基于鏡像測試用例多樣性和改進(jìn)的測試用例選擇策略的自適應(yīng)隨機(jī)測試方法研究
發(fā)布時間：2021-08-31 05:09
軟件測試已經(jīng)被認(rèn)為是軟件工程學(xué)科中最重要的研究領(lǐng)域之一,并且仍然是軟件質(zhì)量評估和改進(jìn)中最被廣泛實(shí)踐和研究的方法之一。通過復(fù)雜的抽象,軟件工程師已經(jīng)將測試集成到軟件開發(fā)價值鏈的每個層次。但是不管怎么努力,“缺陷”還是會在不經(jīng)意間“潛入”軟件,導(dǎo)致軟件最終失效。因此,大量的測試技術(shù)被提出用來增強(qiáng)開發(fā)人員的信心,高質(zhì)量的軟件具有滿足相關(guān)人群各種愿望的特定功能。大多數(shù)這類技術(shù)都關(guān)注于選擇測試用例的策略。如何生成高效的測試用例已經(jīng)成為學(xué)術(shù)界和工業(yè)界關(guān)注的主題。在對這一主題的眾多貢獻(xiàn)中,隨機(jī)測試（RT,Random Testing）是最成功的,并且由于其簡單、靈活、高效、易于自動化等優(yōu)點(diǎn)而受到人們的青睞。最近,越來越多的研究人員開始關(guān)注RT的有效性,他們批評RT未能利用任何程序信息來指導(dǎo)測試過程。RT的機(jī)制建立在軟件使用和行為不能被預(yù)測的前提下,因此評估軟件質(zhì)量的測試輸入也應(yīng)該以不可預(yù)測（隨機(jī)）的方式進(jìn)行選擇。盡管這一論斷仍然正確,但各種經(jīng)驗(yàn)觀察表明,程序員更容易犯某些類型的錯誤,從而造成大多數(shù)程序失效的輸入聚集化。這些引起輸入失效的幾何形狀和分布被稱為失效模式。根據(jù)這些額外信息,這些研究人員試圖...
【文章來源】：江蘇大學(xué)江蘇省
【文章頁數(shù)】：213 頁
【學(xué)位級別】：博士
【文章目錄】：
ABSTRACT
摘要
Chapter 1 Introduction
1.1 Software testing
1.2 Software testing resource and expectations management
1.3 Closely related Software testing terminologies:Mistake,Fault,Error,and Failure
1.4 Test case selection and exhaustive testing
1.5 Automated testing and the test oracle problem
1.6 Testing techniques:black-box vs.white-box testing
1.6.1 Black-box testing
1.6.2 White-box testing
1.7 Aims and organization of this thesis
1.8 Contributions of this thesis:
1.9 Outline of the presentation of this thesis
Chapter 2 Literature Review of Adaptive Random Testing Methods
2.1 Introduction
2.2 Random testing(RT)
2.3 Geometric shapes of failure causing inputs-motivation for ART
2.4 Adaptive Random Testing(ART)
2.5 Classifications of ART algorithms
2.5.1 The exclusion approaches/restrictions
2.5.2 Partitioning methods
2.5.3 Restriction-based methods
2.6 Categories based on effectiveness
2.7 Based on solving the high dimensionality problems of ART
2.8 Based on reducing the overhead cost of ART
2.9 Based on applications of ART to OOP
2.10 Fault-seeded programs tested using ART
2.11 Criticisms and open issues on ART
2.12 An in-depth look at mirror adaptive random testing(MART)
2.12.1 Characteristics of MART
2.12.2 Effects of partitioning scheme on the distribution failure causing inputs
2.12.3 Test case diversity:Implications for the effectiveness of MART
2.12.4 The problem with the mapping functions of MART
2.13 Discussions and conclusions
Chapter 3 Metrics,Simulations and Experiment Design Setup
3.1 Introduction
3.2 Effectiveness and efficiency metrics
3.3 Data collection method
3.4 Error seeding and fault injection strategy
3.5 Simulations and object programs
3.5.1 Simulated programs
3.5.2 Object programs
3.6 System environment of the simulation and experimental studies:
3.7 Conclusion
Chapter 4 A Proactive Approach to Test Case Selection-An Efficient Implementation of Adaptive Random Testing
4.1 Introduction
4.2 Description of FSCS-ART algorithm
4.3 The overhead problem of FSCS-ART
4.4 Overlapping rule
4.4.1 Contextual underpinning
4.4.2 FSCS test case distribution
4.4.3 Predicting the maximin values
4.4.4 Analysis of candidate test case selection and the impact of overlapping rule
4.4.5 Determination of efficiency ratio
4.4.6 Complexity of FSCS-O
4.5 Research questions
4.6 Simulations and experimental studies
4.6.1 Results of simulations
4.6.2 Results of empirical studies
4.7 Conclusions
Chapter 5 A Relational Memory Selection:An implementation of Adaptive Random Sequence to Enhance Effectiveness and Efficiency of Adaptive random testing
5.1 Introduction
5.2 Background
5.3 Adaptive random sequence
5.4 A relational memory selection
5.4.1 Potential threat to effectiveness of RMS
5.4.2 Test case generation using RMS
5.5 Time complexity analysis
5.6 Research questions
5.7 Simulations and experimental studies
5.7.1 Parameter setting
5.7.2 Results of simulation studies
5.7.3 Results of experimental studies
5.8 Conclusion
Chapter 6 Elimination by Linear Association:An Effective and Efficient Static Mirror Adaptive Random Testing
6.1 Introduction
6.2 Dynamic partitioning mirror adaptive random testing(DMART)
6.3 Eliminating the linear association among partitions
6.3.1 Eliminating failure unrelated challenge of MART
6.3.2 ELA partitioning schemes
6.3.3 Selection of source domain
6.3.4 Generation of test cases from source domain
6.3.5 Definition of neighboring partition
6.3.6 Identifying Neighboring partitions
6.3.7 Generating Mirror test cases
6.3.8 Demonstrating of test case generation
6.4 Complexity of EMART
6.5 Research questions
6.6 Simulations and experiments
6.6.1 Setting of parameters
6.6.2 Simulations
6.6.3 Experiment with object programs
6.7 Conclusions
Chapter 7 Random Border Mirror Transform:A Diversity Based Approach to an Effective and Efficient Mirror Adaptive Random Testing
7.1 Introduction
7.2 Diversity of mirror mapping function
7.3 Random border mirror transforms(RBMT)
7.3.1 Creation of virtual mirror partition
7.3.2 Random boarder displacement vector
7.3.3 Schematic diversity transform vector
7.3.4 Virtual mirror transform and transform discriminant(T~d)
7.3.5 Mirror transforms
7.3.6 Addressing the problems of MART
7.3.7 Test case generation using RBMT-ART
7.3.8 The algorithm of RBMT
7.3.9 Complexity analysis of RBMT-ART
7.4 Research Questions
7.5 Simulations and Empirical Studies
7.5.1 Parameter setting
7.5.2 Simulations Results and Analysis
7.5.3 Empirical Analysis
7.6 Discussion and conclusion
Chapter 8 General Conclusion
8.1 General conclusion
8.2 Contributions of this thesis
8.3 Future research directions
REFERENCES
AKNOWLEDGEMENT
PUBLICATIONS
【參考文獻(xiàn)】：
期刊論文
[1]Out of sight, out of mind: a distance-aware forgetting strategy for adaptive random testing[J]. Chengying MAO,Tsong Yueh CHEN,Fei-Ching KUO. Science China（Information Sciences）. 2017(09)
[2]A revisit of three studies related to random testing[J]. CHEN Tsong Yueh,KUO Fei-Ching,TOWEY Dave,ZHOU Zhi Quan. Science China（Information Sciences）. 2015(05)
[3]Worst-Input Mutation Approach to Web Services Vulnerability Testing Based on SOAP Messages[J]. Jinfu Chen,Huanhuan Wang,Dave Towey,Chengying Mao,Rubing Huang,Yongzhao Zhan. Tsinghua Science and Technology. 2014(05)
[4]失效區(qū)域緊致性對適應(yīng)性隨機(jī)測試的性能影響(英文)[J]. 陳宗岳,郭斐菁,孫昌愛. 軟件學(xué)報. 2006(12)
本文編號：3374258
【文章來源】：江蘇大學(xué)江蘇省
【文章頁數(shù)】：213 頁
【學(xué)位級別】：博士
【文章目錄】：
ABSTRACT
摘要
Chapter 1 Introduction
1.1 Software testing
1.2 Software testing resource and expectations management
1.3 Closely related Software testing terminologies:Mistake,Fault,Error,and Failure
1.4 Test case selection and exhaustive testing
1.5 Automated testing and the test oracle problem
1.6 Testing techniques:black-box vs.white-box testing
1.6.1 Black-box testing
1.6.2 White-box testing
1.7 Aims and organization of this thesis
1.8 Contributions of this thesis:
1.9 Outline of the presentation of this thesis
Chapter 2 Literature Review of Adaptive Random Testing Methods
2.1 Introduction
2.2 Random testing(RT)
2.3 Geometric shapes of failure causing inputs-motivation for ART
2.4 Adaptive Random Testing(ART)
2.5 Classifications of ART algorithms
2.5.1 The exclusion approaches/restrictions
2.5.2 Partitioning methods
2.5.3 Restriction-based methods
2.6 Categories based on effectiveness
2.7 Based on solving the high dimensionality problems of ART
2.8 Based on reducing the overhead cost of ART
2.9 Based on applications of ART to OOP
2.10 Fault-seeded programs tested using ART
2.11 Criticisms and open issues on ART
2.12 An in-depth look at mirror adaptive random testing(MART)
2.12.1 Characteristics of MART
2.12.2 Effects of partitioning scheme on the distribution failure causing inputs
2.12.3 Test case diversity:Implications for the effectiveness of MART
2.12.4 The problem with the mapping functions of MART
2.13 Discussions and conclusions
Chapter 3 Metrics,Simulations and Experiment Design Setup
3.1 Introduction
3.2 Effectiveness and efficiency metrics
3.3 Data collection method
3.4 Error seeding and fault injection strategy
3.5 Simulations and object programs
3.5.1 Simulated programs
3.5.2 Object programs
3.6 System environment of the simulation and experimental studies:
3.7 Conclusion
Chapter 4 A Proactive Approach to Test Case Selection-An Efficient Implementation of Adaptive Random Testing
4.1 Introduction
4.2 Description of FSCS-ART algorithm
4.3 The overhead problem of FSCS-ART
4.4 Overlapping rule
4.4.1 Contextual underpinning
4.4.2 FSCS test case distribution
4.4.3 Predicting the maximin values
4.4.4 Analysis of candidate test case selection and the impact of overlapping rule
4.4.5 Determination of efficiency ratio
4.4.6 Complexity of FSCS-O
4.5 Research questions
4.6 Simulations and experimental studies
4.6.1 Results of simulations
4.6.2 Results of empirical studies
4.7 Conclusions
Chapter 5 A Relational Memory Selection:An implementation of Adaptive Random Sequence to Enhance Effectiveness and Efficiency of Adaptive random testing
5.1 Introduction
5.2 Background
5.3 Adaptive random sequence
5.4 A relational memory selection
5.4.1 Potential threat to effectiveness of RMS
5.4.2 Test case generation using RMS
5.5 Time complexity analysis
5.6 Research questions
5.7 Simulations and experimental studies
5.7.1 Parameter setting
5.7.2 Results of simulation studies
5.7.3 Results of experimental studies
5.8 Conclusion
Chapter 6 Elimination by Linear Association:An Effective and Efficient Static Mirror Adaptive Random Testing
6.1 Introduction
6.2 Dynamic partitioning mirror adaptive random testing(DMART)
6.3 Eliminating the linear association among partitions
6.3.1 Eliminating failure unrelated challenge of MART
6.3.2 ELA partitioning schemes
6.3.3 Selection of source domain
6.3.4 Generation of test cases from source domain
6.3.5 Definition of neighboring partition
6.3.6 Identifying Neighboring partitions
6.3.7 Generating Mirror test cases
6.3.8 Demonstrating of test case generation
6.4 Complexity of EMART
6.5 Research questions
6.6 Simulations and experiments
6.6.1 Setting of parameters
6.6.2 Simulations
6.6.3 Experiment with object programs
6.7 Conclusions
Chapter 7 Random Border Mirror Transform:A Diversity Based Approach to an Effective and Efficient Mirror Adaptive Random Testing
7.1 Introduction
7.2 Diversity of mirror mapping function
7.3 Random border mirror transforms(RBMT)
7.3.1 Creation of virtual mirror partition
7.3.2 Random boarder displacement vector
7.3.3 Schematic diversity transform vector
7.3.4 Virtual mirror transform and transform discriminant(T~d)
7.3.5 Mirror transforms
7.3.6 Addressing the problems of MART
7.3.7 Test case generation using RBMT-ART
7.3.8 The algorithm of RBMT
7.3.9 Complexity analysis of RBMT-ART
7.4 Research Questions
7.5 Simulations and Empirical Studies
7.5.1 Parameter setting
7.5.2 Simulations Results and Analysis
7.5.3 Empirical Analysis
7.6 Discussion and conclusion
Chapter 8 General Conclusion
8.1 General conclusion
8.2 Contributions of this thesis
8.3 Future research directions
REFERENCES
AKNOWLEDGEMENT
PUBLICATIONS
【參考文獻(xiàn)】：
期刊論文
[1]Out of sight, out of mind: a distance-aware forgetting strategy for adaptive random testing[J]. Chengying MAO,Tsong Yueh CHEN,Fei-Ching KUO. Science China（Information Sciences）. 2017(09)
[2]A revisit of three studies related to random testing[J]. CHEN Tsong Yueh,KUO Fei-Ching,TOWEY Dave,ZHOU Zhi Quan. Science China（Information Sciences）. 2015(05)
[3]Worst-Input Mutation Approach to Web Services Vulnerability Testing Based on SOAP Messages[J]. Jinfu Chen,Huanhuan Wang,Dave Towey,Chengying Mao,Rubing Huang,Yongzhao Zhan. Tsinghua Science and Technology. 2014(05)
[4]失效區(qū)域緊致性對適應(yīng)性隨機(jī)測試的性能影響(英文)[J]. 陳宗岳,郭斐菁,孫昌愛. 軟件學(xué)報. 2006(12)
本文編號：3374258
本文鏈接：http://sikaile.net/kejilunwen/ruanjiangongchenglunwen/3374258.html
最近更新
教材專著
