CN110209573B - Method for enhancing software fault positioning effect - Google Patents
Method for enhancing software fault positioning effect Download PDFInfo
- Publication number
- CN110209573B CN110209573B CN201910380686.4A CN201910380686A CN110209573B CN 110209573 B CN110209573 B CN 110209573B CN 201910380686 A CN201910380686 A CN 201910380686A CN 110209573 B CN110209573 B CN 110209573B
- Authority
- CN
- China
- Prior art keywords
- fault
- cfnorm
- value
- fault location
- statement
- 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
- G06F11/3668—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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
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)
- Debugging And Monitoring (AREA)
Abstract
A method for enhancing software fault location effect comprises firstly, processing statement coverage matrix to obtain fault location parametersCFNormA value; secondly, the obtained fault location parametersCFNormCombining different indexes with the existing fault locating formulas; and finally, calculating suspicious values of each executable statement by using the improved fault locating formula, and sequencing the statements according to the sequence from high suspicious values to low suspicious values, so as to provide a locating fault basis for programmers. The method of the invention provides a new fault location parameterCFNormFault location parametersCFNormNot only can effectively exertN CF The method has the advantages that the effect of improving the fault positioning effect is improved, and the method can be combined with the traditional fault positioning formula in a product mode, so that the effect of the original technology is improved obviously, and the method has good compatibility and portability.
Description
Technical Field
The invention relates to the technical field of software testing, in particular to a method for enhancing a software fault positioning effect.
Background
The method for eliminating faults in the software is an effective means for ensuring the quality of the software and improving the reliability of the software, and has very important position in the whole software testing process. Only after the fault is accurately positioned, a programmer can analyze and solve the fault, so that how to improve the fault positioning effect becomes a problem of concern of researchers.
Among the existing fault locating technologies, the spectrum-based fault locating technology is favored by more researchers because of the numerous branches and accurate locating. The fault location technology based on the frequency spectrum mainly comprises four objects, namely a program P to be tested, a test case T, frequency spectrum information notification and a suspicious value calculation formula Coefficient. Will already haveClassified test cases T (T successful Or t failed ) The method is implemented on the program P to be tested, spectrum information can be obtained in the running process of the program, and then the spectrum information is input into a formula to obtain suspicious values of each executable statement, wherein the higher the values are, the greater the possibility that the description statement contains errors is.
The spectrum information contains 8 indexes as shown in table 1:
table 1 8 indices included in spectral information
| N CF | The number of times that the failed test case was executed |
| N UF | The number of times that the failed test case was not executed |
| N CS | Number of times that a successful test case is executed |
| N US | The number of times that an unsuccessful test case was executed |
| N C | Number of times the test case is executed |
| N U | Number of times of non-tested case execution |
| N S | Number of successful test cases |
| N F | Number of failed test cases |
Wherein: n (N) CF The larger the value is, the more the number of times that a statement is executed by a failed test case is characterized, which means that the statement is covered by more failed test cases, and the higher the possibility that the statement contains errors is naturally. N (N) CF Almost all are in the molecular position in the existing suspicious value calculation formula, namely N CF In proportion to the probability of statement containing errors, it can be seen that it is highly compatible with the target of fault localization, thus N CF More attention should be paid than to other elements in the spectral information.
Based on this, a new fault location technique DStar is proposed, the calculation formula of the suspicious value is (N) CF )*/(N UF +N CS ). The formula will be N CF Is placed at the molecular position alone and its weight is regulated by index change, and the result shows that N is within a certain range CF The greater the weight of DStar, the better the effect of DStar, which justifies N CF Positive effects on fault location. However, DStar is intended to strengthen N CF A specific formula of the action has single application scene, N CF The function of (2) can only be embodied in a DStar formula, and more fault location technologies based on frequency spectrums cannot be optimized by using the function. In addition, researchers often sort the spectrum information into the form of a statement coverage matrix for convenience of fault localization, as shown in fig. 1. Each row is all the spectral information of a certain sentence, and each column is all the values of a certain element in the spectral information. In the prior art, a statement-oriented mode is adopted, suspicious values of the statement are obtained through processing row information, and few people adopt a spectrum-oriented mode to take column information of a matrix as a study object.
Disclosure of Invention
The invention provides a method for enhancing the fault positioning effect of softwareThe method provides a new fault location parameter CFNorm which can effectively exert N CF The method has the advantages that the effect of improving the fault positioning effect is improved, and the method can be combined with the traditional fault positioning formula in a product mode, so that the effect of the original technology is improved obviously, and the method has good compatibility and portability.
The technical scheme adopted by the invention is as follows:
a method of enhancing the effect of software fault localization comprising the steps of:
firstly, processing a statement coverage matrix to obtain a fault location parameter CFNorm value; secondly, combining the obtained fault locating parameter CFNorm with the existing fault locating formula by different powers; and finally, calculating the suspicious value of each statement by using the improved fault locating formula, and sequencing the statements according to the sequence from high suspicious value to low suspicious value, thereby providing a locating fault basis for programmers.
In the method, the CFNorm value of the fault location parameter is covered by N in a matrix by a statement CF The data in the column is normalized and added with 1, and the normalized value is prevented from being 0.
In this method, for one statement ω in program P, the steps of calculating its CFNorm are as follows:
step 1-normalization of data for NCF columns:
wherein: oldValue (ω) and newValue (ω) are N of the sentence ω, respectively CF The values before and after normalization, max and min are N respectively CF Maximum and minimum values of columns.
Step 2-add 1 to the normalized value:
CFNorm(ω)＝newValue(ω)+1 (2)；
through the above steps, the fault location parameter CFNorm value of each executable statement in the program P can be obtained.
In the method, the method for combining the fault locating parameter CFNorm with the existing fault locating technology is as follows:
AF′＝AF×CFNorm* (3)；
wherein: AF (Algebraic Form) is an algebraic form of the existing fault location formula, CFNorm is the result of performing an exponential operation on the fault location parameter CFNorm, and AF' is the improved fault location formula.
When 0, AF' =af, the larger the value of AF, —the greater the weight of CFNorm, the greater its amplitude of change to AF; with the increase of the value, the effect of the original fault location technology is continuously improved, but when the value is increased to a certain extent, the improvement of the effect is stopped, so that the value should be flexibly selected in the actual use process to obtain the best effect. The suspicious value of each statement is generated by the improved fault locating formula and is submitted to a programmer after being sequenced in a descending order, so that a basis is provided for locating faults.
In the method, the program to be tested P and the test case T are two inputs of a fault positioning technology based on frequency spectrum,
the program under test P is a series of codes in which a bug needs to be found, and is composed of k executable statements s i I=1, 2,3, …, k, notes, empty rows, function and variable declarations, etc. all belong to non-executable statements;
the test case T is j groups of data T input into the program P to be tested j J=1, 2,3, …, m, j sets of data t are set according to whether the desired output is the same as the actual output j Divided into successful test cases (t successful ) And failure test case (t failed )；
The test case T is executed on the program P to be tested, the frequency spectrum information of each sentence can be obtained, and the frequency spectrum information of a plurality of sentences forms a sentence coverage matrix.
In the method, in the sentence coverage matrix, the second row is a sentence s 1 Will N CF ，N UF ，……，N F Inputting the same data into the existing calculation formula to obtain sentence s 1 And so on, the suspicious values of all k executable sentences can be obtained, and the sentences are reduced according to the suspicious valuesThe sentences most likely to contain the bug are presented to the programmer in order.
The invention provides a method for enhancing the fault positioning effect of software, which has the advantages that:
(1): the invention provides a new fault location parameter CFNorm, which considers that a statement is covered by more failure test cases, the higher the possibility of error is contained, the CFNorm passes through N in the spectrum information CF The column data is normalized and then added with 1.
(2): the CFNorm is combined with the traditional fault location technology in a certain index, and the CFNorm can be optimized, so that the purpose of improving the fault location effectiveness is achieved.
(3): CFNorm is a parameter, not an independent technique, which makes N CF The advantages of the method are not limited to a specific technology, but can be added to any technology, so that the effect of more fault location technologies is enhanced, and the method has high flexibility and expandability.
Drawings
Fig. 1 is a sentence coverage matrix diagram.
FIG. 2 is a flow chart of the use of the present invention.
FIG. 3 is a graph of AVE values for various techniques under different conditions.
Detailed Description
Principle analysis:
as shown in table 1 described in the background art, the test case T is executed on the program P to be tested, so that the suspicious value of each executable statement, that is, 8 pieces of data given in the table, can be obtained. Processing the 8 data by using different fault locating techniques will result in different statement suspicious values. Different fault locating technologies are different processing modes of the data, and a calculation formula can be regarded as a specific form of the fault locating technology.
Three existing fault location techniques are exemplified below.
One of the existing fault location techniques (Tarantula):
tarantula', modified Tarantula:
two existing fault location techniques (Ochiai):
ochiai', modified Ochiai:
/>
three existing fault localization techniques (cross stab):
crosstab', modified Crosstab:
it can be seen that the original fault localization technique can be improved by multiplying it by CFNorm.
The program to be tested P and the test case T are two inputs of the fault location technology based on frequency spectrum. The program under test P is a series of codes in which a bug needs to be found, and is composed of k executable statements s i I=1, 2,3, …, k, notes, empty rows, function and variable declarations, etc. all belong to non-executable statements. The test case T is m groups of data T input into the program P to be tested j J=1, 2,3, …, m sets of data t can be set depending on whether the desired output is the same as the actual output j Divided into successful test cases (t successful ) And failure test case (t failed ). The test case T is executed on the program P to be tested, so that spectrum information of each sentence can be obtained, and the spectrum information of a plurality of sentences forms a sentence coverage matrix, as shown in fig. 1.
Such as: the second row in FIG. 1 is statement s 1 Will N CF ，N UF ，……，N F Inputting the same data into the existing calculation formula to obtain sentence s 1 And the like, suspicious values of all k executable sentences can be obtained, the sentences are arranged in descending order according to the suspicious values, namely, the sentences most likely to contain bug are preferentially presented to programmers, and the programmers are helped to rapidly and accurately locate faults.
From the above, N CF The method has a very important role in software fault location, has higher weight than other elements in spectrum information, and the DStar technology is proposed according to the principle. How to let N CF Improving the effect of more fault location technologies, rather than being limited to only one technology, becomes a problem to be solved.
The invention provides a method for enhancing software fault location effect, which provides a new fault location parameter CFNorm capable of setting N CF Combines the advantages of the prior fault locating technology with the advantages of a plurality of traditional fault locating technologies, thereby improving the effect of the corresponding original technology. It should be noted that CFNorm is not a separate technique, but is specific to N CF The new parameters obtained by processing cannot be applied alone and need to be used in combination with the existing fault location technology.
Unlike the "statement-oriented" approach, CFNorm is the product of the "spectrum-oriented" approach. In order to comprehensively consider the conditions of all executable sentences in the program and simultaneously reduce the complexity of index calculation to be performed next, the invention covers the sentences in a matrix with N CF The column data is normalized to convert the jagged values to values between 0 and 1. In addition, CFNorm is factored into the existing suspicious value calculation formula, so it should avoid the occurrence of 0 values, and therefore, the normalized values are all added to 1, resulting in a set of values between 1 and 2, i.e., CFNorm.
For one statement ω in procedure P, the steps for calculating its CFNorm are as follows:
step 1-normalization of data for NCF columns:
where oldValue (ω) and newValue (ω) are N of the sentence ω, respectively CF The values before and after normalization, max and min are N respectively CF Maximum and minimum values of columns.
Step 2-add 1 to the normalized value:
CFNorm(ω)＝newValue(ω)+1 (2)；
through the steps, the CFNorm value of each executable statement in the program P to be tested can be obtained, and then the CFNorm value is applied to the existing fault location technology.
Fig. 2 shows a flow chart of CFNorm usage, and it can be seen that in the prior fault localization procedure, the spectrum information is the only input of the fault localization formula, and in the present invention, CFNorm is extracted from the spectrum information to become the second input of the formula.
The method of combining CFNorm with existing fault localization techniques is as follows:
AF′＝AF×CFNorm* (3)；
wherein: AF (Algebraic Form) is an algebraic form of the existing fault location technique, CFNorm is the result of the exponential operation of CFNorm, and AF' is the fault location technique obtained after improvement. When 0, AF' =af, the larger the value of AF, i.e., the greater the weight of CFNorm, the greater its amplitude of change to AF.
With the increase of the value, the effect of the original fault location technology is continuously improved, but when the value is increased to a certain extent, the improvement of the effect is stopped, so that in the actual use process, the value needs to be flexibly selected to obtain the best effect.
Figure 3 shows AVE values for various techniques under different conditions. AVE is an index for evaluating the failure localization effect, and the lower the value is, the better the effect of the technique is explained. Best and Worst represent the Best case and Worst case, respectively, the Best case representing that the statement containing an error is first checked when the statement containing an error is given the same suspicious value as the other statements, and the Worst case representing that the statement containing an error is last checked when the statement containing an error is given the same suspicious value as the other statements.
Tarantula (=0) has Tarantla ' values of 16.70% and 25.97% in the best case and worst case respectively, and as the values rise, tarantla ' has a decreasing AVE value, while Tarantla ' has the best effect when it=4, and it has AVE values of 11.81% and 21.08% in both cases respectively, which is improved by 4.89% compared to the prior art.
After 4, the AVE values in both cases began to rise back, thus stopping the experiment on them at 10.
The values of AVE at best and worst case for Ochiai (=0) are 13.98% and 23.26%, respectively, with increasing values, the value of AVE at Ochiai 'is continuously decreasing, and when the value of ochiai=3, the effect of Ochiai' is best, with AVE values at both cases being 11.59% and 20.86%, respectively, which are improved by 2.39% and 2.4% compared to the prior art. After 3, the AVE value in both cases began to rise back, thus stopping the experiment on it at 10.
The AVE values for cross stab (=0) at best and worst case are 14.99% and 24.21%, respectively. As the increase continues, the effect of crostab' continues to increase, so we have expanded the experimental scale. The results show that when the AVE value of crostab 'in both cases remained decreasing (the same effect in 11 and 12) and that the effect of crostab' in case of=14 was best when the AVE value in both cases was 12.08% and 21.30%, respectively, increased by 2.91% compared to the prior art, but when the AVE value of crostab 'exceeded 14, the effect of crostab' began to fluctuate, and the AVE value thereof gradually stabilized, thus stopping the experiment on it when the value was=20.
The suspicious value of each statement is generated by the changed fault locating technology and is submitted to a programmer after being sequenced in a descending order, so that a basis is provided for locating faults. If the AF' generates a suspicious value ordering that may help the programmer locate the fault more effectively than the AF, namely: the fewer sentences that need to be checked to find a bug, the better the effect of AF' than AF.
Claims (4)
1. A method for enhancing the fault location effect of software, comprising the steps of:
firstly, processing a statement coverage matrix to obtain a fault location parameter CFNorm value; secondly, combining the obtained fault locating parameter CFNorm with the existing fault locating formula by different indexes; finally, calculating suspicious values of each statement by using the improved fault locating formula, and sequencing the statements according to the sequence from high suspicious values to low suspicious values, so as to provide a locating fault basis for programmers;
the value of the fault location parameter CFNorm is covered by the statement to cover N in the matrix CF Normalizing the data of the column, and adding 1 to obtain the data; in this method, for one statement ω in program P, the steps of calculating its CFNorm are as follows:
step 1-normalization of data for NCF columns:
wherein: oldValue (ω) and newValue (ω) are N of the sentence ω, respectively CF The values before and after normalization, max and min are N respectively CF Maximum and minimum values of columns;
step 2-add 1 to the normalized value:
CFNorm(ω)＝newValue(ω)+1 (2)
through the above steps, the fault location parameter CFNorm value of each executable statement in the program P can be obtained.
2. A method of enhancing software fault localization effects as claimed in claim 1, wherein: the method for combining the fault location parameter CFNorm with the existing fault location technology is as follows:
AF′＝AF×CFNorm * (3)
wherein: AF (Algebraic Form) is an algebraic form of the existing fault localization formula, CFNorm * Is the result of the index operation of the fault location parameter CFNormAF' is an improved fault localization formula;
when 0, AF' =af, the larger the value of AF, —the greater the weight of CFNorm, the greater its amplitude of change to AF; with the increase of the value, the effect of the original fault location technology is continuously improved, but when the value is increased to a certain extent, the improvement of the effect is stopped, the suspicious value of each statement is generated by using the improved fault location formula, and the suspicious value is submitted to a programmer after descending order, so that the basis is provided for locating the fault.
3. A method of enhancing software fault localization effects as claimed in claim 1, wherein: the program under test P and the test case T are two inputs of a spectrum-based fault localization technique,
the program under test P is a series of codes in which a bug needs to be found, and is composed of k executable statements s i I=1, 2,3, k composition, notes, blank lines, functions, variable declarations, etc. all belong to non-executable statements;
the test case T is m groups of data T input into the program P to be tested j J=1, 2,3, …, m, and m sets of data t according to whether the desired output is the same as the actual output j Divided into successful test cases (t successful ) And failure test case (t failed )；
The test case T is executed on the program P to be tested, the frequency spectrum information of each sentence can be obtained, and the frequency spectrum information of a plurality of sentences forms a sentence coverage matrix.
4. A method of enhancing software fault localization effects as claimed in claim 3, wherein: in the sentence coverage matrix, the second row is sentence s 1 Will N CF ，N UF ，N CS ，N US ，N C ，N U ，N S ，N F The data is input into the existing calculation formula,
wherein: n (N) CF Representing the number of times that the failed test case is executed; n (N) UF Representing the number of times that the failed test case was executed; n (N) CS Indicating that it was successfulThe number of times the test case is executed; n (N) US Representing the number of times that an unsuccessful test case is executed; n (N) C The number of times of executing the tested case is represented; n (N) U The number of times of executing the non-tested case is represented; n (N) S Representing the number of successful test cases; n (N) F Representing the number of failed test cases; can obtain statement s 1 And so on, the suspicious values of all k executable sentences can be obtained, the sentences are arranged in descending order according to the suspicious values, and the sentences most likely to contain bug can be presented to the programmer preferentially.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910380686.4A CN110209573B (en) | 2019-05-08 | 2019-05-08 | Method for enhancing software fault positioning effect |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910380686.4A CN110209573B (en) | 2019-05-08 | 2019-05-08 | Method for enhancing software fault positioning effect |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110209573A CN110209573A (en) | 2019-09-06 |
| CN110209573B true CN110209573B (en) | 2023-06-02 |
Family
ID=67787055
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910380686.4A Active CN110209573B (en) | 2019-05-08 | 2019-05-08 | Method for enhancing software fault positioning effect |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110209573B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113127334A (en) * | 2020-01-15 | 2021-07-16 | 阿里巴巴集团控股有限公司 | Data processing method and device, electronic equipment and storage equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103995780A (en) * | 2014-05-30 | 2014-08-20 | 浙江理工大学 | Program error positioning method based on statement frequency statistics |
| WO2015118536A1 (en) * | 2014-02-06 | 2015-08-13 | B.G. Negev Technologies And Applications Ltd., At Ben-Gurion University | Using model-based diagnosis to improve software testing |
| CN105975388A (en) * | 2016-03-28 | 2016-09-28 | 南京邮电大学 | Incremental defect positioning method based on frequency spectrum |
| CN106445801A (en) * | 2016-04-27 | 2017-02-22 | 南京慕测信息科技有限公司 | Method for positioning software defects on basis of frequency spectrum positioning and visualization |
| CN109254924A (en) * | 2018-09-28 | 2019-01-22 | 中国科学院长春光学精密机械与物理研究所 | A kind of software fault positioning method, device, equipment and readable storage medium storing program for executing |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10437702B2 (en) * | 2016-02-29 | 2019-10-08 | B. G. Negev Technologies And Applications Ltd., At Ben-Gurion University | Data-augmented software diagnosis method and a diagnoser therefor |
| US10678673B2 (en) * | 2017-07-12 | 2020-06-09 | Fujitsu Limited | Software program fault localization |
-
2019
- 2019-05-08 CN CN201910380686.4A patent/CN110209573B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015118536A1 (en) * | 2014-02-06 | 2015-08-13 | B.G. Negev Technologies And Applications Ltd., At Ben-Gurion University | Using model-based diagnosis to improve software testing |
| CN103995780A (en) * | 2014-05-30 | 2014-08-20 | 浙江理工大学 | Program error positioning method based on statement frequency statistics |
| CN105975388A (en) * | 2016-03-28 | 2016-09-28 | 南京邮电大学 | Incremental defect positioning method based on frequency spectrum |
| CN106445801A (en) * | 2016-04-27 | 2017-02-22 | 南京慕测信息科技有限公司 | Method for positioning software defects on basis of frequency spectrum positioning and visualization |
| CN109254924A (en) * | 2018-09-28 | 2019-01-22 | 中国科学院长春光学精密机械与物理研究所 | A kind of software fault positioning method, device, equipment and readable storage medium storing program for executing |
Non-Patent Citations (3)
| Title |
|---|
| 基于分层程序频谱的软件故障定位方法研究;叶俊民;《小型微型计算机***》;20150930(第9期);1953-1957 * |
| 基于频谱的软件多故障定位;宗芳芳;《信息科技》;20160716;全文 * |
| 通过增大边际权重提高基于频谱的错误定位效率;谭德贵等;《计算机学报》;20101215(第12期);127-134页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110209573A (en) | 2019-09-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP7095408B2 (en) | Software program fault location | |
| CN110362484B (en) | Method and device for positioning multiple faults of software, electronic equipment and storage medium | |
| EP2597573A1 (en) | Test data generation | |
| CN104572474B (en) | A kind of lightweight location of mistake Implementation Technology based on Dynamic Slicing | |
| US10761961B2 (en) | Identification of software program fault locations | |
| CN110209573B (en) | Method for enhancing software fault positioning effect | |
| CN103914386B (en) | Software defect positioning method based on input parameter characteristic spectrum | |
| US10496519B2 (en) | Method invocation synthesis for software program repair | |
| US20150154238A1 (en) | Systems and Methods for Generating a Cross-Product Matrix In a Single Pass Through Data Using Single Pass Levelization | |
| Businge et al. | Compatibility prediction of Eclipse third-party plug-ins in new Eclipse releases | |
| CN116166967A (en) | Data processing method, equipment and storage medium based on meta learning and residual error network | |
| JP2012230538A (en) | Software evaluation device, software evaluation method and system evaluation device | |
| Shu et al. | Fault localization using a failed execution slice | |
| Jiang et al. | Testing and debugging in continuous integration with budget quotas on test executions | |
| Eder et al. | Selecting manual regression test cases automatically using trace link recovery and change coverage | |
| Sarhan et al. | Effective spectrum based fault localization using contextual based importance weight | |
| US10621073B2 (en) | Method and apparatus for testing software by using static analysis results and computer readable recording medium having program for performing the same | |
| JP4479958B2 (en) | Coverage acquisition system | |
| CN112668890A (en) | Combined empowerment method and system for relay protection comprehensive evaluation | |
| Solanki et al. | Novel Classification of Test Case Prioritization Techniques | |
| Daniel et al. | Improving spectrum-based fault-localization through spectra cloning for fail test cases beyond balanced test suite | |
| JP6660333B2 (en) | Information extraction device, information extraction method, and information extraction program | |
| Badhera et al. | Test case prioritization algorithm based upon modified code coverage in regression testing | |
| Sarhan | Enhancing spectrum based fault localization via emphasizing its formulas with importance weight | |
| Parsa et al. | Statistical debugging using a hierarchical model of correlated predicates |
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 |