White-Box Software Testing using Evolutionary Algorithms
2 likes1,307 views
The document discusses the use of evolutionary algorithms, specifically genetic algorithms and genetic programming, in white-box software testing to automate the generation of test cases. It reviews several surveyed papers that explore different aspects of evolutionary testing, including factors affecting fitness functions and approaches to testing object-oriented software. The take-home messages highlight the challenges of generating test cases and the efficacy of evolutionary algorithms in addressing these challenges.
![photo by [ d i e g o ] 12](https://image.slidesharecdn.com/presentation10-110103151918-phpapp02/85/White-Box-Software-Testing-using-Evolutionary-Algorithms-11-320.jpg)
![photo by ][/][ 16](https://image.slidesharecdn.com/presentation10-110103151918-phpapp02/85/White-Box-Software-Testing-using-Evolutionary-Algorithms-15-320.jpg)
![Evolutionary algorithms can be
used to solve [some of] these
problems
85](https://image.slidesharecdn.com/presentation10-110103151918-phpapp02/85/White-Box-Software-Testing-using-Evolutionary-Algorithms-84-320.jpg)
White-Box Software Testing using Evolutionary Algorithms
- 1. White-Box Software Testing using Evolutionary Algorithms Yaser Sulaiman 1
- 2. Time for a break! 2
- 3. Pex 3
- 4. White-box testing 4
- 5. photo by Simon Goldenberg 5
- 7. Statement Branch Weakest Strongest Path Condition 8
- 8. Coming up with test cases/data isn’t as easy as it sounds 9
- 9. photo by racheocity 10
- 10. Cover as much as you can! 11
- 11. photo by [ d i e g o ] 12
- 12. Looks like an optimization problem.. sort of 13
- 13. So how about automating it? 14
- 14. Evolutionary algorithms 15
- 15. photo by ][/][ 16
- 16. Genetic Algorithms (GA) Evolutionary Algorithms Genetic Programming (GP) 17
- 17. GA 18
- 18. Fitness Selection Function Chromosomes Mutation Crossover 19
- 19. GP 20
- 20. Specialization of GA 21
- 21. Where individuals represent computer programs 22
- 22. GP evolves programs! 23
- 23. stolen from Wikipedia 24
- 24. Mathematically speaking 25
- 25. Given a set of inputs and outputs , a program is sought which satisfies = () 26
- 26. Fitness of a candidate program? 27
- 27. Expected output vs. = () Actual output 28
- 28. 2 = − =1 where ∈ and ∈ 29
- 29. Moving on to the surveyed papers 30
- 30. Brendt* Tonella Wappler* Ahmed* 2002 2004 2006 2008 * et al. 31
- 31. Brendt* Tonella Wappler* Ahmed* 2002 2004 2006 2008 * et al. 32
- 32. “Breeding software test cases with genetic algorithms” 33
- 33. “Breeding software test cases with genetic algorithms” 34
- 34. Fossil record 35
- 35. Relative fitness function 36
- 36. Berndt et al. (2002) 37
- 37. 3 factors affect the fitness function 38
- 38. 1. Novelty measure of uniqueness 39
- 39. 2. Proximity measure of closeness 40
- 40. 3. Severity measure of seriousness 41
- 41. “simple rules, complex behavior” 42
- 42. Explorers, prospectors, and miners 43
- 43. Brendt* Tonella Wappler* Ahmed* 2002 2004 2006 2008 * et al. 44
- 44. “Evolutionary testing of classes” 45
- 45. “Evolutionary testing of classes” 46
- 46. Object-oriented testing 47
- 47. Class Under Test (CUT) & Method Under Test (MUT) 48
- 48. Sequence of object constructions & method calls 49
- 49. Tonella (2004) focuses on representation and operators 50
- 50. A chromosome isn’t just a sequence of input values anymore 51
- 51. $a=A():$b=B():$a.m(int, $b) @ 3 52
- 52. Brendt* Tonella Wappler* Ahmed* 2002 2004 2006 2008 * et al. 53
- 53. “Evolutionary unit testing of object- oriented software using strongly- typed genetic programming” 54
- 54. “Evolutionary unit testing of object- oriented software using strongly- typed genetic programming” 55
- 55. GP that is type-aware 56
- 56. Wappler & Wegener (2006) focus on call sequence feasibility 57
- 57. Wappler & Wegener (2006) 58
- 58. Method Call Dependence Graph (MCDG) 59
- 59. Wappler & Wegener (2006) 60
- 60. A sequence is feasible if it corresponds to an acyclic subgraph of the MCDG 61
- 61. Wappler & Wegener (2006) 62
- 62. Brendt* Tonella Wappler* Ahmed* 2002 2004 2006 2008 * et al. 63
- 63. “GA-based multiple paths test data generator” 64
- 64. “GA-based multiple paths test data generator” 65
- 65. Ahmed & Hermadi (2008) focus on the efficiency of test data generation 66
- 66. Generating multiple test data to cover multiple target paths in one GA run 67
- 67. Requires incorporating the paths within the fitness calculation 68
- 68. // TODO: Redigest the paper. 69
- 69. Let’s compare the surveyed papers 70
- 70. Evolutionary Object- Coverage Paper Algorithm Oriented Criterion Berndt et al. GA Branch (2002) Tonella GA Branch (2004) Wappler & Wegener Strongly-Typed GP Branch (2006) Ahmed & Hermadi GA Path (2008) 71
- 71. A direction for future work 72
- 72. The reverse direction 73
- 73. Evolutionary Program Test Cases Algorithm 74
- 74. Evolutionary Program Test Cases Algorithm 75
- 75. Writing test cases first? 76
- 76. It’s called TDD 77
- 77. photo by nyuhuhuu 78
- 78. photo by Magic Madzik 79
- 79. TDD + GP isn’t a new idea 80
- 80. 81
- 81. Take-home messages 82
- 82. Generating test cases can be challenging 83
- 83. Testing object-oriented software provides even more challenges 84
- 84. Evolutionary algorithms can be used to solve [some of] these problems 85
- 85. Even software developers turn to Mother Nature for inspiration 86