by the data acquisition unit, and classifies each module into low-complexity modules and high-complexity modules according to the code volume, and marks the high-complexity modules as marked modules; The task screening unit sets a screening cycle and transmits the marking module and the screening cycle to the data acquisition unit and the complexity evaluation unit.

3. The intelligent software development system based on Internet big data according to claim 2, characterized in that: The data collection unit receives the screening cycle transmitted by the task screening unit and performs periodic data collection on the marking module; The data acquisition unit periodically monitors the process start and end events of the marking module, records and accumulates the actual running time of the module during the screening cycle; The data collection unit monitors the call events of the tag module and counts the call frequency within the screening cycle; The data collection unit analyzes the code structure in the tag module, counts the number of nested layers of the code, and records the maximum number of nested layers; The data acquisition unit synchronously transmits the actual running time, module call frequency, and code nesting level to the complexity evaluation unit.

4. The intelligent software development system based on Internet big data according to claim 3, characterized in that: The complexity evaluation unit calculates the product of the actual running time and the module calling frequency based on the screening cycle, the actual running time, the module calling frequency, and the number of code nesting layers, and calculates the ratio of the product to the screening cycle to obtain a first complexity coefficient; The second complexity coefficient is obtained by multiplying the number of code nesting layers and the logarithm of the code amount.

5. The intelligent software development system based on Internet big data according to claim 4, characterized in that: The complexity evaluation unit integrates the first complexity coefficient and the second complexity coefficient, performs normalization processing, and constructs a comprehensive complexity coefficient based on a linear regression model; Based on the comprehensive complexity coefficient, the experimental period is set to adjust the data collection frequency. When the comprehensive complexity coefficient is larger, the experimental period is shorter to increase the data collection frequency; when the comprehensive complexity coefficient is smaller, the experimental period is longer to reduce system resource overhead; The complexity evaluation unit transmits the calculated experimental period to the data acquisition unit. The data acquisition unit performs a new data acquisition task according to the experimental period and feeds back the new data set to the complexity evaluation unit.

6. The intelligent software development system based on Internet big data according to claim 5, characterized in that: After receiving the experimental cycle from the complexity evaluation unit, the data collection unit collects the number of branch nodes, the number of loop nodes, and the node activity of the marking module according to the time window within the experimental cycle; Detect and mark the branch control structure inside the module and count its number. If the module contains multiple nested branch structures, count and accumulate them layer by layer to get the number of branch nodes. Detect the loop control structure inside the module and count its number. If the module contains multiple nested loop structures, count and accumulate them layer by layer to get the number of loop nodes. The node activity is obtained by statistical calculation based on the number of calls and cumulative running time of each node in the module; The number of branch nodes, the number of loop nodes, and the node activity of the marking module are normalized, packaged into a data structure, and passed to the complexity evaluation unit.

7. The intelligent software development system based on Internet big data according to claim 6, characterized in that: Substitute the number of branch nodes, the number of loop nodes, and the node activity of the marking module into the nonlinear regression algorithm to calculate the task complexity coefficient; The complexity evaluation unit calculates the task complexity coefficient and passes it to the optimization generation unit.

8. The intelligent software development system based on Internet big data according to claim 7, characterized in that: The optimization generation unit is used to receive the task complexity coefficient input by the complexity evaluation unit and perform weighted calculation with the preset optimization weight coefficient; The generation of optimization weight coefficients is based on a neural network model. By training historical optimization data, the weighted impact of the task complexity coefficient on the optimization of system configuration parameters is determined; The neural network model adopts a multi-layer perceptron structure and is trained and optimized through the back-propagation algorithm.

9. The intelligent software development system based on Internet big data according to claim 8, characterized in that: The optimization generation unit obtains the task complexity coefficient and optimization weight coefficient, and generates a recommended optimization strategy based on weighted calculation; For each configuration parameter, the optimization generation unit calculates its recommended optimization strategy value respectively; After the recommended optimization strategy is generated, the optimization generation unit sends the recommended optimization strategy to the user terminal to guide the user to adjust the system configuration parameters.

Images