CN113391987A - Quality prediction method and device for online software system - Google Patents
Quality prediction method and device for online software system Download PDFInfo
- Publication number
- CN113391987A CN113391987A CN202110693234.9A CN202110693234A CN113391987A CN 113391987 A CN113391987 A CN 113391987A CN 202110693234 A CN202110693234 A CN 202110693234A CN 113391987 A CN113391987 A CN 113391987A
- Authority
- CN
- China
- Prior art keywords
- software system
- quality prediction
- online
- model
- quality
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 99
- 238000011161 development Methods 0.000 claims abstract description 57
- 230000008569 process Effects 0.000 claims abstract description 50
- 238000012549 training Methods 0.000 claims abstract description 33
- 238000012545 processing Methods 0.000 claims description 40
- 238000012360 testing method Methods 0.000 claims description 21
- 230000007547 defect Effects 0.000 claims description 20
- 238000004458 analytical method Methods 0.000 claims description 14
- 230000002159 abnormal effect Effects 0.000 claims description 12
- 238000004590 computer program Methods 0.000 claims description 12
- 238000007781 pre-processing Methods 0.000 claims description 9
- 238000010801 machine learning Methods 0.000 claims description 7
- 238000004422 calculation algorithm Methods 0.000 claims description 6
- 238000012217 deletion Methods 0.000 claims description 5
- 230000037430 deletion Effects 0.000 claims description 5
- 238000013507 mapping Methods 0.000 claims description 5
- 238000010187 selection method Methods 0.000 claims description 5
- 238000007619 statistical method Methods 0.000 claims description 5
- 238000011156 evaluation Methods 0.000 claims description 4
- 238000005070 sampling Methods 0.000 claims description 4
- 238000010586 diagram Methods 0.000 description 7
- 238000004891 communication Methods 0.000 description 5
- 238000001914 filtration Methods 0.000 description 3
- 239000013049 sediment Substances 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 230000008030 elimination Effects 0.000 description 2
- 238000003379 elimination reaction Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000012552 review Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013075 data extraction Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000007477 logistic regression Methods 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000012067 mathematical method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000002085 persistent effect Effects 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3447—Performance evaluation by modeling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/36—Preventing errors by testing or debugging software
- G06F11/3668—Software testing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Data Mining & Analysis (AREA)
- Computer Hardware Design (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Stored Programmes (AREA)
Abstract
The invention provides a quality prediction method and a quality prediction device for an online software system, wherein the method comprises the following steps: determining characteristic information of an online software system to be subjected to quality prediction; inputting the characteristic information of the online software system to be subjected to quality prediction into a quality prediction model to obtain a quality prediction result of the online software system output by the quality prediction model; the quality prediction model is obtained by training a preset regression model after sample data is constructed on the basis of characteristic information of a software system to be online in the development process. The embodiment of the invention realizes high precision of quality prediction of an online software system and continuous dynamic integer adjustment data so as to improve the quality.
Description
Technical Field
The invention relates to the technical field of software quality, in particular to a quality prediction method and a quality prediction device for an online software system.
Background
In the development process of a software system, whether a team adopts a traditional waterfall development model, a rapid prototyping model or the most popular agile development model, a series of design, development and test work is generally required to be carried out on the team, and the overall quality risk is evaluated according to certain specific indexes before online so as to determine whether the team is online or not. However, it is often the case that the software quality performance and expectations after the bring-up are very different. Because the software engineering itself is a system engineering, the factors affecting the quality of the software system are very many, and it is not comprehensive to predict the quality simply from the test passing rate, the remaining defects, and the like.
The technical scheme of the invention is to predict the quality level of a research and development team through the quality of application software developed by the team, but the problems of low quality precision and incapability of dynamic adjustment of a software system to be predicted to be on-line exist.
Disclosure of Invention
The embodiment of the invention provides a quality prediction method and a quality prediction device for an online software system, which are used for solving the problems that the quality precision of the software system to be online predicted is not high and dynamic adjustment cannot be carried out at present.
In a first aspect, an embodiment of the present invention provides a quality prediction method for an online software system, including:
determining characteristic information of an online software system to be subjected to quality prediction;
inputting the characteristic information of the online software system to be subjected to quality prediction into a quality prediction model to obtain a quality prediction result of the online software system output by the quality prediction model;
the quality prediction model is obtained by training a preset regression model after sample data is constructed on the basis of characteristic information of a software system to be online in the development process.
Further, the feature information of the software system to be online in the development process is feature information for representing quality, which is obtained by extracting basic feature information based on the full life cycle historical data of the software to be online in the development process and preprocessing the basic feature information;
the method for constructing sample data based on the characteristic information of the software system to be online in the development process comprises the following steps:
and sampling the characteristic information of the software system to be online in the development process to obtain balance data with set negative and positive proportions, and removing useless characteristics in the balance data by a characteristic selection method to obtain the sample data.
Further, the sample data comprises a training set and a test set;
the quality prediction model is obtained by training a preset regression model after sample data is constructed on the basis of characteristic information of a software system to be online in a development process, and comprises the following steps:
training a preset regression model through a machine learning gradient descent algorithm based on the training set to obtain a model to be tested;
and continuously adjusting the parameters of the model to be tested obtained each time to the current optimal model through supervised learning based on the test set so as to obtain the quality prediction model.
Further, based on the test set, continuously adjusting the parameters of the model to be tested obtained each time to the current optimal model through supervised learning, including:
sequentially evaluating the model to be tested obtained each time through an average absolute error method MAE, an average variance MSE and an R square value;
and sequentially carrying out continuous parameter adjustment on the model to be measured until the current optimal model is obtained on the basis of the evaluation results of the model to be measured obtained each time by the MAE, the MSE and the R square value.
Further, the extracting of the basic feature information based on the full lifecycle historical data of the software to be online in the development process includes:
extracting the full life cycle historical data of each system by adopting a development tool, performing data modeling on the data of each system, and extracting the data of different systems to a data set;
and analyzing the missing value, the abnormal value and the repeated record of the extracted data set, and performing corresponding processing according to the analysis result to obtain basic characteristic information including development stage data.
Further, the corresponding processing according to the analysis result comprises missing value processing, abnormal value processing and duplication elimination processing;
the missing value processing comprises the steps of selecting direct deletion, special value replacement, mean value or median replacement and interpolation modes based on the missing value distribution condition;
the outlier processing comprises taking no processing or processing as missing values based on the identified outlier distribution; wherein the identified outliers are based on statistical analysis and mapping, on boxplot analysis, on models, on distances, on densities, or on clustering;
the deduplication process includes direct deduplication.
Further, the basic characteristic information further comprises on-line defects of the components, and the on-line defects of each component comprise quantity and severity;
the characteristic information for representing quality obtained after preprocessing the basic characteristic information comprises:
and normalizing, standardizing and converting the label into a numerical value for the data in the development stage, and weighting and summing the quantity and the severity of the on-line defects of each component to obtain the quality represented in a digital form, thereby obtaining the characteristic information.
In a second aspect, an embodiment of the present invention provides a quality prediction apparatus for an online software system, including:
the characteristic determining unit is used for determining the characteristic information of the online software system to be subjected to quality prediction;
the quality prediction unit is used for inputting the characteristic information of the online software system to be subjected to quality prediction into a quality prediction model to obtain a quality prediction result of the online software system output by the quality prediction model;
the quality prediction model is obtained by training a preset regression model after sample data is constructed on the basis of characteristic information of a software system to be online in the development process.
In a third aspect, an embodiment of the present invention provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor implements the steps of the quality prediction method of the online software system according to any one of the above-mentioned first aspects when executing the program.
In a fourth aspect, an embodiment of the present invention provides a non-transitory computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the steps of the quality prediction method for an online software system according to any one of the above-mentioned first aspects.
According to the quality prediction method and device of the online software system, provided by the embodiment of the invention, the characteristic information of the online software system is input into the quality prediction model, the quality prediction result of the online software system output by the quality prediction model is obtained, the quality prediction model is obtained through sample data training constructed by data of historical project sediment and is dynamically adjusted, and the quality of the online software system can be predicted more accurately.
Drawings
In order to more clearly illustrate the technical solutions of the present invention or the prior art, the drawings needed for the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.
FIG. 1 is a flow chart of a quality prediction method of an online software system according to the present invention;
FIG. 2 is a schematic diagram illustrating a training process of a predetermined regression model according to the present invention;
FIG. 3 is a schematic diagram illustrating a process for adjusting a current optimal model according to the present invention;
FIG. 4 is a schematic diagram of the basic feature information extraction process provided by the present invention;
FIG. 5 is a schematic diagram of a quality prediction system of an online software system according to the present invention;
fig. 6 is a schematic structural diagram of an electronic device provided in the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is obvious that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The following describes a quality prediction method and apparatus for an online software system according to the present invention with reference to fig. 1 to 6. According to the invention, the quality of the software system in the development process is predicted through the full life cycle historical data of the software system developed and on-line by the team, and a local optimal solution obtained by adjusting the characteristic value can be used for providing suggestions for a quality improvement strategy.
The embodiment of the invention provides a quality prediction method of an online software system. Fig. 1 is a schematic flow chart of a quality prediction method of an online software system according to an embodiment of the present invention, as shown in fig. 1, the method includes:
step 120, inputting the characteristic information of the online software system to be subjected to quality prediction into a quality prediction model to obtain a quality prediction result of the online software system output by the quality prediction model;
the quality prediction model is obtained by training a preset regression model after sample data is constructed on the basis of characteristic information of a software system to be online in the development process.
According to the method provided by the embodiment of the invention, various data in the software development process are collected, the quality prediction model is obtained by adopting a machine learning method and through data training of historical project sediment, and the quality of the software system to be online is predicted more accurately; meanwhile, a strategy direction for improving the quality is continuously provided at different stages of the whole development process in a mode of continuously and dynamically adjusting certain data to obtain the current optimal predicted quality.
Based on any embodiment, the feature information of the software system to be online in the development process is the feature information used for representing quality, which is obtained by extracting basic feature information based on the full life cycle historical data of the software to be online in the development process and preprocessing the basic feature information;
the method for constructing sample data based on the characteristic information of the software system to be online in the development process comprises the following steps: and sampling the characteristic information of the software system to be online in the development process to obtain balance data with set negative and positive proportions, and removing useless characteristics in the balance data by a characteristic selection method to obtain the sample data.
It should be noted that, in order to achieve better learning effect, the data needs to ensure a certain negative and positive ratio, i.e. the data imbalance problem needs to be dealt with. And because the data size in the characteristic preprocessing process is limited, data balance is carried out in a mode of repeating subclasses. But to prevent overfitting, cross-validation and regularization may be used.
The feature selection can reduce useless features to reduce the complexity of a final model, and the final purpose of the feature selection is to obtain a simplified model and improve the calculation speed under the condition of not reducing the prediction accuracy or having little influence on the prediction accuracy. Feature selection is not to reduce training time (in fact, some techniques increase overall training time), but to reduce model scoring time. The commonly used feature selection method: variance filtering, chi-square filtering, mutual information filtering, recursive feature elimination, and the like.
In any of the above embodiments, the sample data comprises a training set and a test set;
the quality prediction model is obtained by training a preset regression model after constructing sample data based on the characteristic information of the software system to be online in the development process, and as shown in fig. 2, the quality prediction model comprises the following steps:
in the field of machine learning, a regression model is a model widely used in the field of supervised learning, and is used to predict a relationship between an input variable (independent variable) and an output variable (dependent variable), and particularly, when a value of an input variable changes, a value of an output variable changes accordingly. The regression model is just a function representing the mapping between input variables to output variables. The invention selects logistic regression to train the model, and adopts gradient descent algorithm as the model learning algorithm.
And step 220, continuously adjusting parameters of the model to be tested obtained each time to the current optimal model through supervised learning based on the test set so as to obtain the quality prediction model.
It should be noted that, the parameters of the model to be measured are adjusted to perform continuous training until the generalization capability of the model is no longer improved. When the whole process is terminated, the obtained model is the current optimal model, the online quality performance of the components in the development process can be predicted by using the model, and when the predicted quality is lower than a threshold value, quality risk early warning can be given; meanwhile, suggestions for avoiding quality risks are given by continuously trying to adjust the values of the relevant features.
Based on any of the above embodiments, based on the test set, the continuous parameter adjustment is performed on the model to be tested obtained each time to the current optimal model through supervised learning, as shown in fig. 3, including the following steps:
and 320, sequentially carrying out continuous parameter adjustment on the model to be measured until the model is the current optimal model according to the evaluation result of the model to be measured obtained each time based on the MAE, the MSE and the R square value.
It should be noted that, for the regression model, generally, methods such as an average absolute error Method (MAE), an average variance (MSE), an R-square value and the like are used to evaluate the quality of the model, and the smaller MAE and MSE are, which indicates that the model to be tested describes that the test set has better accuracy; the larger the R square value is, the better the fitting effect of the model to be tested is.
Based on any of the above embodiments, the extracting of the basic feature information based on the full lifecycle historical data of the software to be online in the development process, as shown in fig. 4, includes the following steps:
specifically, data modeling: since the development tools, management tools, test tools, persistent integration tools, monitoring tools, etc. used by different companies are different, in order to extract data from various tools, it is necessary to model the data in advance so that the data from different systems can be extracted into one data set.
And 420, analyzing the missing value, the abnormal value and the repeated record of the extracted data set, and performing corresponding processing according to the analysis result to obtain basic characteristic information including development stage data.
Specifically, data extraction: all relevant data in the component development process and after online are extracted from each system, the extraction mode can be an API (application programming interface), a database or file export mode and the like, and the whole process is actually standard ETL. And (3) data analysis: for the extracted data set, the missing value, the abnormal value and the repeated record are mainly analyzed, and the adopted mathematical method comprises the following steps: sum, arithmetic mean, variance, standard deviation, covariance matrix, correlation coefficient matrix, skewness, kurtosis, and the like.
According to any embodiment of the above embodiments, the corresponding processing according to the analysis result includes missing value processing, abnormal value processing and duplicate removal processing;
the missing value processing comprises the steps of selecting direct deletion, special value replacement, mean value or median replacement and interpolation modes based on the missing value distribution condition;
the outlier processing comprises taking no processing or processing as missing values based on the identified outlier distribution; wherein the identified outliers are based on statistical analysis and mapping, on boxplot analysis, on models, on distances, on densities, or on clustering;
the deduplication process includes direct deduplication.
Specifically, the following data cleansing process is performed according to the analysis result:
1) missing value processing: different strategies should be adopted for different data scenes, the distribution situation of the missing values should be judged firstly, and then methods such as direct deletion, special value replacement, mean value or median replacement, interpolation and the like are adopted according to the situation.
2) Abnormal value processing: for simple cases, statistical analysis and drawing identification can be adopted, and for complex cases, box type drawing analysis, model-based, distance-based, density-based, clustering-based and other methods can be adopted for identification. And for the identified abnormal value, the abnormal value can be processed by adopting no processing or being regarded as a missing value according to the distribution situation.
3) And (3) duplicate removal treatment: and deleting the repeated data directly.
According to any of the above embodiments, the basic feature information further includes on-line defects of the components, and the on-line defects of each component include number and severity;
specifically, the quality of the developed system after being online is predicted, the predicted dimensionality needs to reach the component level, and the prediction result has more guiding significance. For each component, all relevant attributes are gathered from different internal systems, including but not limited to:
the method comprises the following steps: state staying, state distribution, development time, test time, change rate and test return rate;
the technical characteristics are as follows: development language, framework complexity, whether external software is relied on, software rating is relied on;
code: the total line number, the submission times, the review passing times and the construction passing rate;
use case: total number, pass number of review, pass rate of execution and coverage rate of demand;
internal defects: state retention, number, state distribution, legacy severity distribution, defect escape rate, defect return rate, defect effectiveness rate;
automated testing: unit test coverage, unit test pass rate, interface automation pass rate, UI automation pass rate and code coverage;
and (3) online defects: quantity, severity distribution.
The characteristic information for representing quality obtained after preprocessing the basic characteristic information comprises: and normalizing, standardizing and converting the label into a numerical value for the data in the development stage, and weighting and summing the quantity and the severity of the on-line defects of each component to obtain the quality represented in a digital form, thereby obtaining the characteristic information.
Specifically, the extracted data in the development stage is normalized, normalized and subjected to label-to-value processing, the online defects of one component are weighted and summarized according to the quantity and the severity to obtain the quality represented in a numerical form, and a quality threshold value is set, and the quality higher than the value is considered to be poor. That is, the quality of a component is weighted by the number and severity of defects on the line.
The quality prediction method of an online software system provided by the invention is described below, and the quality prediction system of the online software system described below and the quality prediction system of the online software system described above can be referred to correspondingly.
Fig. 5 is a schematic structural diagram of a quality prediction system of an online software system according to an embodiment of the present invention, and as shown in fig. 5, the system includes a feature determination unit 510 and a quality prediction unit 520;
the characteristic determining unit 510 is configured to determine characteristic information of an online software system to be quality predicted;
the quality prediction unit 520 is configured to input the feature information of the online software system to be quality predicted into a quality prediction model, and obtain a quality prediction result of the online software system output by the quality prediction model;
the quality prediction model is obtained by training a preset regression model after sample data is constructed on the basis of characteristic information of a software system to be online in the development process.
According to the system provided by the embodiment of the invention, various data in the software development process are collected, the quality prediction model is obtained by adopting a machine learning method and through data training of historical project sediment, and the quality of the software system to be online is predicted more accurately; meanwhile, a strategy direction for improving the quality is continuously provided at different stages of the whole development process in a mode of continuously and dynamically adjusting certain data to obtain the current optimal predicted quality.
Based on any embodiment, the feature information of the software system to be online in the development process is the feature information used for representing quality, which is obtained by extracting basic feature information based on the full life cycle historical data of the software to be online in the development process and preprocessing the basic feature information;
the method for constructing sample data based on the characteristic information of the software system to be online in the development process comprises the following steps:
and sampling the characteristic information of the software system to be online in the development process to obtain balance data with set negative and positive proportions, and removing useless characteristics in the balance data by a characteristic selection method to obtain the sample data.
In any of the above embodiments, the sample data comprises a training set and a test set;
the quality prediction model is obtained by training a preset regression model after sample data is constructed on the basis of characteristic information of a software system to be online in a development process, and comprises the following steps:
training a preset regression model through a machine learning gradient descent algorithm based on the training set to obtain a model to be tested;
and continuously adjusting the parameters of the model to be tested obtained each time to the current optimal model through supervised learning based on the test set so as to obtain the quality prediction model.
Based on any one of the above embodiments, based on the test set, continuously adjusting the parameters of the model to be tested obtained each time to the current optimal model through supervised learning, including:
sequentially evaluating the model to be tested obtained each time through an average absolute error method MAE, an average variance MSE and an R square value;
and sequentially carrying out continuous parameter adjustment on the model to be measured until the current optimal model is obtained on the basis of the evaluation results of the model to be measured obtained each time by the MAE, the MSE and the R square value.
Based on any one of the above embodiments, extracting the basic feature information based on the full lifecycle historical data of the software to be online in the development process includes:
extracting the full life cycle historical data of each system by adopting a development tool, performing data modeling on the data of each system, and extracting the data of different systems to a data set;
and analyzing the missing value, the abnormal value and the repeated record of the extracted data set, and performing corresponding processing according to the analysis result to obtain basic characteristic information including development stage data.
According to any embodiment of the above embodiments, the corresponding processing according to the analysis result includes missing value processing, abnormal value processing and duplicate removal processing;
the missing value processing comprises the steps of selecting direct deletion, special value replacement, mean value or median replacement and interpolation modes based on the missing value distribution condition;
the outlier processing comprises taking no processing or processing as missing values based on the identified outlier distribution; wherein the identified outliers are based on statistical analysis and mapping, on boxplot analysis, on models, on distances, on densities, or on clustering;
the deduplication process includes direct deduplication.
According to any of the above embodiments, the basic feature information further includes on-line defects of the components, and the on-line defects of each component include number and severity;
the characteristic information for representing quality obtained after preprocessing the basic characteristic information comprises
And normalizing, standardizing and converting the label into a numerical value for the data in the development stage, and weighting and summing the quantity and the severity of the on-line defects of each component to obtain the quality represented in a digital form, thereby obtaining the characteristic information.
Fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present invention, and as shown in fig. 6, the electronic device may include: a processor (processor)610, a communication Interface (Communications Interface)620, a memory (memory)630 and a communication bus 640, wherein the processor 610, the communication Interface 620 and the memory 630 communicate with each other via the communication bus 640. The processor 610 may call logic instructions in the memory 630 to perform a quality prediction method for an online software system, the method comprising: determining characteristic information of an online software system to be subjected to quality prediction; inputting the characteristic information of the online software system to be subjected to quality prediction into a quality prediction model to obtain a quality prediction result of the online software system output by the quality prediction model; the quality prediction model is obtained by training a preset regression model after sample data is constructed on the basis of characteristic information of a software system to be online in the development process.
In addition, the logic instructions in the memory 630 may be implemented in software functional units and stored in a computer readable storage medium when the logic instructions are sold or used as independent products. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
In another aspect, an embodiment of the present invention further provides a computer program product, where the computer program product includes a computer program stored on a non-transitory computer-readable storage medium, the computer program includes program instructions, and when the program instructions are executed by a computer, the computer can execute the quality prediction method of an online software system provided by the above methods, where the method includes: determining characteristic information of an online software system to be subjected to quality prediction; inputting the characteristic information of the online software system to be subjected to quality prediction into a quality prediction model to obtain a quality prediction result of the online software system output by the quality prediction model; the quality prediction model is obtained by training a preset regression model after sample data is constructed on the basis of characteristic information of a software system to be online in the development process.
In yet another aspect, an embodiment of the present invention further provides a non-transitory computer-readable storage medium, on which a computer program is stored, where the computer program is implemented by a processor to execute the quality prediction method of the online software system provided in the foregoing aspects, and the method includes: determining characteristic information of an online software system to be subjected to quality prediction; inputting the characteristic information of the online software system to be subjected to quality prediction into a quality prediction model to obtain a quality prediction result of the online software system output by the quality prediction model; the quality prediction model is obtained by training a preset regression model after sample data is constructed on the basis of characteristic information of a software system to be online in the development process.
The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A quality prediction method of an online software system is characterized by comprising the following steps:
determining characteristic information of an online software system to be subjected to quality prediction;
inputting the characteristic information of the online software system to be subjected to quality prediction into a quality prediction model to obtain a quality prediction result of the online software system output by the quality prediction model;
the quality prediction model is obtained by training a preset regression model after sample data is constructed on the basis of characteristic information of a software system to be online in the development process.
2. The quality prediction method of the online software system according to claim 1, characterized in that the feature information of the online software system to be developed is feature information for representing quality, which is obtained by extracting basic feature information based on the full lifecycle historical data of the online software to be developed and preprocessing the basic feature information;
the method for constructing sample data based on the characteristic information of the software system to be online in the development process comprises the following steps:
and sampling the characteristic information of the software system to be online in the development process to obtain balance data with set negative and positive proportions, and removing useless characteristics in the balance data by a characteristic selection method to obtain the sample data.
3. The method for predicting the quality of the online software system according to claim 1 or 2, wherein the sample data comprises a training set and a test set;
the quality prediction model is obtained by training a preset regression model after sample data is constructed on the basis of characteristic information of a software system to be online in a development process, and comprises the following steps:
training a preset regression model through a machine learning gradient descent algorithm based on the training set to obtain a model to be tested;
and continuously adjusting the parameters of the model to be tested obtained each time to the current optimal model through supervised learning based on the test set so as to obtain the quality prediction model.
4. The method for predicting the quality of the online software system according to claim 3, wherein the step of continuously adjusting the parameters of the model to be tested obtained each time to the current optimal model through supervised learning based on the test set comprises the following steps:
sequentially evaluating the model to be tested obtained each time through an average absolute error method MAE, an average variance MSE and an R square value;
and sequentially carrying out continuous parameter adjustment on the model to be measured until the current optimal model is obtained on the basis of the evaluation results of the model to be measured obtained each time by the MAE, the MSE and the R square value.
5. The quality prediction method of the online software system according to claim 2, wherein the extracting of the basic feature information based on the full lifecycle historical data of the software to be online in the development process comprises:
extracting the full life cycle historical data of each system by adopting a development tool, performing data modeling on the data of each system, and extracting the data of different systems to a data set;
and analyzing the missing value, the abnormal value and the repeated record of the extracted data set, and performing corresponding processing according to the analysis result to obtain basic characteristic information including development stage data.
6. The method for predicting the quality of the online software system according to claim 5, wherein the corresponding processing according to the analysis result comprises missing value processing, abnormal value processing and duplicate removal processing;
the missing value processing comprises the steps of selecting direct deletion, special value replacement, mean value or median replacement and interpolation modes based on the missing value distribution condition;
the outlier processing comprises taking no processing or processing as missing values based on the identified outlier distribution; wherein the identified outliers are based on statistical analysis and mapping, on boxplot analysis, on models, on distances, on densities, or on clustering;
the deduplication process includes direct deduplication.
7. The method for predicting the quality of the online software system according to claim 2 or 5, wherein the basic feature information further comprises online defects of the components, and the online defects of each component comprise the number and the severity;
the characteristic information for representing quality obtained after preprocessing the basic characteristic information comprises:
and normalizing, standardizing and converting the label into a numerical value for the data in the development stage, and weighting and summing the quantity and the severity of the on-line defects of each component to obtain the quality represented in a digital form, thereby obtaining the characteristic information.
8. A quality prediction device of an online software system is characterized by comprising:
the characteristic determining unit is used for determining the characteristic information of the online software system to be subjected to quality prediction;
the quality prediction unit is used for inputting the characteristic information of the online software system to be subjected to quality prediction into a quality prediction model to obtain a quality prediction result of the online software system output by the quality prediction model;
the quality prediction model is obtained by training a preset regression model after sample data is constructed on the basis of characteristic information of a software system to be online in the development process.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps of the quality prediction method of an online software system according to any one of claims 1 to 7 when executing the program.
10. A non-transitory computer readable storage medium, on which a computer program is stored, wherein the computer program, when being executed by a processor, implements the steps of the quality prediction method of an online software system according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110693234.9A CN113391987A (en) | 2021-06-22 | 2021-06-22 | Quality prediction method and device for online software system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110693234.9A CN113391987A (en) | 2021-06-22 | 2021-06-22 | Quality prediction method and device for online software system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113391987A true CN113391987A (en) | 2021-09-14 |
Family
ID=77623396
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110693234.9A Pending CN113391987A (en) | 2021-06-22 | 2021-06-22 | Quality prediction method and device for online software system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113391987A (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2366782A1 (en) * | 1999-04-30 | 2000-11-09 | E.I. Du Pont De Nemours And Company | Distributed hierarchical evolutionary modeling and visualization of empirical data |
| CN109240929A (en) * | 2018-09-18 | 2019-01-18 | 百度在线网络技术（北京）有限公司 | Software quality prediction method, apparatus, terminal and computer readable storage medium |
-
2021
- 2021-06-22 CN CN202110693234.9A patent/CN113391987A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2366782A1 (en) * | 1999-04-30 | 2000-11-09 | E.I. Du Pont De Nemours And Company | Distributed hierarchical evolutionary modeling and visualization of empirical data |
| CN109240929A (en) * | 2018-09-18 | 2019-01-18 | 百度在线网络技术（北京）有限公司 | Software quality prediction method, apparatus, terminal and computer readable storage medium |
Non-Patent Citations (2)
| Title |
|---|
| 彭友松: "机器学习中的数据不平衡问题", pages 1, Retrieved from the Internet <URL:http://blog.sciencenet.cn/blog-54276-377102.html> * |
| 王颖;李金;王磊;徐成振;才忠喜;: "基于机器学习的microRNA预测方法研究进展", 计算机科学, no. 02 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11650968B2 (en) | Systems and methods for predictive early stopping in neural network training | |
| CN111639783A (en) | Line loss prediction method and system based on LSTM neural network | |
| WO2017071369A1 (en) | Method and device for predicting user unsubscription | |
| CN115587543A (en) | Federal learning and LSTM-based tool residual life prediction method and system | |
| CN112632179A (en) | Model construction method and device, storage medium and equipment | |
| CN113887126A (en) | Welding spot quality analysis method and device, terminal equipment and medium | |
| CN114962390A (en) | Hydraulic system fault diagnosis method and system and working machine | |
| CN113110961B (en) | Equipment abnormality detection method and device, computer equipment and readable storage medium | |
| CN112182056A (en) | Data detection method, device, equipment and storage medium | |
| CN113742993A (en) | Method, device, equipment and storage medium for predicting life loss of dry-type transformer | |
| CN113151842A (en) | Method and device for determining conversion efficiency of wind-solar complementary water electrolysis hydrogen production | |
| CN117034197A (en) | Enterprise power consumption typical mode analysis method based on multidimensional Isolate-detection multi-point detection | |
| CN115238583B (en) | Business process remaining time prediction method and system supporting incremental log | |
| CN111612149A (en) | Main network line state detection method, system and medium based on decision tree | |
| Fan | Data mining model for predicting the quality level and classification of construction projects | |
| CN113391987A (en) | Quality prediction method and device for online software system | |
| CN111611117B (en) | Hard disk fault prediction method, device, equipment and computer readable storage medium | |
| CN113657656A (en) | Loan data analysis and prediction method and device | |
| CN113610225A (en) | Quality evaluation model training method and device, electronic equipment and storage medium | |
| CN113139332A (en) | Automatic model construction method, device and equipment | |
| CN112445632A (en) | HPC reliability evaluation method based on fault data modeling | |
| CN109409720B (en) | Personalized auditing method based on big data and deep learning and robot system | |
| CN116780524B (en) | Industrial enterprise short-term load prediction method based on LSTM deep learning | |
| CN117194963B (en) | Industrial FDC quality root cause analysis method, device and storage medium | |
| CN109359133B (en) | Personalized identification method based on big data and deep learning and robot system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |