Recent studies have shown a remarkable need for testing automation techniques in the context of mobile applications. The main contributions in literature in the field of testing automation regard techniques such as Capture/Replay, Model Based, Model Learning and Random techniques. Unfortunately, only the last two typologies of techniques are applicable if no previous knowledge about the application under testing is available. Random techniques are able to generate effective test suites (in terms of source code coverage) but they need a remarkable effort in terms of machine time and the tests they generate are quite inefficient due to their redundancy. Model Learning techniques generate more efficient test suites but often they do not not reach good levels of coverage. In order to generate test suites that are both effective and efficient, we propose in this paper AGRippin, a novel Search Based Testing technique founded on the combination of genetic and hill climbing techniques. We carried out a case study involving five open source Android applications that has demonstrated how the proposed technique is able to generate test suites that are more effective and efficient than the ones generated by a Model Learning technique.

AGRippin: a novel search based testing technique for Android applications

AMALFITANO, DOMENICO;AMATUCCI, NICOLA;FASOLINO, ANNA RITA;TRAMONTANA, PORFIRIO
2015

Abstract

Recent studies have shown a remarkable need for testing automation techniques in the context of mobile applications. The main contributions in literature in the field of testing automation regard techniques such as Capture/Replay, Model Based, Model Learning and Random techniques. Unfortunately, only the last two typologies of techniques are applicable if no previous knowledge about the application under testing is available. Random techniques are able to generate effective test suites (in terms of source code coverage) but they need a remarkable effort in terms of machine time and the tests they generate are quite inefficient due to their redundancy. Model Learning techniques generate more efficient test suites but often they do not not reach good levels of coverage. In order to generate test suites that are both effective and efficient, we propose in this paper AGRippin, a novel Search Based Testing technique founded on the combination of genetic and hill climbing techniques. We carried out a case study involving five open source Android applications that has demonstrated how the proposed technique is able to generate test suites that are more effective and efficient than the ones generated by a Model Learning technique.
9781450338158
9781450338158
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/11588/628839
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 12
  • ???jsp.display-item.citation.isi??? ND
social impact