CN112115045B - Failure prediction method for complex software system - Google Patents
Failure prediction method for complex software system Download PDFInfo
- Publication number
- CN112115045B CN112115045B CN202010838115.3A CN202010838115A CN112115045B CN 112115045 B CN112115045 B CN 112115045B CN 202010838115 A CN202010838115 A CN 202010838115A CN 112115045 B CN112115045 B CN 112115045B
- Authority
- CN
- China
- Prior art keywords
- failure
- characterization
- software system
- complex software
- data
- 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.)
- Active
Links
Images
Classifications
-
- 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
- G06F11/3604—Software analysis for verifying properties of programs
- G06F11/3608—Software analysis for verifying properties of programs using formal methods, e.g. model checking, abstract interpretation
-
- 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
- G06F11/3672—Test management
- G06F11/3684—Test management for test design, e.g. generating new test cases
-
- 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
- G06F11/3672—Test management
- G06F11/3688—Test management for test execution, e.g. scheduling of test suites
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Debugging And Monitoring (AREA)
Abstract
The invention discloses a failure prediction method of a complex software system, which starts from the state and behavior characteristics of the complex software system, analyzes and extracts failure characterization measurement elements of the complex software system, searches for characterization rules expressed inside and outside the complex software system before the complex software system fails, obtains root causes generated by the characterization rules, establishes a failure prediction model, and realizes failure prediction of the complex software system, thereby providing a basis for health management during system operation, avoiding the occurrence of system failure behaviors, and ensuring the credible operation of the system.
Description
Technical Field
The invention relates to the technical field of software reliability, in particular to a failure prediction method for a complex software system.
Background
With the high-speed development of computer software technology, various software systems are in a variety, the functions completed by the systems tend to increase day by day, and the systems interact with other systems, equipment, sensors and people increasingly, so that the concept of a complex software system is formed. These complex software systems will incur huge losses once they fail. In traditional software engineering and system engineering, verification and validation (V & V) are mainly used to ensure that a complex software system reaches a state of near zero failure and strict compliance with requirements before deployment, thereby ensuring the trusted operation of the system. Experience has shown, however, that in practice the system cannot reach this near failure-free state at all. To predict and prevent a complex software system from failing, it is first understood what is the failure, why the failure occurs, what state and behavior is before the failure occurs, what behavior of the system is under what circumstances and what conditions are triggered to cause the failure. The complex software system failure can be accurately and effectively predicted only by solving the problems of the software failure representation rule and the failure mechanism. Failure characterization refers to the representation of some status and behavior characteristics associated with a failure, which is a type of metric that reflects whether a failure has occurred in a system. The failure characterization rule refers to the correlation between the failure characterization metric element and the failure type. The failure mechanism is a mechanism process that a defect is introduced into a code due to human errors, the defect becomes a fault under a certain condition, and the fault is activated to become a failure in an operating state.
Regarding failure characterization, the prior art mainly focuses on code and development processes, and is relatively deficient in analyzing the state and behavior characteristics of a system in time/space, whole/local, software/hardware and other dimensions. In the research of failure mechanism, the prior art fails to consider software and hardware interaction behavior and dynamic runtime information, and fails to perform comprehensive mechanism analysis on failure caused by new characteristics of a complex software system, so that the failure mechanism analysis is unreasonable and insufficient, and the accuracy of failure prediction of the complex software system is further influenced.
Therefore, how to accurately predict the impending failure in a complex software system by using the failure characterization rule is an urgent problem to be solved by those skilled in the art.
Disclosure of Invention
In view of the above, the present invention provides a failure prediction method for a complex software system, which starts from the state and behavior characteristics of the complex software system, analyzes and extracts a failure characterization metric element of the complex software system, finds a characterization rule expressed inside and outside the complex software system before the complex software system fails, obtains a root cause generated by the characterization rule, establishes a failure prediction model, and implements failure prediction of the complex software system, thereby providing a basis for health management during system operation, avoiding occurrence of system failure behavior, and ensuring reliable operation of the system.
In order to achieve the purpose, the invention adopts the following technical scheme:
a failure prediction method for a complex software system comprises the following specific steps:
step 1: analyzing the state and behavior characteristics of the complex software system in a multi-dimensional manner, and extracting failure measurement meta-information;
step 2: constructing a failure representation knowledge base according to the failure measurement meta-information;
and step 3: constructing a failure representation rule set of the complex software system based on the failure representation knowledge base, and obtaining a failure mechanism of the complex software system through quantitative analysis of a multiple-distance regression model and qualitative analysis of failure reasons;
and 4, step 4: constructing a failure prediction model of the complex software system based on the failure characterization rule set and the failure characterization knowledge base;
and 5: and inputting the software defect data of the complex software system to be detected into the failure prediction model to obtain a prediction result.
Preferably, the software failure types are mainly divided into functional failure and performance failure, and the multi-dimensional analysis comprises four dimensions of internal/external, time/space, whole/local and software/hardware.
Preferably, in the step 2, according to the failure measurement meta information, corresponding failure data is collected from international universal public data sets (UCI, weka, AWS, MLC, promise, IBM, NASA, etc.), if the data acquired from the open source website cannot meet the requirements and failure data capable of accurately covering dimensions of internal/external, time/space, whole/local, software/hardware cannot be obtained, information on measurement purpose, measurement method, measurement meta interpretation, measurement scale type, measurement data source, etc. of each failure characterization measurement meta is analyzed in a manner of cooperating with an enterprise, corresponding test cases are designed for different complex software systems, the failure data is acquired by means of testing, etc., then the failure data is subjected to layering, structuring and organizing processing, and constructing the failure characterization knowledge base.
Preferably, in the step 3, the failure characterization rule set is constructed based on the failure characterization database of the complex software system, the root cause generated by the failure characterization rule is obtained according to the incidence relation between the failure characterization rules in the failure characterization rule set, and the failure mechanism of the complex software system is obtained by a method combining the multiple regression model quantitative analysis and the failure cause qualitative analysis.
Preferably, according to the failure characterization rule and the failure characterization knowledge base, the failure prediction model is constructed by adopting algorithms such as KNN, SVM, neural network and Softmax; selecting and optimizing the failure prediction model, comparing by adopting a machine learning algorithm to realize selection, and optimizing by class imbalance, cross validation, error analysis and the like;
and preprocessing the data in the failure characterization knowledge base, including data cleaning, data standardization, data integration, data discretization and the like.
Preferably, in the step 5, the failure characterization rule set is combined with the failure prediction model to perform prediction, the software defect data with measurement elements of the complex software system is input into the failure prediction model, the output data includes accuracy, recall rate and false alarm rate, and finally the prediction result includes non-failure, functional failure, performance failure and all failure.
According to the technical scheme, compared with the prior art, the failure prediction method for the complex software system is disclosed and provided, the failure characterization rule is obtained according to the requirement of typical complex software system failure prediction, and the failure characterization rule is utilized to carry out failure prediction on impending failure in the operation of the complex software system. Therefore, differential failure prediction aiming at a complex software failure mechanism is realized, the efficiency and the accuracy of the software failure prediction are improved, and the occurrence of software failure is effectively avoided.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
FIG. 1 is a schematic diagram of the structure provided by the present invention;
FIG. 2 is a schematic diagram illustrating a comparison of accuracy box diagrams of a complex software system failure prediction method and a common method according to the present invention;
FIG. 3 is a schematic diagram illustrating a comparison between a complex software system failure prediction method and a recall rate box diagram of a conventional method according to the present invention;
FIG. 4 is a schematic diagram showing a comparison of false alarm rate box diagrams of the complex software system failure prediction method and the common method provided by the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. 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 embodiment of the invention discloses a failure prediction method for a complex software system, which comprises the following specific steps:
s1: analyzing the state and behavior characteristics of the complex software system in a multi-dimensional manner, and extracting failure measurement meta-information; the software failure types are mainly divided into functional failure and performance failure, failure characterization is carried out on four-dimensional analysis including internal/external, time/space, integral/local and software/hardware to obtain failure characterization measurement elements, and failure characterization measurement element information is extracted from the failure characterization measurement elements; the failure is characterized in that some appearance attributes and parameters which can be perceived exist when the complex software system is about to fail or has failed;
s2: constructing a failure representation knowledge base according to the failure measurement meta-information;
s21: collecting corresponding failure data from international universal public data sets (UCI, weka, AWS, MLC, promise, IBM, NASA, etc.) according to the failure measurement meta-information;
s22: judging whether the failure data reach the coverage standard, if the data acquired in the public data set can not accurately cover the failure data of each dimension of internal/external, time/space, whole/local and software/hardware, analyzing the information of the measurement purpose, the measurement method, the measurement element explanation, the measurement scale type, the measurement data source and the like of each failure representation measurement element in a mode of cooperation with an enterprise, setting corresponding test cases for different complex software systems, acquiring the failure data in a test mode and the like, and constructing a failure data set; otherwise, directly constructing a failure data set;
s23: after layering, structuring and organizing the failure data set, carrying out knowledge extraction and knowledge representation, thereby constructing a failure representation knowledge base;
s3: constructing a failure representation rule set of the complex software system based on a failure representation knowledge base, and obtaining a failure mechanism of the complex software system through quantitative analysis of a multiple-distance regression model and qualitative analysis of failure reasons;
s31: classifying functional failures and performance failures based on a failure characterization database of a complex software system, and mining failure characterization rules from software failure behaviors and state characteristics;
s32: judging whether the failure representation rule is effective or not, and if so, constructing a failure representation rule set; otherwise, carrying out information feedback knowledge iteration and updating the failure representation knowledge base;
s33: performing correlation analysis and principal component analysis on the failure characterization rule set, and obtaining a failure mechanism by combining a multiple regression analysis model and a failure characterization mapping function with failure cause qualitative analysis; the failure mechanism carries out information feedback knowledge iteration and updates a failure representation knowledge base
S4: constructing a failure prediction model of the complex software system based on the failure characterization rule set and the failure characterization knowledge base;
s41: preprocessing data in a failure representation knowledge base, wherein the failure representation knowledge base data comprise failure data, failure measurement elements, a failure representation rule set, a failure mechanism and the like, and the data preprocessing comprises data cleaning, data standardization, data integration, data discretization and the like;
s42: constructing a failure prediction model by adopting algorithms such as KNN, SVM, neural network and Softmax according to the preprocessed data; selecting and optimizing a failure prediction model, comparing by adopting machine learning algorithm SVM, Logistic, BN, IBK, AdaBoost, Bagging, PART, J48, RF algorithm and the like to realize selection, and realizing optimization through class unbalance, cross validation, error analysis and the like;
s5: inputting software defect data of a complex software system to be detected into a failure prediction model to obtain a prediction result;
and predicting by combining the failure representation rule set with a failure prediction model, inputting software defect data with measurement elements of the complex software system into the failure prediction model, outputting data comprising accuracy, recall rate and false alarm rate, and finally outputting prediction results comprising non-failure, functional failure, performance failure and average failure.
Examples
Selecting a certain model of unmanned aerial vehicle control system, performing static analysis by using a software static analysis tool Testbed, extracting software failure characterization measurement elements and software defect data according to a software static test report, and dividing a failure data label into functional failure and performance failure according to a defect type and defect description in a detected software defect data collection list. The common "defect types" are classified into function implementation types, logics, programs, algorithms, data, documents and others, and specifically, a certain piece of defect data is classified into a functional failure or a performance failure, and reference is made to "defect description" in detail. In the verification of the present example, 9414 pieces of defect data are extracted, wherein 826 pieces of defect data and 8588 pieces of defect data are extracted, 604 pieces of functional failure data and 401 pieces of performance failure data are extracted from the defect data. 30 failure characterization measurement elements are extracted:
ExecutablereformattedLines numeric;
NumberofBasicBlocks numeric;
AverageLengthofBasicBlocks numeric;
ProcedureEntryPoints numeric;
ProcedureExitPoints numeric;
TotalComments numeric;
CommentsinHeaders numeric;
CommentsinDeclarations numeric;
CommentsinExecutableCode numeric;
BlankLines numeric;
TotalComments/Exe.Lines numeric;
HeaderComments/Exe.Lines numeric;
DeclarationComments/Exe.Lines numeric;
CodeComments/Exe.Lines numeric;
Knots numeric;
CyclomaticComplexity numeric;
EssentialKnots numeric;
EssentialCyclomaticComplexity numeric;
ProcedureStructured(SPV)numeric;
NumberofLoops numeric;
DepthofLoopNesting numeric;
NumberofOrder1Intervals numeric;
MaximumIntervalNesting numeric;
Reducible(Intervals)numeric;
GlobalsinProcedure numeric;
FileFanin string;
FanOut string;
failure type 3:
Defect_gn numeric;
Defect_xn numeric;
defect_final numeric。
wherein "Defect _ gn" indicates a functional failure, and "Defect _ xn" indicates a performance failure.
Selecting 15 pieces of data as a display, as shown in table 1, the last column "defect _ final" is a defect type label, if the defect is "bug", the corresponding number type is 1, if the defect is not, the corresponding number type is "clear", and the corresponding number type is 0.
Table 1 example verification data example
After defect data are collected and preprocessed through data cleaning, data standardization and the like, failure characterization rule verification based on data driving is conducted, feature clustering and feature sorting are conducted through a FECAR feature selection method, failure mechanism verification based on data mining is conducted through PCA and regression analysis, and the results are shown in table 2:
table 2 Experimental exemplification of failure characterization rules and mechanisms
On the basis of a failure representation rule and a failure mechanism, a complex software system failure prediction model is constructed, nine common machine learning classification methods are selected, namely SVM, Logistic, BN, IBK, AdaBoost, Bagging, PART, J48 and RF are used for verifying whether the software failure representation rule and the failure mechanism model exist or not, 10-fold cross verification is adopted in the experimental process, and other excessive data preprocessing does not exist. Wherein, "N" indicates that the failure characterization rule and the failure mechanism algorithm test are not performed, and "Y" indicates that the experimental results are shown in Table 3 after the failure characterization rule and the failure mechanism algorithm test are performed.
Table 3 experimental demonstration results
From table 3, it can be seen that, from the perspective of functional failure, performance failure or total failure, the prediction results through the failure characterization rule and the failure mechanism model are significantly improved, and the detailed information of accuracy (Pre), Recall rate (Recall), false alarm rate (PF), mean and error between different classifiers, and the like is shown in fig. 2-4, where Y represents the prediction results obtained by using the method of the present invention, and N represents the prediction results obtained by using the prediction model without the failure characterization rule and the failure mechanism algorithm. It can also be seen from the figure that the results of the prediction model with the failure characterization rule and the failure mechanism are better than those of the prediction model without the failure characterization rule and the failure mechanism algorithm, both from the viewpoint of different failure types and from the measurement of different prediction indexes (Pre, Recall and PF).
According to the current common software failure prediction evaluation index, the failure prediction of a typical product in practical application can be carried out by utilizing the proposed failure representation rule, and the prediction accuracy is not lower than 85%; after the failure prediction model is optimized, the recall rate of the prediction model is not lower than 80%, the false alarm rate is not higher than 20%, and the accuracy of model prediction is improved.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (6)
1. A failure prediction method for a complex software system is characterized by comprising the following specific steps:
step 1: analyzing the state and behavior characteristics of the complex software system in a multi-dimensional manner, and extracting failure measurement meta-information;
step 2: constructing a failure representation knowledge base according to the failure measurement meta-information;
and step 3: constructing a failure representation rule set of the complex software system based on the failure representation knowledge base, and obtaining a failure mechanism of the complex software system through quantitative analysis of a multiple regression model and qualitative analysis of failure reasons;
and 4, step 4: constructing a failure prediction model of the complex software system based on the failure characterization rule set and the failure characterization knowledge base;
and 5: inputting software defect data of a complex software system to be detected into the failure prediction model to obtain a prediction result;
the step 3 specifically comprises the following steps: s31: classifying functional failures and performance failures based on a failure characterization database of a complex software system, and mining failure characterization rules from software failure behaviors and state characteristics;
s32: judging whether the failure representation rule is effective or not, and if so, constructing a failure representation rule set; otherwise, carrying out information feedback knowledge iteration and updating the failure representation knowledge base;
s33: performing correlation analysis and principal component analysis on the failure characterization rule set, and obtaining a failure mechanism by combining a multiple regression analysis model and a failure characterization mapping function with failure cause qualitative analysis; and (4) carrying out information feedback knowledge iteration on the failure mechanism, and updating a failure representation knowledge base.
2. The method of claim 1, wherein the types of software failures are classified into functional failures and performance failures; the multi-dimensional analysis comprises four dimensions of internal/external, temporal/spatial, global/local and software/hardware.
3. The method according to claim 2, wherein in step 2, corresponding failure data is collected from the public data set according to the failure metric meta-information; if the acquired failure data cannot accurately cover the failure data of the four dimensions, analyzing the measurement purpose, the measurement method, the measurement element explanation, the measurement scale type and the measurement data source information of each failure representation measurement element, setting corresponding test cases aiming at different complex software systems, and acquiring the failure data by testing the test cases; and constructing the failure representation knowledge base after layering, structuring and organizing the failure data.
4. The method according to claim 1, wherein in the step 3, the failure characterization rule set is constructed based on the failure characterization database of the complex software system, the root cause of the failure characterization rule set is obtained according to the incidence relation between the failure characterization rules in the failure characterization rule set, and the failure mechanism of the complex software system is obtained by a method combining the multiple regression model quantitative analysis and the failure cause qualitative analysis.
5. The failure prediction method of a complex software system according to claim 1, wherein in the step 4, the failure prediction model is constructed by using a machine learning algorithm according to the failure characterization rule set and the failure characterization knowledge base; selecting and optimizing the failure prediction model, comparing by adopting a machine learning algorithm to realize selection, and optimizing by class imbalance, cross validation and error analysis; and preprocessing the data in the failure characterization knowledge base.
6. The method according to claim 1, wherein the failure prediction model is combined with the set of failure characterization rules to perform prediction in step 5, the software defect data with measurement elements of the complex software system is input into the failure prediction model, the output data includes accuracy, recall rate and false alarm rate, and finally the prediction results include non-failure, functional failure, performance failure and all failure.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010838115.3A CN112115045B (en) | 2020-08-19 | 2020-08-19 | Failure prediction method for complex software system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010838115.3A CN112115045B (en) | 2020-08-19 | 2020-08-19 | Failure prediction method for complex software system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112115045A CN112115045A (en) | 2020-12-22 |
CN112115045B true CN112115045B (en) | 2022-03-18 |
Family
ID=73804152
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010838115.3A Active CN112115045B (en) | 2020-08-19 | 2020-08-19 | Failure prediction method for complex software system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112115045B (en) |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106021097A (en) * | 2016-05-09 | 2016-10-12 | 北京航空航天大学 | Software reliability index interval estimation method based on test characteristics |
CN106776316A (en) * | 2016-12-15 | 2017-05-31 | 中国电力科学研究院 | A kind of power information software defect prevention method |
CN107832219A (en) * | 2017-11-13 | 2018-03-23 | 北京航空航天大学 | The construction method of software fault prediction technology based on static analysis and neutral net |
CN108345544A (en) * | 2018-03-27 | 2018-07-31 | 北京航空航天大学 | A kind of software defect distribution analysis of Influential Factors method based on complex network |
CN108664402A (en) * | 2018-05-14 | 2018-10-16 | 北京航空航天大学 | A kind of failure prediction method based on software network feature learning |
CN110597735A (en) * | 2019-09-25 | 2019-12-20 | 北京航空航天大学 | Software defect prediction method for open-source software defect feature deep learning |
CN111338972A (en) * | 2020-03-30 | 2020-06-26 | 北京计算机技术及应用研究所 | Machine learning-based software defect and complexity incidence relation analysis method |
CN111506504A (en) * | 2020-04-13 | 2020-08-07 | 扬州大学 | Software development process measurement-based software security defect prediction method and device |
Family Cites Families (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7797141B2 (en) * | 2002-10-22 | 2010-09-14 | The Boeing Company | Predictive analysis of availability of systems and/or system components |
US8988236B2 (en) * | 2010-05-27 | 2015-03-24 | University Of Southern California | System and method for failure prediction for rod pump artificial lift systems |
US9038030B2 (en) * | 2012-07-26 | 2015-05-19 | Infosys Limited | Methods for predicting one or more defects in a computer program and devices thereof |
CN103093095A (en) * | 2013-01-14 | 2013-05-08 | 湖州师范学院 | Software failure time forecasting method based on kernel principle component regression algorithm |
CN105045719B (en) * | 2015-08-24 | 2018-06-19 | 中国科学院软件研究所 | Method and device based on the change prediction regression test failure for repairing defect |
US10970146B2 (en) * | 2018-03-09 | 2021-04-06 | Seagate Technology Llc | Adaptive fault prediction analysis of computing components |
CN110825615A (en) * | 2019-09-23 | 2020-02-21 | 中国科学院信息工程研究所 | Software defect prediction method and system based on network embedding |
-
2020
- 2020-08-19 CN CN202010838115.3A patent/CN112115045B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106021097A (en) * | 2016-05-09 | 2016-10-12 | 北京航空航天大学 | Software reliability index interval estimation method based on test characteristics |
CN106776316A (en) * | 2016-12-15 | 2017-05-31 | 中国电力科学研究院 | A kind of power information software defect prevention method |
CN107832219A (en) * | 2017-11-13 | 2018-03-23 | 北京航空航天大学 | The construction method of software fault prediction technology based on static analysis and neutral net |
CN108345544A (en) * | 2018-03-27 | 2018-07-31 | 北京航空航天大学 | A kind of software defect distribution analysis of Influential Factors method based on complex network |
CN108664402A (en) * | 2018-05-14 | 2018-10-16 | 北京航空航天大学 | A kind of failure prediction method based on software network feature learning |
CN110597735A (en) * | 2019-09-25 | 2019-12-20 | 北京航空航天大学 | Software defect prediction method for open-source software defect feature deep learning |
CN111338972A (en) * | 2020-03-30 | 2020-06-26 | 北京计算机技术及应用研究所 | Machine learning-based software defect and complexity incidence relation analysis method |
CN111506504A (en) * | 2020-04-13 | 2020-08-07 | 扬州大学 | Software development process measurement-based software security defect prediction method and device |
Non-Patent Citations (3)
Title |
---|
Study of software reliability prediction based on GR neural network;Yumei Wu 等;《The Proceedings of 2011 9th International Conference on Reliability, Maintainability and Safety》;20110812;688-693 * |
基于迁移学习的跨项目软件缺陷预测关键技术研究;邱少健;《中国博士学位论文全文数据库 信息科技辑》;20200615;I138-11 * |
软件可靠性加速测试方法研究;吴玉美;《计算机工程与应用》;20060831;40-43 * |
Also Published As
Publication number | Publication date |
---|---|
CN112115045A (en) | 2020-12-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11494295B1 (en) | Automated software bug discovery and assessment | |
CN106844217B (en) | Method and device for embedding point of applied control and readable storage medium | |
CN110825644B (en) | Cross-project software defect prediction method and system | |
KR20210100600A (en) | software testing | |
US10740216B1 (en) | Automatic bug classification using machine learning | |
US20210064518A1 (en) | Methods Circuits Devices Systems and Functionally Associated Machine Executable Code For Automatic Failure Cause Identification in Software Code Testing | |
Banerjee et al. | Automated triaging of very large bug repositories | |
US11550707B2 (en) | Systems and methods for generating and executing a test case plan for a software product | |
Pitakrat et al. | A framework for system event classification and prediction by means of machine learning | |
Shankar et al. | Towards observability for production machine learning pipelines | |
US20180232299A1 (en) | Composing future tests | |
CN111108481A (en) | Fault analysis method and related equipment | |
Gholamian et al. | A comprehensive survey of logging in software: From logging statements automation to log mining and analysis | |
Murali et al. | Scalable statistical root cause analysis on app telemetry | |
Rathee et al. | Clustering for software remodularization by using structural, conceptual and evolutionary features | |
Falessi et al. | Preserving order of data when validating defect prediction models | |
CN114691403A (en) | Server fault diagnosis method and device, electronic equipment and storage medium | |
CN112115045B (en) | Failure prediction method for complex software system | |
Valueian et al. | Constructing automated test oracle for low observable software | |
CN116302984A (en) | Root cause analysis method and device for test task and related equipment | |
Shilpika et al. | Toward an in-depth analysis of multifidelity high performance computing systems | |
Jeya Mala et al. | Critical components testing using hybrid genetic algorithm | |
JP6547341B2 (en) | INFORMATION PROCESSING APPARATUS, METHOD, AND PROGRAM | |
CN105814546A (en) | Method and system for assisting in the verification and validation of an algorithm chain | |
Abdelkarim et al. | TCP-Net++: Test Case Prioritization Using End-to-End Deep Neural Networks-Deployment Analysis and Enhancements |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |