Genetic-based web regression testing: an ontology-based multi-objective evolutionary framework to auto-regression testing of web applications
| Autor: | Maryam Nooraei Abadeh |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
business.industry
Computer science media_common.quotation_subject 020207 software engineering 02 engineering and technology Ontology (information science) Machine learning computer.software_genre Fault detection and isolation Evolutionary computation Management Information Systems Autoregressive model Hardware and Architecture 020204 information systems Regression testing 0202 electrical engineering electronic engineering information engineering Web application Quality (business) State (computer science) Artificial intelligence business computer Software Information Systems media_common |
| Zdroj: | Service Oriented Computing and Applications. 15:55-74 |
| ISSN: | 1863-2394 1863-2386 |
| DOI: | 10.1007/s11761-020-00312-y |
| Popis: | Regression testing is one of the most critical activities in the software maintenance process and its importance is twofold for evolutionary applications, e.g., modern flexible web-based applications. By increasing the complexity of application due to the rapid change, automatic evolutionary testing approaches are essential to find solutions providing different trade-offs between testing objectives by applying evolutionary computation. This paper proposes a model-based regression test case generation framework, as an optimization solution, by impressively taking the advantages of the genetic algorithms (GAs), called genetic-based web regression testing (GbWRT). The aim of the paper is twofold. Firstly, a meta-ontology has been designed based on an in-deep assessment to capture testing challenges caused by the inherent dynamic properties of web applications. Secondly, the multi-objective fitness functions of GbWRT are defined built on top of the meta-ontology in terms of the most important meta-model features. GbWRT minimizes exploration and exploitation in regression testing using an incremental change adaption technique implemented in the proposed GA. This approach allows a new incremental regression testing strategy to solve fault detection effectiveness and the coverability problems which are extendable to different domain-specific modeling environments. GbWRT is evaluated using the proposed fitness functions on two experimental case studies. Also, the results of comparison with three non-evolutionary and evolutionary regression testing methods indicate that the GbWRT is competitive with the state of the art regarding solution quality to perform in web-based non-stationary environments. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink