Skip to content

Latest commit

 

History

History
71 lines (53 loc) · 3.33 KB

README.md

File metadata and controls

71 lines (53 loc) · 3.33 KB
Raw

Uncertainty-Aware Test Prioritization: Approaches and Empirical Evaluation

Get Started

Usage: options

 -a,--algorithm <arg>       which algorithm you want to employ for
                            priortizing tests (default is SPEA2)
 -d,--dir <arg>             where you want to save results (default is
                            `uncertainty_prioritization_solutions`)
 -f,--testsfile <arg>       [REQUIRED] where you have tests to prioritize
 -r,--repetition <arg>      how many times you want to repeat the
                            experiment (default is 5)
 -s,--strategy <arg>        which strategy you want to employ for
                            priortizing tests (default is 6)
 -t,--timeContraint <arg>   what maximum percentage of time budget you
                            want to apply for (default is 1)

Example

foo.json represents an example of uncertainty-wise tests that has 5 tests (see executedUTCs) and associates with 5 uncertainties (see alluncertainites) and 2 indeterminacy sources (see allIndSs).

using -f to specify where your tests are, then run the command

$ java -jar uncertaintyprioritization.jar -f example/foo.json

Done, solutions can be found under uncertainty_prioritization_solutions

You will get solutions generated using SPEA2 with strategy Problem 6: f(PET, PTR, AUM, ANU) and TB100 under default path, i.e., uncertainty_prioritization_solutions as

uncertainty_prioritization_solutions/
├─ foo_uc1_TB100_SPEA2_ET_CTR_UM_NU_001_FUN.tsv
├─ foo_uc1_TB100_SPEA2_ET_CTR_UM_NU_001_VAR.tsv
├─ ...

Used libraries

If you want to build the tool with source code, you need to add libraries as follows

  • org.eclipse.emf.common_2.28.0.v20230223-0922.jar
  • org.eclipse.emf.ecore_2.33.0.v20230226-0921.jar
  • org.eclipse.emf.mapping.ecore2xml_2.12.0.v20230211-1150.jar
  • org.eclipse.uml2.common_2.5.0.v20221116-1811.jar
  • org.eclipse.uml2.types_2.5.0.v20221116-1811.jar
  • org.eclipse.uml2.uml_5.5.0.v20221116-1811.jar
  • org.eclipse.uml2_5.5.3.v20221116-1811.jar

Experiment and analysis

We conducted the experiment with

  • five algorithms: RS, CellDE, MOCell, NSGAII, and SPEA2
  • 10 strategies: Problem 1 - Problem 10
  • 10 time constraints: TB10 - TB100 specified as time budget
  • repetition: 100
  • five use cases: AW1-AW4 and GS1

In this package, we also provide Appendix A (summarized results for all use cases) and Appendix B (statistical results in detail for each use case, i.e., AW1, AW2, AW3, AW4, GS1). To better find the results for each use case, a navigation table is provided in Appendix B as navigation